Gallup, Inc.

GALLUP, INC. 349 NLRB No. 113 1213 Gallup, Inc. and United Steelworkers of America, AFL–CIO, CLC. Cases 16–CA–19898, 16–CA– 19898–2, 16–CA–19898–3, 16–CA–19898–4, and 16–CA–20028 May 31, 2007 DECISION AND ORDER BY CHAIRMAN BATTISTA AND MEMBERS SCHAUMBER AND KIRSANOW On May 25, 2001, Administrative Law Judge Richard J. Linton issued the attached decision. The General Counsel filed exceptions and a supporting brief, and the Respondent filed cross-exceptions, a supporting brief, and a brief in response. The National Labor Relations Board has delegated its authority in this proceeding to a three-member panel. The Board has considered the decision and the record in light of the exceptions and briefs and has decided to affirm the judge’s rulings,1 findings,2 and conclusions and to adopt the recommended Order as modified. 1 We deny the Respondent’s cross-exception based on its trial mo- tion to strike the testimony of all four alleged discriminatees on the ground that the General Counsel assertedly failed to comply with sub- poenas that were served on those individuals. On special appeal, in two orders dated March 1 and 31, 2000, the Board denied that motion, finding that the Respondent had failed to file a request for the subpoe- naed material with the General Counsel as required under 29 CFR § 102.118. The Respondent has proffered no new reason for us to revisit this issue. 2 No exceptions were filed to the judge’s findings that the Respon- dent violated Sec. 8(a)(1) of the Act by restricting employees from posting, distributing, or possessing union-related literature; by remov- ing and confiscating such material; by requiring employees to notify a supervisor before distributing union literature; and by instructing new employees to report attempts by other employees to speak to them about union matters. In addition, no cross-exceptions were filed to the judge’s findings that the Respondent did not violate Sec. 8(a)(1) by issuing a no- harassment statement directed at organizing activity, restricting access to copying equipment, or on additional occasions removing posted union literature or restricting posting and distribution. We also find the allegation that Supervisor Chad Murray removed union literature from an interviewer’s desk and told employees they could not distribute such literature, and the allegation that Lead Super- visor Gisela Uria-Ruiz told Sherri Lee, on May 14, 1999, that distribu- tion of union literature at the workplace was forbidden, to be cumula- tive of other violations the judge found to which the Respondent does not except. We therefore do not reach these allegations. The General Counsel excepted to the judge’s failure to find that the Respondent, by Supervisor Heidi Roberts, violated Sec. 8(a)(1) on three occasions when she told employees who acted as mentors for new interviewer trainees not to talk to the trainees about the Union during training sessions. We find merit in the General Counsel’s exception for the following reasons. The antiunion mentors wore antiunion T-shirts and buttons. However, Roberts addressed herself only to the prounion mentors, who wore no insignia of any kind, and told them to refrain from talking about the Union. This discriminatory conduct violated Sec. 8(a)(1). I. MAY 19 REMOVAL OF POSTED LITERATURE The complaint alleged that on May 19, Supervisor Heidi Roberts removed union flyers from the walls of employees’ cubicle walls and desks, and told employees they could not post union literature. However, as the judge found, the record evidence pertained only to other misconduct allegedly committed by Roberts on different dates. Since the General Counsel neither amended the complaint in this respect nor cited any evidence to sup- port the complaint allegation, the judge correctly dis- missed it. II. JUNE 4 RESTRICTION ON DISTRIBUTION The judge found from the credited evidence that Su- pervisor Gisela Uria-Ruiz, on June 4, told prounion in- terviewers that they could only distribute their literature with permission from a supervisor. Their literature was union literature. The judge found this to be lawful on the ground that Uria-Ruiz’ statement “applied to all distribu- tions” and therefore did not constitute disparate enforce- ment. We disagree and find that Uria-Ruiz’ statement violated Section 8(a)(1) of the Act. As the judge found, no supervisor had previously told the Respondent’s employees of any “unwritten” restric- tion, and the Respondent had permitted numerous post- ings and distributions on other matters. In fact, the judge found that distribution and posting had been “practically unrestricted,” and that “[t]he variety of endeavors [for which posting was allowed] seems endless,” but that this tolerant policy “changed once the union literature began to appear.” The judge also found, without exception from the Respondent, that the Respondent’s asserted pol- icy restricting distribution and posting “in fact was newly created in order to suppress the protected activities of the supporters of the Union.” In this setting, although Uria-Ruiz may not have re- ferred explicitly to “union” literature when she imposed an oral restriction on distribution, she was applying a new and restrictive policy specifically to a union distri- bution. And as noted above, the Respondent had previ- ously not restricted any nonunion-related distribution or posting. Under these circumstances, we find that Uria- Ruiz disparately enforced the Respondent’s policy against union distributions, in violation of Section 8(a)(1) of the Act. St. Francis Medical Center, 340 NLRB 1370, 1378 (2003) (finding violation where employer prohib- ited union distributions but did not prohibit nonunion distributions). DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1214 ORDER3 The National Labor Relations Board adopts the rec- ommended Order of the administrative law judge, as modified below, and orders that the Respondent, Gallup, Inc., Houston, Texas, its officers, agents, successors, and assigns, shall take the action set forth in the Order as modified. 1. Substitute the following paragraph for 1(e) and re- letter the subsequent paragraphs accordingly. “(e) Instructing employees not to talk to new inter- viewers about the union.” 2. Substitute the attached notice for that of the admin- istrative law judge. APPENDIX NOTICE TO EMPLOYEES POSTED BY ORDER OF THE NATIONAL LABOR RELATIONS BOARD An Agency of the United States Government The National Labor Relations Board has found that we vio- lated Federal labor law and has ordered us to post and obey this notice. FEDERAL LAW GIVES YOU THE RIGHT TO Form, join, or assist a union Choose representatives to bargain with us on your behalf Act together with other employees for your bene- fit and protection Choose not to engage in any of these protected activities. WE WILL NOT restrict you from posting, distributing, or possessing union-related literature, flyers, or notices anywhere in the Houston, Texas office that employees are permitted to post, distribute, or possess other non- work-related materials. WE WILL NOT remove or confiscate posted or distrib- uted union-related materials from places where employ- ees may post, distribute, or possess other nonwork- related items. WE WILL NOT require you to notify a supervisor before distributing union literature at work where no such noti- fication is required before you distribute other nonwork materials. WE WILL NOT instruct or request new employees, or trainees, to report to supervision any attempt by other employees to speak to them about union matters as the 3 We will modify the remedial notice in accordance with Ishikawa Gasket America, Inc., 337 NLRB 175 (2001), enfd. 354 F.3d 534 (6th Cir. 2004). new employees, or trainees, are working, where the in- struction or request does not also extend to other non- work-related matters. WE WILL NOT instruct employees that they cannot talk to new interviewers about the union. WE WILL NOT in any like or related manner interfere with, restrain, or coerce you in the exercise of the rights guaranteed you by Section 7 of the Act. WE WILL, within 14 days from the date of the Board’s Order, rescind all rules restricting you from posting, dis- tributing, or possessing union literature, flyers, or notices in the Houston, Texas office more so than we restrict such posting, distributing, or possessing other nonwork- related materials, and contemporaneously with the re- moval of such restrictions, WE WILL notify you in writing that such rules have been rescinded and that you are free to post, distribute, or possess union-related materials in the same manner that you are free to post, distribute, or possess any other nonwork-related items. WE WILL, within 14 days from the date of the Board’s Order, notify you, especially including new employees, or trainees, that you are as free, during your working time, to talk with other employees about union matters and without notifying supervision of such talk, to the extent that you are permitted to talk about any other non- work-related matter without such notification. GALLUP, INC. Tamara J. Gant, Esq. for the General Counsel. William A. Harding, Esq. and Jack L. Shultz, Esq. (Harding, Shultz & Downs), of Lincoln, Nebraska, for the Respon- dent, Gallup. Douglas P. Fennell, Org. (Steelworkers), of Houston, Texas, for the Charging Party. DECISION STATEMENT OF THE CASE RICHARD J. LINTON, Administrative Law Judge: At the height of the Union’s organizing campaign, Gallup brought in its CEO, Jim Clifton, who delivered an impassioned speech to the assembled employees. Partly because some of Clifton’s ex- pressions used words such as “guns,” the Government attacks the speech as unlawful, arguing that, in effect, one message of the speech was that the entire office should rise up and expel this enemy before it destroys Gallup. The argument suggests that the Union’s supporters may well have recalled the opening stanza from Lord Byron’s, The Destruction of Sennacherib:1 The Assyrian came down like the wolf on the fold, And his cohorts were gleaming in purple and gold; And the sheen of their spears was like stars on the sea, When the blue wave rolls nightly on deep Galilee. 1 See 2 Kings 19:35; 2 Chron. 32: 21; Isaiah 37:36. GALLUP, INC. 1215 This is a discharge case. In the summer of 1999,2 Gallup fired four employees who are the alleged discriminatees here. Finding in favor of the Government as to only a few of the allegations of coercive statements (and not as to CEO Clifton’s speech), I also dismiss as to all four of the dischargees. I presided at this 16-day trial in Houston, Texas beginning November 16, 1999 and concluding on June 27, 2000. Trial was pursuant to the September 30, 1999 order consolidating cases, consolidated complaint, and notice of hearing (the com- plaint), issued by the General Counsel of the National Labor Relations Board through the Acting Regional Director for Re- gion 16 of the Board. Such pleading is based on a series of charges filed by the United Steelworkers of America, AFL– CIO, CLC (the Union), beginning with the charge (later amended), filed May 24 in Case 16–CA–19898, and ending with the charge (later amended) filed August 12 in Case 16– CA–20028. The pleadings establish that the Board has both statutory and discretionary jurisdiction over Gallup, a Delaware corporation, that Gallup is a statutory employer, and that the Union is a statutory labor organization. The pleadings also establish that Gallup has a place of business in Houston, Texas where it is engaged in market research and consulting services for various companies. Among its half-dozen services, Gallup, also known as The Gallup Organization, no doubt is best known for The Gallup Poll that Dr. George Gallup Sr. developed in the 1930s. (JX 23 at 11)3 [JX 23 is a photocopy of the printed and paper-bound RX 35, the “Interviewer Training Manual” (ITM) at Gallup. To better understand the pagination of JX 23, and for back- ground purposes, the parties have agreed (by fax or letter not part of the record) that RX 35, marked at 7:1365 but not of- fered, may be received in evidence. I now receive RX 35 into evidence for the limited purpose stated. Accordingly, counting the cover as page 1 and the cover’s inside as page 1A, the “Welcome” page becomes page 2 and the table of contents pages 3 and 4. As shown on page 1A, the publication date of the manual is January 1999. Although JX 23 contains a few hand printed marginal comments, I have disregarded them as not being part of the exhibit.] “The Gallup Organization as we know it today is the combi- nation of two great companies which came together in 1987.” (JX 23 at 11) The two companies were The Gallup Organiza- tion and Selection Research, Inc. (SRI), with SRI acquiring Gallup. (JX 23 and RX 35 at 12–13) Presumably, Gallup, Inc. is the parent organization of The Gallup Organization, but the record is unclear. Although SRI did the acquiring, the com- bined firm, wisely it appears, puts The Gallup Organization front and center. [In a speech, discussed later, that CEO Clifton gave to the Houston employees on May 26, Clifton describes how SRI purchased the Gallup organization from Dr. Gallup’s estate. JX 6 at 14–16.] As to all these names, including the combined company (whatever its legal name might be), I here 2 Unless otherwise indicated, all dates are for 1999. 3 References to the 16-volume transcript of testimony are by volume and page. Exhibits are designated GCX for the General Counsel’s, and RX for those of Respondent Gallup. apply the short name of Gallup unless there is a need to refer to a specific name. As of the January 1999 publication of the interviewer train- ing manual (ITM) (JX 23 at 2), Gallup operated with “over 2,000 associates in 30 countries.” (JX 23 at 13) Of this num- ber, over 1500 are interviewers (consumer, with over 100 ex- ecutive interviewers). Gallup’s surveys “cover 70% of the entire world’s population.” (JX 23 at 15) At its website, www.Gallup.com, we learn that Gallup’s World Headquarters office is located in Princeton, New Jersey, and that its North American Operations Center is situated at 301 South 68 Street Place in Lincoln, Nebraska. Further, Gallup’s corporate headquarters also are located at the South 68th Street address in Lincoln. I take official notice of the web- site only for background information. Reliability of its data collection process is of the utmost im- portance to Gallup. In Gallup’s “Welcome” page to new inter- viewers from Jim Clifton, President and CEO (the website shows his name as James K. Clifton), Clifton writes (JX 23 at 2): Gallup has a commitment to quality that no other company in our industry can match. Nothing is more important to us than the accuracy and reliability of our data. And three pages later the training manual (ITM) declares to the new interviewer (JX 23 at 5): The “Gallup” name is the most trusted in market research. Known worldwide for The Gallup Poll, our company’s mis- sion is “Helping People Be Heard.” Remember the part about “the most trusted” name in market research, for when we get to my summary of CEO Clifton’s May 26 speech to the Houston employees, we will see that Clifton uses that asserted reputation as his primary theme in order to emphasize his message. The complaint alleges that, beginning about May 14, 1999 and continuing to about August 8, 1999, Gallup, by several named (including CEO Clifton as to his May 26 speech) and unnamed supervisors, engaged in various coercive acts in viola- tion of Section 8(a)(1) of the Act. Gallup denies. The com- plaint also alleges that Gallup unlawfully discharged four em- ployees. Admitting the fact of the discharges, Gallup denies that such violated Section 8(a)(3) of the Act. Named in com- plaint paragraph 25, the four, with their discharge dates, are: Lynne Zieler – June 22, 1999 Janice Rinehart – June 24, 1999 Sherri Lee – June 30, 1999 Patrick Snyder – August 6, 1999 The organizing campaign to obtain union representation ap- pears to have been rather brief. From late April to May 13, 1999, employees Sherri Lee, Lynne Zieler, and others met with Doug Fennell, the Union’s organizer, and formed an organizing committee of 20 employees. (5:885–886; 6:1023–1025) Be- fore July 1, 1999 (when she became Director of Latin American Interviewing), Gisela Uria-Ruiz was the Director of Consumer Interviewing for Gallup’s Houston office. (1:37) Before mid- May Gallup had an interviewing office in Bellaire (one of the DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1216 small incorporated areas surrounded by the City of Houston), but that office was closed on May 14, 1999 and its operations merged into the Houston (north) office at 14405 Walters Road, on the north side of Houston. The old Bellaire office is not involved in this case. (1:41; 11:2144) The (combined) Houston office occupies the second floor (about 17,000 square feet) of a 10-story building. (11:2156, Barlow) A floor diagram (RX 38) is in evidence. The office has some 200 employees, including 7 supervisors. (1:48, 112; 2:226) The employee interviewers work in cubicles, about 200 in number. (3:553, Lambert; 4:742, Snyder) The cubicles stand about shoulder height. (5:880, Snyder; 14:2810, Geiger) As with the other management representatives, Uria-Ruiz’s title also is called supervisor or manager. During this early time frame, Uria-Ruiz was the “lead” supervisor. Each of the seven supervisors reports to Director of Interviewing Jean Timmerman at the corporate office in Lincoln, Nebraska. (2:226; 13:2474–2475) As “lead” supervisor, Uria-Ruiz was not in charge of the other supervisors, for all the supervisors are equal. However, one served as “lead” or liaison in communi- cating with Timmerman on certain project and budgetary mat- ters. (2:225–227) In July 1999, Heidi Roberts (maiden name Russo, 15:2933) became the lead supervisor. (15:2932) Effec- tive July1 Uria-Ruiz became Director of Latin American Inter- viewing. (1:7–8, 37) As we shall see, “Mission Control” is the designated name for a small area, just outside the break room in the top center section of the floor diagram (RX 38), where a supervisor is located and available for assisting, consulting, and directing. (2:345, Uria-Ruiz; 11:2148, Barlow) A bulletin board (the mission control bulletin board) is located on the outside wall of Supervisor Chad Murray’s office and faces the mission control desk. (15:2976, Murray; RX 38) The mission control bulletin board is about 4 feet high, 8 feet wide, and is divided into eight sections, with one of the sections designated as “Other.” (1:48; 2:346–349, Uria-Ruiz; 3:547, Lambert) About 30 minutes before a mandatory meeting scheduled by Uria-Ruiz for 5 p.m. on May 14, Lynne Zieler, Janice Rinehart, Sherri Lee, and several other members of the organizing com- mittee entered Uria-Ruiz’s office and, delivering a paper (GCX 49) warning that managers and supervisors should not violate the law, announced the fact of the organizing campaign. (1:37; 5:888–890, 947–950; 6:986–988, 993–994, 1026–1027; 7:1323–1326; 8:1687–1691; 10:1934; 15:3036) At the 5 p.m. mandatory meeting that May 14, attended by between 100 and 200 interviewers (6:994, Zieler; 15:3037, Uria-Ruiz), Uria-Ruiz made an announcement pertaining to the Bellaire closing, and then said that rumors of some union orga- nizing had now been confirmed, that a group of interviewers had just informed her that there was an organizing drive in progress. Uria-Ruiz encouraged employees to attend any union meetings, to get the facts, and to ask questions. However, Uria- Ruiz made it clear that the “Union is not what Gallup is.” At about that point organizing committee members Janice Rinehart, Sherri Lee, and Mark Anthony “Money” Anderson announced that the first (general) union meeting would be held at a local Holiday Inn that Sunday (May 16). (2:281; 15:3037– 3039) Minutes of the short meeting are in evidence as Joint Exhibit 18. (2:281–282) By letter dated May 16 (RX 5), the Union, by organizer Doug Fennell, notified Uria-Ruiz and Gallup of the in-house organizing committee at the Houston facility on Walters Road, that such activity was protected by law, and that “We look for- ward to a fair and amicable organizing campaign, and a bright future for all at Gallup.” At the bottom of the letter all 20 members of the organizing committee signed their names, with that of Janice Rinehart leading the first column. Counting Rinehart, all four of the alleged discriminatees are listed (along with 16 others not alleged as discriminatees). That Monday, May 17, Uria-Ruiz testified, Janice Rinehart and Lynne Zieler delivered courtesy copies of the foregoing letter (plus a copy of a second letter) to Uria-Ruiz, informing her that the original would arrive by certified mail. It did. (2:283–284) The orga- nizing committee named itself the “We Care About You Com- mittee.” (11:2097, Rinehart; 16:3155, Snyder) Throughout the organizing campaign, some employees sup- ported the Union, while other employees opposed it. (5:953, Zieler) Oddly, it appears that the members of the organizing campaign were “very surprised” to learn that a substantial number of employees opposed unionization of the office. (6:1033, Lee) Record evidence is rather limited about how or when the organizing campaign ended. The campaign (both for and against) appears to have been quite active during May and June, apparently subsiding thereafter, especially so after the organizing committee lost the first three (Lynne Zieler, Janice Rinehart, and Sherri Lee) of the alleged discriminatees to ter- mination in late June. In any event, the campaign never gener- ated enough employee support to persuade the Union to file a petition for an election, and the parties stipulated (2:314) that no election petition was ever filed. Of the 28 witnesses who testified, 15 were called by the Government (one, Uria-Ruiz, called under FRCP 611(c), was recalled by Gallup), which then rested (11:2119), and 13 (in- cluding Uria-Ruiz) were called by Gallup, which then rested (15:3132). There was brief rebuttal and surrebuttal by wit- nesses recalled to testify. The parties have stipulated (RX 87) to the admission of documents containing corrections for each of the 16 volumes of the transcript (GCX 1 (bbb – rrr); RXs 80-86). (RX 86 appar- ently supersedes RX 81.) I had indicated (16:3149–3150) that there was no need to correct typographical errors where the meaning or correct word is clear. Moreover, what sometimes, perhaps frequently, appears to be an error is really the reflection of a mispronunciation by the witness, lawyer, or judge. When the meaning is obvious, the printed error ordinarily needs no formal correction. Also, when the name of a witness is mis- spelled, it need be corrected only once for all such entries. Repeated corrections are unnecessary. The stipulated correc- tions sometimes go beyond these guidelines. Nevertheless, approving the stipulation (RX 87), I now receive in evidence RXs 80–87. GCX 1(bbb) through (rrr), with (sss) being the index, were received earlier. (16:3171) The General Counsel filed a motion, dated November 13, 2000, to reopen the record in order to introduce certain addi- tional documents. Gallup filed its opposition, dated November GALLUP, INC. 1217 27. By order dated November 30, I denied the Government’s motion to reopen the record. By its March 13, 2001 motion, the Government requests that I take official notice of an attached (copy of the) 25-page March 7, 2001 decision, JD(ATL)–16–01, of Administrative Law Judge Pargen Robertson in Gallup, Case 16–CA–20442, a case involving Gallup’s operation in Austin, Texas. Gallup objects by its opposition of March 19. In granting the General Counsel’s motion only to a limited extent, I take official notice of Judge Robertson’s decision (which, as Gallup argues, is not final and from which Gallup intends to appeal to the Board) for the single limited purpose of recognizing it as a background event as to Gallup generally. Also for background purposes, I take official notice that the Government has filed a petition for injunctive relief under Sec- tion 10(j) of the Act involving the matters in this case. That proceeding, styled National Labor Relations Board v. Gallup, Inc., C.A. No. H-99-4223 in the United States District Court for the Southern District of Texas, Houston Division, is pending before the Honorable Ewing Werlein, Jr., United States District Judge. The parties have advised me that oral argument was set for March 2, 2001 before Judge Werlein, and I have informed counsel of an estimated date for the issuance of my decision in this case. On the entire record, including my observation of the de- meanor of the witnesses, and after consideration of the post- trial briefs filed by the General Counsel and by Gallup, I make the following findings and conclusions. Aside from its 171- page primary brief, Gallup submitted a 55-page brief in reply to the Government’s 52-page brief. Judges do not always accept reply briefs,4 and accepting such is at the judge’s discretion. Fruehauf Corp., 274 NLRB 403, 403 fn. 2 (1985). Having reviewed Gallup’s unopposed motion for leave to file, dated October 5, 2000, and as such reply brief arrives at a very early stage in the decision process, I grant Gallup’s motion and I have considered its reply brief. See Salem Electric Co., 331 NLRB 1575, 1578 fn. 3 (2000). Complying with my trial re- quests (11:2042; 16:3184–3186), Gallup’s briefs were quite helpful because in them Gallup lists many facts (giving the names of the witnesses, and the volume number and page of the transcript), as well as arguing the factual and legal issues. Fi- nally, in February of this year the parties, at my request, filed supplemental briefs on a question (arising in Janice Rinehart’s case) governed by the Board’s decision, on July 10 last year, in Epilepsy Foundation of Northeast Ohio, 331 NLRB 676 (2000). The supplemental briefs were helpful. FINDINGS OF FACT A. Procedural Matters—Tapes, Subpenas, and Special Appeals Before the trial opened on November 16, 1999, Gallup served subpenas duces tecum (SDT) on the four alleged dis- criminatees directing each of them to produce any audio and video tape recordings that they may have made (or be in pos- session of) at or near Gallup’s place of business between Janu- 4 See, for example, NACCO Materials, 331 NLRB 1245, 1246 fn. 2 (2000). ary 1, 1999, and the date [November 3, 1999] of the subpena. As it turns out, audio tape recordings, fitting the description in the SDT, had been made, but they had been turned over to the General Counsel (that is, to the Government’s trial attorney) before the SDTs were served on the four dischargees. The issue came up shortly after cross-examination had begun respecting Patrick C. Snyder, the first of the four dischargees to testify. Following, on request, the General Counsel’s produc- tion of Snyder’s three pretrial affidavits (4:751), Gallup in- quired about a reference in one of the affidavits to a tape re- cording of a conversation. (4:767–768) This led to the disclo- sure that tapes, which had relevant information (4:777), had been made and turned over to the General Counsel before Gallup’s SDT (RX 21; 4:774–775) was served on Snyder. (4:769, 771, 778) Gallup demanded production of the tapes that now were in the General Counsel’s possession. The Gen- eral Counsel refused under either the Jencks rule (29 CFR 102.118)5 or under the subpena to Snyder (4:777) and (4:778): MS. GANT: Your Honor, this witness cannot produce what he does not have, and his testimony cannot be stricken for failure to produce something that is not in his possession. JUDGE LINTON: Can a witness hide something from evidence by turning it over to, say, his lawyer? MS. GANT: But, Your Honor, there’s no contention that this witness is hiding something by turning it over to his lawyer. These tapes were given to us before any sub- poenas were served on these individuals. There’s no evi- dence of that [hiding]. JUDGE LINTON: Well, I’ll grant the motion if the tapes are not going to be produced. Gallup moved to strike all testimony of Snyder concerning conversations with supervisors. (4:775–778) Repeating her position, the Government asserted that the tape was not produc- ible under Jencks, and she pointed out that no SDT had been served upon her. (4:786) Essentially granting Gallup’s motion to strike (4:787–788), I relied on Bannon Mills, 146 NLRB 611 (1994), and on my analysis that to do otherwise would be equivalent to permitting a party to hide evidence, in advance of any subpena, by deposit- ing it with his lawyer as a “King’s X” sanctuary (4:783–784),6 something the courts say a client may not do. I also stated that to take the time for Gallup to serve the General Counsel with a subpena (and to request that the General Counsel in Washing- ton, D.C. grant the Government’s trial attorney, or the Regional Director, permission to produce the tapes) would be nothing more than an exercise in wasting time. (4:778, 782–785) The General Counsel announced her intention to file a special ap- peal. (4:780–781) As this development also applied to the remaining witnesses, I adjourned the trial sine die. (4:815) 5 “Those statements were not made to me. Therefore I have no obli- gation to furnish them under Jencks.” (4:772–773) 6 A lawyer cannot be used as a shield against discovery of relevant documents. See E.S. Epstein, The Attorney-Client Privilege and the Work-Product Doctrine 116-117 (3d ed., ABA, 1997). DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1218 During the adjournment the General Counsel filed her spe- cial appeal of November 24, and Gallup filed its December 2 brief in opposition. Other filings were made, including Gallup’s motion to reconsider, in which confirmation of a fact was established. By its order of March 1, 2000, the Board (Members Liebman and Hurtgen, with Member Brame dissent- ing) granted the General Counsel’s special appeal and reversed. The Board’s order (with the dissent) is rather lengthy at 11 pages, but the bottom line is this: Gallup would have to take the route of a written request as required by 29 CFR 102.118. (The dissent argued that the General Counsel should have moved to quash the SDTs or simply have alerted Gallup, when the SDTs were served, that she, not the witnesses, had possession of the tapes, so that Gallup could then have made a written request that consent be granted for the General Counsel to produce the tapes. As of the Board’s March 1 order, however, it was not clear when the General Counsel learned that the SDTs had been served.) Responding to Gallup’s motion for reconsideration, the Gen- eral Counsel candidly stated that, before trial, she was aware that the SDTs had been served. The Board’s subsequent order, as corrected on May 31, 2000, ruled as before. (Member Brame, dissenting, would have granted Gallup’s motion for reconsideration.) In its May 31 order, the Board also makes clear that my ruling that a 102.118 request was “unnecessary” was the point of the first order. Thus, “It was precisely that ruling that was reversed by the Order of March 1.” Finally, I report a possibly significant dictum by the Board in its order of March 1. First, recall from Delta Mechanical, Inc., 323 NLRB 76, 77 (1997), that Jencks does not apply to voices of an event in progress captured on a tape recording because such recordings are “direct evidence” of the event and as direct evidence such tapes would be “subject to a subpena duces te- cum.” However, in footnote 3 of its March 1, 2000 order, the Board writes: Although Sec. 102.118 contains a Jencks exception to the re- quirement for requests for General Counsel’s consent, no party contends that the tapes fall within this exception. [This was clarified in the second order, denying the motion for re- consideration, to state “The Judge found that the rule did not apply, and no party requested special permission to appeal that ruling.”] We therefore do not pass on this issue. How- ever, we note that, in our view, the mere fact that a witness’ statement is made on tape, rather than in a writing, is not per se a basis for excluding it from the Jencks rule. If all the Board means by its last sentence of footnote 3 is that a witness “statement” submitted, or taken by a Board agent, on a tape recording and describing past events, as distin- guished from an event in progress captured on a tape recording, then nothing is new. Clearly that tape recording would be pro- ducible under the Jencks rule. But does the context of the en- tire footnote 3 suggest that the Board is signaling a change? (After all, the tapes here were not statements dictated on tape rather than placed in an affidavit. The tapes here captured voices of events in progress—direct evidence.) That is, by its footnote 3, is the Board indicating that it now disapproves of the cited rule of Delta Mechanical? True, such an indication in footnote 3 would be dictum. But dictum “can be persuasive authority.” Ayoub v. I.N.S., 222 F.3d 214, 215 (5th Cir. 2000). It seems unlikely that the Board intends, by its footnote 3, to indicate any disapproval of or departure from Delta Mechani- cal. Overall (and also in light of the Board’s order of May 31), therefore, I interpret footnote 3 as suggesting nothing new, and that its reference to a “statement” on tape simply is a reference to a typical statement that, instead of typed, is dictated (describ- ing past events) onto a tape recording. And such a “statement,” as footnote 3 asserts, is producible under the Jencks rule. As for a tape recording by which a witness has captured, on the tape, voices (audio) or sights and sounds (video) from some event in progress, then that, under Delta Mechanical, would be a tape of “direct evidence” producible under subpena duces tecum. And, per the instant case, make sure that a written re- quest also is made under 28 CFR 102.118. If permission is not granted to produce, then (the Board states in its March 1 order at 2 fn. 4), the presiding judge “should decide any issues flow- ing from the General Counsel’s response,” such as, presumably, a petition to revoke the SDT. B. Allegations of Coercion 1. Introduction Complaint paragraphs 8 through 24 allege coercive acts. Aside from paragraph 23, concerning CEO Clifton’s speech of May 26, most of the paragraphs deal with distributing or post- ing union literature, and most of the dates are for May 1999. 2. Supervisor Gisela Uria-Ruiz a. May 14, 1999 (1) Facts Complaint paragraph 8 alleges that, about May 14, Gallup, by Uria-Ruiz, told employees that “distribution of literature was prohibited at the center.” Gallup denies. Following the 5 p.m. mandatory meeting, and with the other supervisors present in her office, Uria-Ruiz called their com- mon superior, Jean Timmerman, at the Lincoln corporate of- fice, and spoke with Timmerman and Jane Miller, director of operations. (2:326–327) When the conference call ended and Uria-Ruiz opened her door, Sherri Lee was there posting a flyer about a union meeting on Sunday, May 16, that Janice Rinehart (JX 18 at 1) had announced earlier at the 5 p.m. mandatory meeting. Uria-Ruiz testified that she told Lee that such distri- bution was prohibited, that it was not authorized, that “[w]e never had allowed distribution in the office.” (1:48–49; 2:327; 15:3040) Sherri Lee replied that Gallup had permitted the distribution of a religion flyer about a prayer meeting. Uria-Ruiz advised Lee that neither she nor any of the managers had approved the distribution of the religion flyer. Lee responded that it made no difference because such distribution had set a precedent and that now “we have the right to distribute.” Uria-Ruiz said that she would check with Gallup’s lawyers and get back to Lee. (2:327–328, 336; 15:3040) The religion flyer (RX 17), announcing nondenominational prayer meetings to begin May 13, apparently was distributed by Michael Lee, an executive interviewer, on May 13. (2:328, Uria-Ruiz) Uria-Ruiz testified that no supervisor approved GALLUP, INC. 1219 Michael Lee’s distribution of the religion flyer. (2:327; 15:3041) Uria-Ruiz testified that before May 27, Gallup had followed an unwritten policy that prohibited distribution of nonwork material in the office work areas. (2:327, 360) For example, Uria-Ruiz described an occasion in about January 1999 when Jack Burrows, a former executive interviewer, asked Uria-Ruiz if he could distribute his business card. (Burrows sold health and dental insurance on the side.) Uria-Ruiz denied the re- quest, although she permitted him to leave some cards at the front desk and to publish a note in Gallup’s newsletter, “Cur- rent,” because selling insurance was something commercial or “business” in nature. (2:334–335, 347) Following her consultation with Gallup’s lawyers, Uria-Ruiz testified (2:337; 15:3041), Gallup relaxed its policy and on May 27 Uria-Ruiz issued (1:48; 2:331, 337) to all interviewers a memo reading (JX 2): We have had many questions about distribution of written material by employee’s on matters which do not involve business. There is no change in policy, but this memo is for the purpose of making sure everyone knows the policy. 1. In general—Gallup allows non-business material to be distributed by employees at work. We allow such dis- tribution in the break room and at interviewer desks. We recently allowed such distribution by a prayer group and by pro-union employees. A supervisor must be notified before any such material is distributed. 2. Exception—This right to distribute does not apply to the new interviewer desks. 3. Prohibition—Employees distributing material or lit- erature are prohibited from disrupting employees who are working. If a co-worker is placing a call or otherwise working, you may not interrupt them. 4. Posting—Such material can be posted at your work area or in the break room. It may also be posted on the bulletin board by Mission Control if approved in advance by a supervisor. We will continue to follow this long-standing policy if it is not abused. Please respect the rights of your co- workers. Notwithstanding Uria-Ruiz’s testimony about an “unwritten” policy prohibiting the distribution of nonwork materials in the office work areas, there is an abundance of testimony that no supervisor had ever told anyone of such “unwritten” policy. Moreover, the evidence is that distribution and posting was practically unrestricted. (3:588, Robinson; 3:608–609, Juneau; 4:685, 687–688, Snyder; 5:895, Zieler; 6:1038, 1041–1042; 7:1334, Lee) A partial exception existed, at times, when there might be a partial removal of some items posted by employees in their work cubicles just before a visit by one of Gallup’s clients. Indeed, Supervisor Uria-Ruiz asserts that she, or other su- pervisors, approved much of what was done. Thus, before the advent of the union organizing in mid-May 1999, Uria-Ruiz approved postings on Gallup’s main bulletin board for sales of dogs and a notice of a free rabbit. (2:347–348) Such postings on the bulletin board, which always have required supervisory approval, are posted on a specific section denominated, “Other.” (1:48; 2:346–349) Notice of a Cuban festival had been posted there in the past, but Uria-Ruiz declined to approve it on the last request. (2:348) Uria-Ruiz reports that notice of the Cuban festival was posted in the break room (2:350), where Gallup always has permitted practically anything to be posted (2:345–346, 348, 352). Uria-Ruiz concedes that Girl Scout cookies will always be sold, but she contends that such is different from someone so- liciting for “a type of business.” (2:351) As to the latter, in addition to Uria-Ruiz’s rejection of the sale of insurance, men- tioned earlier, Gallup (it is not clear whether Uria-Ruiz or an- other supervisor acted) rejected the sales effort of Herbalife in 1998. (2:351–352) But where insurance and Herbalife solicitations for sales were barred, Janice Rinehart succeeded in obtaining the per- mission of Supervisor Douglas Barlow to distribute color flyers (GCXs 52, 53) advertising water pillows that she and her hus- band sold in a sideline business that they operated. Rinehart did not post any of the copies. (8:1701–1703; 10:1935) This occurred either in 1997 (8:1701, Rinehart), 1998 (3:606, Jun- eau), or early 1999 (5:894, Zieler). Then there is the gourmet sauce and meatballs that Sherri Lee sold in 1998 for a person who distributed such items as a commercial venture. Lee did not request permission before she posted (breakroom; around the start up computers; and in the hallways) and distributed literature promoting the product. Supervisors were aware of her actions because at least one supervisor, Dennis Welch, bought some of the product. (6:1044–1045, 1341) The variety of endeavors seems endless. As Patrick Snyder and Sherri Lee describe, there were fundraisers for employees who had cancer (6:1042), or who were otherwise sick (4:682; Nancy Sico in 1998, 6:1043), or who simply needed financial aid (4:682; 6:1042). Several of the supervisors, including Uria- Ruiz, contributed to the collection for one of the cancer victims. (6:1043) Notices were posted on doors and walls as well as in the cafeteria and break room. (4:683) There also were the usual sales of cars, boats, and furniture with the notices posted on the cubicles of the employees or placed on many of the desks of the other employees. (2:369, Trovato; 4:682, Snyder; 6:1038, Lee) Last, but certainly not least, we have postings inside the em- ployees’ cubicles. Here by practice, and by policy, the postings were virtually unrestrained—as some of the sexually suggestive photographs in evidence (GCX 26) reveal. On one occasion, in February 1997 (about a month after she was hired), interviewer Catherine Wagley complained to Supervisor Tom Langenegger about a large photo of a woman in a very small bikini. Lan- genegger told Wagley that the interviewers could post anything they wanted to in their individual booths. (2:431–432, 434– 435) The photos in evidence (GCX 26), and the testimony of Supervisor Douglas Barlow (12:2234), confirm Langenegger’s pronouncement. As the aisleway side of the cubes are open, the postings in- side the cubes are visible to anyone walking by. (2:434) When a Gallup client was scheduled to come through on a tour, the interviewers were informed of this and asked to tidy up their DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1220 cubes. (2:320–321, Uria-Ruiz; 2:401, 436, Wagley; 12:2233– 2234, Barlow) In practice, this generally meant straightening up papers and discarding any trash, not removing any photos or other nonwork items. (2:402, Wagley; 3:522–523, Lambert; 3:634–635, Juneau) Thus, normally most of the nonwork items, even the “cheesecake” and “beefcake” photos posted inside the individual cubes, remained untouched. (2:402–403, 435–436, Wagley; 3:522–523, 558–559, Lambert; 4:740–741, Snyder) However, when the “cheesecake” disrobed to the point of displaying nudity, such as exposing a woman’s breasts, then it appears that some effort was made by some supervisors to come by, remove such nude photos (or ask that they be re- moved temporarily), and place them in the interviewer’s desk drawer on those occasions whenever a client was about to visit. (5:864, Snyder; 15:3010, Solis).7 Although supervisor Doug Barlow agrees that it was the de- sire to remove such displays of nudity before any client visit, he concedes that he never instructed any interviewer to remove any such items. And even though other supervisors have told him that they have done so, he has no personal knowledge con- firming their reports. (12:2234, 2387) Before May 1999, Su- pervisor Chad Murray testified, Murray had never removed literature from an interviewer’s cube except before a client visit or simply to clear off a desk to free it up for availability. (15:2986–2987) In light of the foregoing, and all the record, it seems clear, and I find, that before May 1999, Gallup’s policy and practice was to permit employees to distribute and post, for example, sales of personal items such as cars, boats, furniture, and pets, to announce fundraising drives for needy employees, and to solicit contributions from one and all on behalf of such em- ployees. (Permission was required to post in the “Other” sec- tion of the main bulletin board.) As for the individual booths of the interviewers, postings inside such cubes were virtually un- restricted with the sole exception that sometimes, in preparation for client visits, employees would be asked to remove any pho- tos of nudity, or supervisors, on occasion, would remove such photos temporarily, while leaving posted the other nonwork items. However, things changed once the union literature be- gan to appear. In making the foregoing findings, I also have considered the fact that the Houston office has a Gallup-sponsored program called “Community Builders.” [Although Patrick Snyder (4:760) and Sherri Lee (7:1292) each claims to have gotten the program started in Houston, both agree that Lee was the chair- person of the committee once the program was launched in Houston.] Under this Gallup-sponsored program, Gallup matches any funds collected to help the community. Flyers that circulate announcing, for example, fundraising activities such as bake sales or garage sales do not need approval because that is part of the company-sponsored activities of Community Builders. (2:329–330, Uria-Ruiz) A separate bulletin board is, 7 About early 1996, Bobbie Solis was asked, in preparation for a cli- ent’s visit, to remove a cartoon (RX 64) that the client might find offen- sive. (15:3010–3011) The cartoon shows two older women, one in a string bikini with her breasts sagging straight down below the bikini’s top. or at least was, maintained outside a supervisor’s office just for the program. (15:2972, Roberts) Even though the record is less than fully developed respecting the point, it appears that the personal postings described in the record were separate and unrelated to the Community Builders Program. (Indeed, the parties have not even briefed the point.) (2) Conclusion As Supervisor Uria-Ruiz told Sherri Lee that Uria-Ruiz would check with Gallup’s lawyers, and as the meeting notice that Lee and others were distributing and posting was, it ap- pears, posted and distributed, I find nothing coercive in Uria- Ruiz’s statement on this occasion. Accordingly, I shall dismiss complaint paragraph 8. Some additional evidence bears on complaint paragraph 24 (regarding a directive to remove, temporarily, posted union items during a client visit in August). I address complaint paragraph 24 later. b. May 27, 1999—The distribution memo (1) Facts Complaint paragraph 15 alleges: About May 27, 1999, Respondent issued a memorandum prohibiting distribution of literature at new interviewers’ desks, requiring supervisory notification prior to any distribu- tion of literature at work, and requiring supervisory approval prior to posting literature in the breakroom, at employees’ work stations, and on the bulletin board by “Mission Con- trol.” By its answer, Gallup “partially admits” the allegation, “but denies the portion of complaint paragraph 15 concerning super- visor approval prior to posting literature in the break room and at employees’ workstations.” As one of its affirmative de- fenses, Gallup pleads (defense VI): At all material times, the Respondent continued to apply its pre-existing policy concerning the distribution of literature and the posting of material and made no exceptions concern- ing the distribution or posting of either pro-union or anti- union material except that Respondent relaxed its preexisting policy concerning prior approval for distribution of literature after the unauthorized distribution in early May 1999 by em- ployee Mike Lee of certain religious oriented-material. Fol- lowing such distribution, the Respondent relaxed its policy and only required prior notification for the distribution of non- work-related literature in its facility. Complaint paragraph 27 alleges that the conduct alleged in several paragraphs, including paragraph 15, constitutes a viola- tion of Section 8(a)(1) of the Act. Gallup denies. (2) Discussion Citing (Br. at 15) “Eaton Technologies, 322 NLRB [848] (1997)” [the page cite is 848, and the jump cite is 853–854], the General Counsel argues for finding a violation. However, I find merit only respecting the notification required to distribute. Even then, however, I find no merit to the extent the allegation attacks item “2. Exception” (respecting distribution on the desks of trainees) of Uria-Ruiz’s May 27 memo (JX 2). GALLUP, INC. 1221 Respecting the postings, complaint paragraph 15 reflects a misreading of Uria-Ruiz’s May 27 memo in that, as Uria-Ruiz testified (2:318, 345–348), postings at workstations and in the break room are essentially unrestricted. The approval restric- tion in paragraph 4 (JX 2) pertains to the mission control bulle- tin board. Crediting Uria-Ruiz’s unrebutted testimony, I find that prior approval has always been required for posting on mission control’s bulletin board in the “Other” section. No evidence was presented that, as in Eaton, such requirement was honored only, or even mainly, in the breach. Later I address the matter of distributions at the desks of the trainees (new interviewers). As I there find, the evidence is insufficient to show that management ever knowingly permitted distributions at the desks of the trainees. Accordingly, I find no merit to complaint paragraph 15 to the extent it attacks the “Ex- ception” at item 2 pertaining to distributions at the desks of the trainees. In light of the foregoing, the merit I find to complaint para- graph 15 is limited to the notification requirement specified in item 1, “In General,” of Gallup’s May 27 memo (JX 2). Before the advent of the Union’s organizing campaign, such distribu- tions were unrestricted. Accordingly, to this aspect of com- plaint paragraph 15, I find merit, and therefore find that the May 27 notification requirement violated Section 8(a)(1) of the Act. I shall dismiss the balance of complaint paragraph 15. c. June 4, 1999—The disputed “too” (1) Facts About June 4, complaint paragraph 20 alleges, Gallup, by Supervisor Uria-Ruiz, “instructed employees that they were required to secure supervisory approval before distributing union literature and that they were prohibited from such distri- bution at new interviewers’ desks.” By its answer, Gallup “partially admits,” but denies “the al- legation concerning supervisory approval before distributing union literature.” As its affirmative defense number IX, Gallup avers: At all material times, the Respondent has adhered to its pre- existing policy concerning the prohibition against distributing any material on new interviewer desks which is not directly related to the orientation process for new interviewers, and made no exception with respect to the distribution of either pro-union or anti-union material. Up to a point, the parties roughly agree on what happened, and the meeting in question was tape recorded. Unfortunately, there were half a dozen persons in the meeting, and frequently several of them spoke at the same time. That is true of the critical point on the tape. The event begins about 5 p.m. that June 4 when Eileen Jun- eau and a couple of the other members of the Union’s organiz- ing committee were about to begin distributing copies of a un- ion leaflet (GCX 31) to the desks of the interviewers. Before she began passing out her copies, Juneau, pursuant to the rule set forth in Supervisor Uria-Ruiz’s May 27 memo (JX 2), showed supervisor Chad Murray a copy and notified him that she and the others would be passing out copies of the flyer at that time.8 Murray [or perhaps it was Roberts] said, “Okay.” Juneau distributed her copies in a couple of areas, visited the ladies room, and moments later learned that she was being paged to report to mission control. At mission control Supervi- sor Charles Davenport informed Juneau that she was to report to supervisor Uria-Ruiz’s office. (3:625–628, 659–660, Jun- eau) Before we consider the events in Uria-Ruiz’s office, examine first the union leaflet (GCX 31) that Juneau and the others were distributing that June 4. On a legal-size sheet of paper, the message of the leaflet urges employees to sign union authoriza- tion cards, such as the copy of the one reproduced at the top of the leaflet, so that the NLRB would conduct an election. The card reproduced at the top of the leaflet appears to be a copy of an authorization card of the United Steelworkers of American, AFL–CIO, CLC. The copy (not overmarked by “Sample” or “Not Valid” or some similar term) is marked off by border lines on all four sides, and has lines for (among other items) name, phone, address, date, and signature. Five questions and an- swers appear under the reproduced card and pertain to the pur- pose or effect of a signed card. Turn now to the meeting in Supervisor Uria-Ruiz’s office. Present in Uria-Ruiz’s office that June 4 were Supervisor Uria-Ruiz, Supervisor Chad Murray, and telephone interview- ers Eileen Juneau, Sherri Lee, Janice Rinehart, and Lynne Zieler. (1:49–50, Uria-Ruiz; 3:628, Juneau; 8:1710–1711, Rinehart) The voices of this meeting appear on a tape (GCX 45) running in Janice Rinehart’s tape recorder that was secretly carried by Zieler. (5:902; 6:982, Zieler; 8:1710–1711, Rinehart) Rinehart did a hand-transcription (RX 37, with- drawn, 10:1943, 1947; RX 78, 16:3147) and submitted that to the Houston office of NLRB Region 16. (10:1940–1942). Other than making a glancing comparison at the trial, Rinehart never made a line-by-line comparison of the typed transcript (GCX 44) with her handwritten (RX 37) version. (10:1942) Based on her brief scan of the 17-page (only one word on page 17) typed transcript (GCX 44) while in the witness chair, Rinehart did testify that the typed version matches the tape. (8:1712) That statement is mostly worthless. Had there been only two participants, or were the tape of a speech, perhaps someone with a very good memory could credibly testify about such a matching. With this tape, and with several of the atten- dees frequently overspeaking one another, such authentic- cating testimony is less than valuable. The record also contains Gallup’s 16-page typed version (RX 67) of the tape recording. There are many differences between the two typed (GCX 44; RX 67) transcript versions. Fortunately, that generally is not a problem because at trial the parties announced that they rely on their transcripts only as to the one critical point in dispute. (10:1944–1947; 15:3053–3055; 16:3139–3144) To my ques- tion of what I should do if I could not resolve the disputed point by listening to the tape, the General Counsel replied that, in the absence of the parties utilizing some enhancement system, I 8 Supervisor Heidi Roberts suggests that it was she to whom Juneau gave such notification. (15:2957–2958, 2970–2971) I need not resolve the difference. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1222 would have to rely on the “credibility of the witnesses.” (16:3144) By my fax to counsel on June 28, 2000, I informed them that I had listened several times to the disputed portion. In relevant part, I informed them: This morning I played the disputed portion of GCX 45 (GCX 44 at 5; RX 67 at 4), that being the specific line, “To pass around [too].” After several tries, all I can be sure of are the first three words, “To pass around.” Then a lot of fast talking by one or more persons. The “too” may or may not be there. At this point it appears that I will have to rely on the witnesses as to what was said. A more recent effort yields a result no better than the one de- scribed above, although some of the surrounding words can be understood. While some enhancing service might have the technical equipment and ability to separate and slow the voices so as to permit piecing together who said what, without the benefit of that service, the mass confusion of voices on the tape at the critical point causes me not to rely on the tape concerning the disputed “too.” The disputed point arises in this fashion. As the two tran- scripts reflect, for the first few minutes the group discusses the reproduction of the union authorization card on the leaflet. Supervisor Uria-Ruiz’s concern was whether the reproduction should have “Void” printed across the face to keep it from be- ing a valid blank card. This appears to be tied to her apparent concern that authorization cards, as a form of solicitation, should not be posted, and that there has to be permission to post. (1:51; 15:3048–3049) Both transcripts then show, in leading into the critical part, that Uria-Ruiz states that she has seen what the (union supporters) are passing around. “No one has come to me and said can we post it/this. And yesterday I asked Heidi [Roberts], did somebody ask you because that’s one of the rules, that they have to ask permission.” (GCX 44 at 4; RX 67 at 4) Juneau responds, id., “To have it posted.” (Juneau’s re- sponse is an agreement that, under the May 27 memo, any em- ployee or employee group had to ask permission to post some- thing on the bulletin board at mission control.) To this Uria- Ruiz’s disputed passage reads, id. ”To pass around [too].” That is, if the “too” is included, it would show Uria-Ruiz replying that permission was required not only to post on the bulletin board at mission control, but it also was required if there was just going to be a distribution. (The latter would be contrary to the May 27 memo which states, in paragraph 1, that notification is required before distribution.) The Government’s version contains the “too,” indicating that permission (as well as notification) is also needed to distribute. (GCX 44 at 5) Taking the opposite view, that Uria-Ruiz did not say the “too” because Uria-Ruiz knew full well that permis- sion was not needed to distribute, Gallup’s version does not have the “too.” (RX 67 at 4) The parties stipulated that the next four lines read (16:3139): Juneau:But: But not to pass it out. Uria-Ruiz: Yes ma’am! Juneau: On that memo, no ma’am. Uria-Ruiz: Okay. Juneau’s reference to “that memo” is to Uria-Ruiz’s May 27 memo (JX 2), and Juneau is telling Uria-Ruiz that in the May 27 memo Uria-Ruiz specifically states that permission is re- quired to post (on the bulletin board at mission control), but that for distribution, only notification is required. At that point, although the next line is not in full agreement on the two ver- sions, Juneau apparently states that she will go get the memo. Indeed, she testified that she went and retrieved a copy of JX 2. (3:629–630) As both transcripts reflect, although their entries are not fully in accord, Juneau returned shortly and apparently showed the memo (JX 2) to Uria-Ruiz, followed by references to notification and to posting. (GCX 44 at 6; RX 67 at 5) Al- though even here the tape has several voices talking at once, both transcripts are fairly close to agreement on what was said. According to Juneau, when she brought in the memo and pointed out the notification and the posting paragraphs (and this seems to be reflected on the transcripts), Uria-Ruiz assertedly replies (3:631, 672): Well, I understand that, but you still have to get it approved. The Government’s version of the transcript has Uria-Ruiz’s response reading (GCX 44 at 6): I understand that, [and here Gallup’s version picks up and generally agrees for the rest of the response] but you see that’s fine and the same thing goes to them, let me tell you one thing. Can you lower it a little bit? The notification, the noti- fication doesn’t give, doesn’t give the right to anybody to come and say, this is what we are doing, (slammed paper down on the desk) you think. To this, on both transcripts, Juneau states that she does not do it that way, and that she merely handed the item to Uria- Ruiz. Supervisor Uria-Ruiz replies that she was describing what she had seen in the past. The point here is that both transcript versions, and especially the Government’s, do not support Juneau’s testimony that Uria- Ruiz said, “Well, I understand that, but you still have to get it approved.” Indeed, the Government’s beginning phrase (Gallup’s version has “CNDW”—cannot determine word(s), 8:1715—for the important opening phrase) virtually matches Juneau’s testimonial version, “I understand that, but you . . . .” First, it seems illogical that Uria-Ruiz, looking there at her own memo of May 27 (JX 2), would say that after notification that they still had “to have it approved.” That makes no sense. By contrast, the transcript version shows that “the same goes for them” is an apparent reference to the antiunion group that had posted the “Grim Reaper” flyer (GCX 48), so named because it contains (3:630) a sketch of the “grim reaper.” (When Juneau left the meeting to get a copy of Uria-Ruiz’s May 27 memo, the conversation in the meeting, as the transcript versions reflect, turned to whether a supervisor had given permission for the antiunion group to post the “Grim Reaper” flyer.) In light of everything, I do not credit Juneau’s assertion in this regard. On this point, I find that the Government’s version of the transcript shows the response that Supervisor Uria-Ruiz gave. (GCX 44 at 6) That brings us back to the main point in dispute. Did Uria- Ruiz also say that the rules (the rules stated in her memo of GALLUP, INC. 1223 May 27) required permission not only to post, but also to dis- tribute, as in, “To pass around, too”? Did Uria-Ruiz somehow get confused and incorrectly restate the rules that just a few days earlier she had issued her memo to make sure that “every- one knows the policy”? (JX 2) Note this testimony of Uria-Ruiz. Called as one of Gallup’s last witnesses, Supervisor Uria-Ruiz testified that in the ex- change quoted earlier (“Juneau: But, not to pass it out. Uria- Ruiz:Yes ma’am!”) that she, Uria-Ruiz, was “agreeing with” Juneau. Thus (15:3047): Q. Okay. Please continue. A. Okay. So Eileen says, “To have it posted.” And I was saying, “To pass around you have to have prior notifi- cation.” I didn’t even get to say that. I got interrupted. Okay. And that’s when she said, “But not to pass it out.” And I say, “Yes, ma’am.” I’m agreeing with her, that, yes, not to pass around; they don’t have to go ahead and [ob- tain] permission to pass around. Q. So if you look at page 4 of Respondent’s 67 [GCX 44 at 5], Eileen Juneau was saying essentially that you [any employee] didn’t have to get permission to distribute literature, and you were agreeing with her. A. Right. There are two problems with this assertion by Uria-Ruiz that by her “Yes ma’am” she was telling Juneau that Supervisor Uria-Ruiz agreed that no permission was needed to distribute. First, as we see in court testimony all the time, a question (or a declaration) that contains a negative (as in, “But, not to pass it out.”) virtually always, when the witness intends to confirm, elicits a reply that is negative in form so as to conform, or af- firm, the negative in the question. Here, therefore, the usual affirmation would have been, “No ma’am”—unless the tone would have indicated an intent to contradict. Judges frequently have to urge the lawyers to omit negatives from their questions, even on cross examination, in order to obtain clarity in the tes- timony. Second, clearly Eileen Juneau understood Uria-Ruiz’s “Yes ma’am” to be a contradiction of what Juneau had just said (that permission was not needed to distribute), and that is why Jun- eau left to get a copy of Uria-Ruiz’s May 27 memo (JX 2). Third, when she was testifying as a witness called by the Government under FRE 611(c), Uria-Ruiz asserted that at the meeting she said that as to distribution the people distributing had to give prior notification. To this Juneau assertedly said, “Oh, that’s not so.” Uria-Ruiz replied, “Yes, it is so.” Uria- Ruiz there reports that Juneau left the office and returned with a copy of Uria-Ruiz’s May 27 memo. Looking at the memo at the meeting, Uria-Ruiz pointed out that one of the paragraphs called for prior notification on distributions. (1:50–51, 54) As to this third point, the transcripts do not support Uria- Ruiz. At that point the participants were talking about permis- sion, not notification. Even the tape can be understood so as to confirm that after Uria-Ruiz completed her statement about “ . . . because that’s one of the rules that they have to ask per- mission” [to post], Juneau (as identified by “E” for Eileen on the transcripts) says, “To have it posted.” To this Uria-Ruiz (as identified by both transcripts) states, “To pass around” [too]. (The “too” added in GCX 44 is disputed.) As to the topics being covered at this spot, I find the tape and the transcripts more reliable than Uria-Ruiz’s version. I therefore do not credit Uria-Ruiz on the point of the topic or her description of what was said at that point. Although Chad Murray, the other supervisor in the meeting, testified as one of Gallup’s witnesses on other subjects, he was not asked specifically about this meeting. He does assert that, after May 27, he had no discussions with any employees about the posting or distribution of union materials. (15:2981) In fairness, Murray probably interpreted that question as asking whether he, as the only supervisor present, had any such dis- cussions. Did Uria-Ruiz add the “too”? Eileen Juneau asserts that Uria-Ruiz did. (3:629) And on redirect (3:667), “That I had to get permission to pass it out and to post it.” This assertion, however, strains under the weight added during cross- examination. Thus, she also states that Uria-Ruiz said that employees must give notice before making a distribution. (3:660) And on recross examination (3:699), Juneau agrees that Uria-Ruiz also said that the employees needed to notify (man- agement) before making a distribution. The problem with Juneau’s concessions about notification statements is that the transcripts do not show Uria-Ruiz saying that about notification. As already summarized, the brief refer- ences to notification are after Juneau returns to the meeting with a copy of the May 27 memo. When Juneau points to the item there about notification, Uria-Ruiz shifts the focus slightly by asking that the union group lower their gusto a bit when delivering such notice, for notification does not mean that someone can come in and “slam” (editorial addition by Janice Rinehart at GCX 44 at 6, although internally consistent with dialogue that follows) a flyer down on the desk while saying, “This is what we are doing.” The General Counsel claims that three of the four employees (Juneau, Rinehart, and Zieler) testified “unequivocally” that Uria-Ruiz said that, for distribution, not only was notification necessary, but also permission. Sherri Lee, although present, only recalled that whatever action the employees were seeking to do was being prohibited by Uria-Ruiz. (Br. at 17–18) Hav- ing surveyed the “unequivocal” report of Juneau, we turn now to the testimony of Janice Rinehart (which, incidentally, is not quoted on this topic or cited by any page reference in the Gov- ernment’s brief). Recall that Rinehart left the meeting moments after Juneau went to get her copy of Uria-Ruiz’s May 27 memo. And, as shown by the transcripts (GCX 44 at 5; RX 67 at 4–5), it was Janice Rinehart who then raised the subject of whether man- agement had given permission for the “Grim Reaper” flyer (GCX 48) to be posted. Once that question was answered, Rinehart departed right at the line ending with her editorial addition, “I left the room.” (8:1712, 1717; GCX 44 at 5) That means that Rinehart was present during the critical ex- change that includes the disputed “too.” However, Rinehart is never asked what was said, and she never addresses that spe- cific exchange. Lynne Zieler’s description is brief (5:900): DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1224 Gisela [Uria-Ruiz] told Eileen [Juneau] that she did not like the way that she was handing out the pro-union literature [this is Zieler’s individual interpretation of what the transcripts show Uria-Ruiz said, but it certainly is not a quote] and she was not getting permission to hand it out. And Eileen said, “We do not have to have permission to hand the literature out. We only have to have permission if we want to post it.” And Gisela said, “No, ma’am.” And Eileen said, “I’ve got your memo, I’ll go get it.” She left the office at that point. While she does not quote a “too,” Zieler’s brief description generally is consistent with Juneau’s (direct examination) ver- sion as well as the version reflected on the Government’s tran- script, including the stipulated lines following the disputed “too.” Sherri Lee initially stumbles in her description, at first show- ing Juneau disputing as to the posting policy. But Lee corrects herself to focus on the policy about distribution, asserting that the May 27 memo “did not specifically state that we had to get approval; only that we needed to show them [management] the documents.” (6:1036) Of course, this does not report what was said. Lee tells us that Juneau went for her copy of the May 27 memo. Moments later, asked whether the flyer under discus- sion at the meeting was a posting or a distribution, Lee rambles off on a tangent. At this critical juncture, and with no objection to the leading nature of the question, the General Counsel asks, and Lee replies (6:1037): Q. But whatever you were doing with this document this day, Uria-Ruiz was contending that it needed — that the Company had to give permission first? A. Yes. Because the situation calls for clarity in the testimony, and clarity is lacking here, I find Lee’s description to be of no value, and I attach no weight to the “Yes” elicited by the Gen- eral Counsel’s leading question. The General Counsel, sum- ming up, asserts (Br. at 18), “The specific testimony of these four witnesses [actually only two, for Rinehart does not address the topic, and Lee, unable to recall any specifics, is unable to add any clarity to the evidence] should be credited over the self-serving denial of one supervisor.” (2) Discussion At this June 4 meeting, I find, Supervisor Uria-Ruiz, appar- ently becoming momentarily confused, told the four telephone interviewers (Eileen Juneau, Sherri Lee, Janice Rinehart, and Lynne Zieler) that for distributions they needed permission as well as notification. I make this finding without regard to the disputed word “too.” Even without that disputed word, both transcripts reflect that Uria-Ruiz’s phase “To pass around,” in relation to the remarks immediately before and after, are ad- dressing the subject of permission. Juneau asserts that such applied to posting. Uria-Ruiz said, “To pass around . . . .” Whatever Uria-Ruiz said at the end, or would have said, I do not credit her testimony that it pertained to authorization and (as I described earlier) that she agreed with Juneau’s next statement, “But, not to pass it out.” The fact is that the nature of the language on the undisputed portion of both transcripts, and the undisputed fact that Juneau went to get a copy of Uria-Ruiz’s May 27 policy memo, dem- onstrates that Juneau obviously disagreed with Uria-Ruiz’s statement. Accordingly, I find that Supervisor Uria-Ruiz said, in effect, that distributions require permission as well as notifi- cation. Nevertheless, I find no merit to this portion of complaint paragraph 20. First, the allegation avers that Uria-Ruiz singled out “union” literature. Not so. Uria-Ruiz was repeating (incor- rectly at that moment, as I have just found) the policy she had announced in her May 27 memo. When Juneau, moments later, returned with a copy of that May 27 memo, Uria-Ruiz fails to correct her mistake, and her comments swerve a bit off course to address the manner of notification. Neither Juneau nor any of the others confront Uria-Ruiz at that point to demand that she correct her earlier misstatement. The point is, however, that Uria-Ruiz’s misstatement applied to all distributions—not just to those of the Union. Accord- ingly, paragraph 20’s allegation of disparity is without merit, and I shall dismiss that first part of complaint paragraph 20. The second part of complaint paragraph 20—that employees were prohibited from distributing union literature at the new interviewers’ desks—is, again, a mischaracterization of the facts. Neither Supervisor Uria-Ruiz nor her May 27 memo (JX 2) singled out the distributions of the union supporters. And as I find elsewhere respecting complaint paragraphs 9 and 15, there is no evidence that, before the mid-May advent of the Union, Gallup permitted distributions to the desks of trainees. As I dismiss similar allegations elsewhere, I shall dismiss the second part of complaint paragraph 20. Having dismissed both parts of the paragraph, I now shall dismiss complaint paragraph 20 in its entirety. 3. Supervisor Heidi Roberts a. May 18 and June 1, 1999 (1) Introduction Complaint paragraph 9 alleges that on May 18 and June 1, 1999, Gallup, by Supervisor Heidi Roberts, “instructed em- ployees not to place union literature on new interviewers’ desks and not to talk to those employees about the Union.” By its answer, Gallup “partially admits” the allegation “but notes that Supervisor Heidi Roberts instructed employees not to talk to new interviewers about the union during ‘buddy sessions’ which are working time orientations sessions.” I summarize these two allegations (in one count) separately, one pertaining to distributions, and the other to the gag rule. The distributions incident pertains to the May 18 date, and the gag rule pertains to both dates. (2) Distribution (a) Facts Since January 1998, Uria-Ruiz testified (2:338), Gallup’s Houston office has designated one area of the office as the new interviewer section (NIS), an area where new interviewers sit during their 2-week training period. (1:52–53; 2:337–338, Uria-Ruiz) Actually, as Lead Supervisor Heidi Roberts ex- plains (15:2944–2945), the first 2 days for the trainees are in a classroom, followed by 3 days of intensive training doing sur- veys, with the aid of “buddies” (experienced interviewers serv- GALLUP, INC. 1225 ing as mentors, 2:337; 3:546; 5:895), with the second week remaining there to work. Supervisor Heidi Roberts (who also is a trainer, 15:2932) explains that the NIS consists of 18 cubes situated, on the floor diagram in evidence (RX 38), at the upper center just to the right of the breakroom. (15:2935, 2938) Three of those cubes are used by two buddies and one blind interviewer. (15:2968– 2969). The area is designated by signs, hanging from the ceil- ing, bearing the legends, “New Interviewers” and “Welcome To Gallup,” and laminated notes, providing helpful hints about work procedures, are posted in the booths in the section. The rest of the office does not have “Welcome” signs, and the cubes elsewhere have no laminated notes. (1:53; 15:2938, 2963) Notwithstanding this designation of the NIS, Supervisor Uria-Ruiz acknowledges that at any given time there probably would be some experienced interviewers seated in the NIS in order that they could have access to one of the autodialers pre- sent in the NIS but not present at their own cubes. (1:53–54; 2:337) Roberts confirms (15:2969), as does Patrick Snyder (5:862–864). Janice Rinehart testified that at least half the employees seated in the NIS one night that she was there were experienced interviewers. (8:1700–1701) In fact, as dischargee Patrick Snyder informs us, there were times when new employees would outnumber the cubes in the NIS, and they would search for available cubes outside the designated area. (5:852–853) As for the “designated” area, there were no signs on the cubes (on the “desks”) in the desig- nated area declaring that such cubes were for new interviewers. (But Snyder, who testified at the rebuttal stage, did not there dispute the testimony about the signs handing from the ceiling nor did he contest the evidence about the laminated notes.) For that matter, new employees did not wear anything, such as a button, announcing that the employee was a “New Person” (“New Interviewer”). Thus, it was typical that when employees made distributions, copies simply were dropped “on each desk as you go.” (5:851–853) Around May 18, Janice Rinehart testified, Rinehart placed a stack of union flyers, numbering about 200, on a shelf in the dialer cubicle that she was using in the NIS. She also push- pinned one of the flyers to the inside wall of the cube, and she observed that Beverly Robinson, who was sitting behind her that day in the NIS, also had push-pinned one of the flyers to the inside wall of her booth. (8:1695–1698) Robinson con- firms, although she places the date as about May 20, and re- ports that the flyer was notice of a union meeting. (3:564–565) As shown by a copy of her airline ticket and hotel receipt (RX 60), and as she confirms (15:2951–2952), Roberts left early the morning of Thursday, May 20 for Stamford, Connecticut (fly- ing into New York’s LaGuardia), and returned late Monday, May 24. From this, I find that Robinson’s “about May 20” is consistent with, and occurred on, the approximate date of May 18 as given by Janice Rinehart. Although, on cross examination, Robinson could not confirm receiving or reading Uria-Ruiz’s May 27 memo (JX 2) regard- ing postings and distributions, she acknowledges that since the May 27 date she has never been restricted from posting any- thing in the breakroom. (3:601) (Respecting complaint para- graph 9, the only relevance of the latter item is that it tends to confirm that the incident of her posting the notice of a union meeting in the cubicle actually took place before the issuance of Uria-Ruiz’s memo of May 27, JX 2.) On this May 18 occasion, and apparently while Rinehart was on a break (8:1698), Robinson testified that she observed Su- pervisor Roberts come by and remove the flyers that Robinson and Rinehart had pinned to the inside of their cubes, and saw Roberts take the stack of union notices that were on the shelf in Rinehart’s cube. (3:565) As Roberts removed these items, she told Robinson (3:565–566): You are not allowed to post union literature in the new inter- viewer section. Moments later Robinson reported the matter to Rinehart. With “Money” Anderson accompanying her, Rinehart and Anderson confronted Roberts, informing Roberts that Rinehart had paid for the copies (presumably the cost charged by Kinko’s as Rinehart had done with earlier flyers, 8:1691–1692) and that Rinehart wanted them back. Roberts said that she had thrown them in the trash, but told Rinehart that she would re- trieve the copies for her. About 20 minutes later Roberts came to Rinehart, laid the copies on her desk,9 and told Rinehart, “These cannot be in the new interviewers’ section.” (8:1698– 1699) Supervisor Roberts freely acknowledges that this incident occurred, although she does not recall the date. Roberts testi- fied that a legal-size union flyer, “Organize, The Truth About Strikes!” (GCX 28), could have been the flyer involved in the incident. (15:2933–2936) Eileen Juneau identified the “Strikes!” flyer (GCX 28) (it also gave notice of a union meet- ing on May 23), which bears the handwritten notation “Friday 5–21–99,” as a flyer that she and others put on all the desks (3:607–608, 657) and that she also posted on the bulletin board by Mission Control (3:609–611). Although the “Strikes!” flyer may not have been the one in- volved in the removal by Supervisor Roberts of the 200 or so copies from Rinehart’s “desk” in the NIS, I need not find that it is the specific one. (Neither Rinehart nor Robinson identified any exhibit as the specific flyer, although, as mentioned, Rob- inson reports that the flyer announced the date of a union meet- ing, and the union meeting date specified on General Counsel 28 is May 23, a date just a few days after the May 18 incident.) At trial the General Counsel announced (3:611) that GCX 28 was offered in support of complaint paragraph 11, an allegation about a May 21 incident that I reach later. (b) Discussion If, as Supervisor Uria-Ruiz testified (2:327, 360), before May 27 Gallup actually had followed an unwritten policy pro- hibiting distribution of nonwork material in the office work areas, it would be home free. The evidence I earlier summa- rized, however, is the other way as to the office generally. So what about the NIS specifically? As to the NIS itself, the evi- dence is a bit fuzzy. True, Uria-Ruiz reports her (hearsay) un- derstanding that Michael Lee, when distributing the prayer- 9 The “desk” inside the cubes is a horizontal board about 14 inches across and about 40 inches side to side. (8:1696) DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1226 meeting flyer (RX 17) on May 13, retrieved it from the NIS when told he could not distribute it there (2:338), but there is no objective evidence, such as documentation of warnings given (2:338), or a copy of a memo that issued, bearing a date that precedes the mid-May advent of the union organizing. For its strongest evidence of no restriction respecting the NIS, the Government apparently relies on the distribution by Janice Rinehart of the water pillow flyers (GCXs 52, 53), and the “mixed use” (that is, by both experienced employees as well as the trainees) of the NIS. Rinehart testified that in September 1997 she placed copies of the color flyers on all the desks after receiving permission from Supervisor Barlow to make the dis- tribution. (8:1701–1703; 10:1935) Presumably “on every desk” (8:1703) includes the cubes in the NIS. On one hand, it is not at all clear that Barlow’s approval to distribute meant that the distribution could extend to the cubes in the NIS. On the other, however, in the absence of evidence as to some practice or understanding of an articulated policy, permission to distrib- ute, with no limitation stated, reasonably is interpreted as per- mission to distribute everywhere. However, recall Uria-Ruiz’s testimony that the NIS has existed since January 1998. (2:338) According to Janice Rinehart, she distributed her color flyers on the water pillows some 4 months earlier. (8:1701) Thus, Rinehart’s color flyers do not constitute a test of the integrity of the NIS. The “mixed use” concept perhaps could be a valid point ex- cept for one thing. Recall that Uria-Ruiz’s May 27 memo about distributions and postings (JX 2) provides an exception as to the new interviewers desks. Not “area,” but “desks.” Now Uria-Ruiz, in her testimony (1:50–52; 2:337–338), seems to use the terms “area” and “desks” interchangeably, and certainly Supervisor Roberts made no distinction on May 18 between the NIS and the desks located in the NIS. But the written policy does make a distinction, and that distinction could make a dif- ference. Thus, if, as the May 27 written policy states, the (pre- existing) exception applied to “desks,” then an experienced employee seated in the NIS reasonably might not be considered as seated at a desk that is off limits to any distribution. But if the exception applied to the NIS as a whole, then no distribu- tions could ever be made inside the perimeter of the NIS, even if on a given day everyone seated at the cubes in the NIS were experienced interviewers. The short answer would seem to be that Gallup framed and issued its memo of May 27, and any ambiguity in the language chosen should be construed against it. The fact is, however, that there is nothing ambiguous about the term “desks,” nor is any practical confusion created by reading the memo on the basis that the term “desks” was intended to be a distinction from the terms “area” or “section.” Once again, there is a prac- tical reason for interpreting this as an intended distinction. Recall the testimony that at times some of the trainees had to find cubes outside the NIS. Would that make their temporary location, outside the NIS, as fair game for any distributions? It would not have to, and surely, possibly with some clarification issued or markings on the desks themselves, Gallup could have notified the employees that a trainee’s desk was off limits pe- riod, whether located inside out outside the NIS. Similarly, if an experienced interviewer had to work in the NIS in order to have an autodialer on a given day, then that use would remove that desk from the May 27 rule’s exception, and the experi- enced employee could take with her a photo of her family and push-pin that photo (or a union flyer) to the inside wall of the cube without violating the May 27 memo. Turning back to the incident, recall that, contrary to the stan- dard practice, when posted items are removed, of placing the removed items in the desk drawer, Supervisor Roberts threw Rinehart’s private property into the trash. Clearly, I find, Rob- erts would not have thrown family photos or flyers announcing a church bazaar, the sale of a car, boat, or furniture, into the trash. Roberts, I find, would have preserved such items. But as the flyers pertained to the Union, then, in the view of Supervi- sor Roberts and, I find, of Gallup, they were garbage and be- longed in the trash. Does this literal trashing (even though on later demand, retrieved and returned) add weight to the Gov- ernment’s case? Certainly, I find, it shows animus. Note this. The allegation of paragraph 9, first part, is one of distribution (“not to place union literature on new interviewers’ desks”), whereas the evidence describes action against posting and the mere possession of union flyers in the NIS. This is something of a variance (unobjected to) from the allegation, but I find it an immaterial variance under Board law. Thus, the matter was fully litigated, and the allegation pled and any viola- tion found on this are closely linked. Parts Depot, Inc., 332 NLRB 670, 674 fn. 22 (2000). In short, I find that experienced interviewers did not lose their right—even if the May 27 written policy as to the NIS was preexisting (as I have described, the evidence does not show whether it was or not)—to possess union literature even when seated at a cube in the NIS, or to post a Union flyer on the in- side wall of the cube where they were working in the NIS. Janice Rinehart and Beverly Robinson were not sharing cubes with trainees, nor were they working with trainees. They were at their own temporary cubes that happened to be in the NIS. The animus reflected in Supervisor Roberts’ trashing of Janice Rinehart’s (and Beverly Robinson’s) personal property by the disparity of treatment accorded that property (union flyers), as opposed to the traditional preservation of private property, fully reveals that the real reason Gallup removed and trashed the leaflets was not that such personal property was nonwork mate- rial located within the NIS, but that it consisted of flyers an- nouncing the date and place of an upcoming Union meeting (and perhaps describing the Union’s position on one or more other topics). Further supporting this finding, which I make, is the fact that Supervisor Roberts simply ignored the in-your-face-presence of antiunion T-shirts and buttons being worn by some of the bud- dies (those who were opposed to the Union) right in the pres- ence of the new interviewer trainees during the training ses- sions themselves—when the union supporting buddies were not wearing any union insignia. Even though such T-shirts and buttons did not violate the letter of the (future) May 27 memo, ignoring and permitting their presence right in front of the trainees certainly violated the claimed basis for Gallup’s “long- standing” policy exception about distributions (or talking) as it supposedly pertained to need to protect the trainees from all intrusions of nonwork matters. GALLUP, INC. 1227 In short, as Janice Rinehart and Beverly Robinson were free to post, and possess, union flyers in their regular cubes (be- cause Gallup policy and practice permitted employees to have and to post virtually anything in the privacy of their own cubes), their posting of and possession of such in their tempo- rary cubes in the NIS was likewise protected conduct. This is particularly true here with the showing of animus displayed in the trashing of the union flyers, and when Gallup permitted antiunion “buddies” to wear loud antiunion T-shirts and buttons right in the face of the new interviewer trainees when the Union supporters were not wearing any union insignia. Such trashing and such permissive conduct clearly demonstrated that Gallup’s “longstanding” policy in fact was newly created in order to suppress the protected activities of the supporters of the Union. Accordingly, I find that by such May 18 removal of the Union flyers by Supervisor Roberts, Gallup violated Section 8(a)(1) of the Act by denying employees the right to post and to possess such union flyers in circumstances where, under standard Gallup practice, nonwork materials could be posted and pos- sessed, all as alleged by complaint paragraph 9, as modified at trial by the full litigation of the trial modification. (3) The gag rule (a) Facts Lynne Zieler, one of the four dischargees in this case, worked as a telephone interviewer, and also served as a “buddy”—a mentor for the new interviewers. (5:884, 895) Supervisor Heidi Roberts was, and is, a trainer for the new interviewers. (15:2932, 2944) As the trainer for her classes, Supervisor Roberts also instructs the buddies in their duties and responsibilities. Usually before every training session with the new interviewers, Roberts held a briefing with the buddies. (15:2940–2941) Zieler attended a buddy briefing held about May 18 at which Supervisor Roberts was the trainer. (5:896) There were bud- dies present who supported the Union, and also buddies who opposed the Union. (5:953) Zieler reports that the prounion buddies were not wearing, and never wore, union buttons, in- signia, emblems, or T-shirts. (5:899, 952–953, 955)10 In con- trast, the buddies opposed to the Union were wearing T-shirts showing big circles around the word “union” and a red slash mark through the circle. These antiunion buddies also were wearing (antiunion) buttons. (5:896, 897) Zieler asserts that Roberts told the prounion buddies that she was counting on them not to talk to the new interviewers about anything to do with the Union, that she did not want to scare them (the new interviewers), nor were they to place any literature on the desks of the new interviewers. (5:896) Specifically addressing the prounion buddies, Roberts said (5:954): I’m trusting you guys to not talk about the Union during buddy sessions and talk about the Union, in fact, to the new interviewers or place anything on their new interviewer desks because we do not want to scare them away. 10 Zieler understands that, after she and Janice Rinehart were fired, the remaining members of the union organizing committee wore but- tons with the photos of Zieler and Rinehart. (5:953) Testimony of other witnesses confirms this understanding. Even so, Zieler admits that Roberts said that she did not want anyone talking about the Union during the (training) time, but simply wanted the buddies to talk with the new interviewers about how to do their jobs. (5:954) Yet, Roberts did not tell those wearing antiunion slogans, buttons, or T-shirts that they were not to wear any such items when they began their training sessions with the new interviewers. (5:897, 954) About June 1, dischargee Sherri Lee reports, Supervisor Roberts told Lee essentially the same in a one-on-one conversa- tion that Roberts initiated in the reception area before a training class for the new interviewers. Roberts explained that she wanted to avoid confusing or frightening the new interviewers by such union talk. Lee told Roberts that she had no intention of talking to the new interviewers about the Union during train- ing sessions, and that she would place no flyers on their desks if the antiunion interviewers likewise refrained from such distri- bution. (6:1028–1030; 7:1329–1332) For her part, Supervisor Roberts confirms that she did cover this topic with the buddies, except she told all the buddies— about 10 or 12 present—and she dates the occasion as the brief- ing session before the June 3 class. She does not remember whether anyone was wearing buttons, emblems, or insignia favoring or opposing the Union. Roberts also does not think that she saw the Union’s May 16 letter (RX 5) bearing the sig- natures of the 20 members of the Union’s in-plant organizing committee, and thinks that she did not see that letter until a month or two before she testified. Even then she did not read the entire letter. She believes that her June 3 statements to the buddies was the only occasion that she addressed that subject with the buddies. And on that occasion she explained that the buddies were free to talk about the Union with the new inter- viewers during breaks, but not during the training sessions themselves, for that is work time for both the trainees and the buddies, so they were not to discuss the Union during that working time. (15:2939–2944) (b) Complaint paragraph 18 Respecting the gag rule, a separate allegation appears to du- plicate this issue. Thus, complaint paragraph 18 alleges that, on an unknown date in about May 1999, immediately prior to a training session, Gallup, by Supervisor Heidi Roberts, told buddies “that it was impermissible to mention union to new interviewers during training sessions.” (Roberts testified that her earlier testimony applied to this allegation. 15:2953–2954) By its answer, Gallup denies and affirmatively pleads that it followed its preexisting policy that prohibited, during new in- terviewer training sessions, discussion of any nonwork topics [emphasis added], and made no exception for talk regarding the Union. On brief (Br. at 45), Gallup acknowledges that Supervi- sor Roberts admits telling buddies that they were not to talk with new interviewers about the Union (during training ses- sions), but that she disputes other claims. At trial the General Counsel explained that testimony by in- terviewer Glen Lambert was offered in support of complaint paragraph 18. (3:517) Lambert reports essentially the same rule of no talking, and no distributing of literature, about the Union that the other testimony reflects (3:517, 521, 545, 547– 548), with two differences—Lambert’s description does not DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1228 show Roberts as singling out the union supporters, and he re- calls (3:517, 548) that Roberts told the buddies they were not to wear any buttons. Lambert recalls this latter point because Sharri Patty was wearing a red and white antiunion button of some 2.5 inches (diameter) and she never removed it the entire evening and nothing was said to her about removing it. (3:518– 519, 548) Lambert did not ask any supervisor why Patty was permitted to wear the antiunion button because, he explains, he is not very confrontational. (3:548–549) Although I generally credit Lambert, I find that he was mis- taken about a reference to buttons, perhaps mentally including that with the no talk and no literature. I note that he seemed less than positive about it on cross-examination (3:548)—“I would say yes;” “I will say yes.” And his tone was not even as strong as those words. Moreover, Lynne Zieler testified that Supervisor Roberts did not address the matter of what the bud- dies wore. (5:899) And in the individual conversation that Sherri Lee had with Roberts, Lee does not describe Roberts as mentioning anything about the wearing of buttons or other in- signia. Of course, Lee’s conversation with Roberts was private, but the absence of any buttons remark there, and the absence of any reference to buttons, insignia, or T-shirts in Zieler’s version, suggest a pattern that is more consistent with a finding that buttons simply were not mentioned. I find that they were not mentioned, but I do not credit Roberts that she does not re- member seeing any buttons or insignia. I also find that Super- visor Roberts addressed her comments to all the buddies. As for the date or dates of Roberts’ comments at the buddy briefing or briefings, I credit the testimony as given. Thus, and contrary to the version of Roberts that she gave this instruction only once, I find that she gave it at least twice at group brief- ings (once around May 18, then again, as Roberts concedes, on June 3), plus in the individual conversation with Sherri Lee about June 1. It makes sense that Roberts would have voiced it at an early stage—during May—and then later, after Supervisor Uria-Ruiz issued the memo (JX 2) about distributions (nothing to be placed on the new interviewer desks) and postings. (c) Discussion Based on these facts, as I have found, what is the legal con- clusion? Gallup argues (Br. at 152) no violation on the basis of its (formerly oral, now written as JX 2) policy prohibiting dis- tributions to new interviewer desks, with the reasons for the policy being “that the new interviewers need to concentrate on their work, learn without interruption and review their training material (15:2944, Roberts; 2:337–339, Uria-Ruiz).” As Gallup cites, Roberts and Uria-Ruiz did so testify. That reason is an impressive and persuasive basis for the policy—if the testimony is credited. However, in the cited testimony of Uria-Ruiz and Roberts, Roberts closes her description of for- bidden things—things that Gallup wanted to keep away from the trainees so as not to confuse them or disturb their training— with “or looking at pictures.” (15:2944) If Roberts and Gallup were so interested in keeping the new interviewers totally focused on their training during working time that they did not want the trainees to see any (nonwork, presumably) “pictures,” then why did Supervisor Roberts and Gallup stand mute when the buddies opposing the Union wore their antiunion T-shirts and buttons (presumably the same large, 2.5 inch, button that Sharri Patty wore) during the training classes? As Lynne Zieler credibly testified, all that antiunion insignia—permitted by Gallup—simply prompted the trainees to ask, “What is going on here?” (5:954) In short, I do not credit either Uria-Ruiz or Roberts as to this work-only focus of the training sessions. Thus, there is no evidence (aside from the discredited testimony of Uria-Ruiz and Roberts) even suggesting that the trainees were forbidden from talking about anything other than work during the work- ing time of their training sessions (and notwithstanding the portion of Gallup’s answer, set forth above in bold, claiming that such in fact preexisted). On the other hand, neither is there any evidence that the new interviewers were allowed to talk about nonwork topics during the working time of their training sessions. Although there is evidence that some trainees would wind up seated outside the NIS during all or part of a particular evening, or even be found wandering around looking for a cube to claim, even that evidence does not reflect that those individ- ual trainees were free to talk about nonwork subjects during the actual training periods. In short, what I find here is a failure of the Government’s evidence to establish that, before the advent of the Union, the new interviewer trainees, as distinguished from the regular interviewers, were permitted to talk, when time permitted, about the weather, sports, the theater, church bazaars, sales of cars, dogs, furniture, or Girl Scout cookies, or any other non- work topic of conversation. The Government must show dis- parity. Thus, in the cases of Opryland Hotel, 323 NLRB 723, 728–729, 731 (1997); Industrial Wire Products, 317 NLRB 190, 190 (1995); and Teksid Aluminum Foundry, 311 NLRB 711, 713–714 (1993), the evidence reflects that employees could talk about other nonwork topics, but not about the union. Gag rules (no union talk allowed) in those circumstances vio- late Section 8(a)(1) of the Act. Nor does it seem reasonable to infer that, even though regu- lar employees at Gallup are free (if they are not on a call with a consumer) to talk about nonwork topics during their working time, the trainees (who for a few days must focus intensely on their training) also share in that freedom to talk during working time. Positive evidence is required. No disparity having been shown respecting the new interviewers, I shall dismiss com- plaint paragraphs 9 (as to the no talking aspect) and 18. b. May 19, 1999 Complaint paragraph 10 alleges that, about May 19, Gallup, by Supervisor Heidi Roberts “removed union flyers from the walls of employees’ cubicles and from employees’ desks, in- structing employees that they were not allowed to post union literature in the center.” [Emphasis added.] By its answer, Gallup denies. Although her testimony is not fully positive, Supervisor Roberts also essentially denies. (“I don’t remember that. No.” 15:2950) The General Counsel (Br. at 11) names Beverly Robinson and Janice Rinehart as the supporting wit- nesses for this allegation, and gives a short description of their testimony in words tracking the language of the allegation (no transcript volumes or page numbers cited). GALLUP, INC. 1229 The apparent reason that the General Counsel omitted any ci- tation to the record, where the assertedly supporting testimony might be found, is that the only testimony that Rinehart and Robinson gave about this applied to complaint paragraph 9 and the NIS (not to the “center”)11—a matter I already have ad- dressed. When the Government’s lawyer, after studying the record, finds no record support for a complaint allegation, the proper course of action is to move to withdraw the allegation based on lack of evidence. The better and more courteous prac- tice is not even to wait until the brief to do this, but to do so by a separate document in advance of the brief so that opposing counsel does not waste time briefing the matter. But to treat it in the brief as if (by an attempted verbal sleight of hand, actu- ally) there is supporting testimony (but carefully not citing to the pages of the transcript), when there is none, is highly im- proper. Better is expected of a lawyer, and particularly of a prosecutor for the Government.12 In the absence of any supporting evidence for the allegation, I shall dismiss complaint paragraph 10. c. Confiscated and discarded union literature “In May 1999,” complaint paragraph 16 alleges, Gallup, by Supervisor Heidi Roberts, “confiscated and discarded union literature from a new interviewer’s desk and told employees that union literature was impermissible in that area.” By its answer, Gallup “partially admits” the allegations “but denies the allegations concerning confiscation and discarding the un- ion literature in question.” Rather, Gallup alleges, “such litera- ture was provided by Supervisor Heidi Roberts to union sup- porter Janice Rinehart at her request.” This allegation is almost a duplication of the first portion of complaint paragraph 9, as to which I have found merit. As the supporting evidence is the same, I find merit to this allegation which specifically alleges the confiscation and discarding of the union literature. Even though the material later, on request, was retrieved from the trash and returned to Rinehart, such action would tend to inhibit employees in the exercise of their protected rights and therefore constitute a violation of Section 8(a)(1) of the Act. See Albertson’s v. NLRB, 161 F.3d 1231, 1238 (10th Cir. 1998). I therefore find merit to complaint para- graph 16 (and to the conclusory paragraph associated with it, paragraph 27). d. Told trainees to report union talk (1) Facts “On an unknown date about May 1999,” complaint para- graph 19 alleges, Gallup, by Supervisor Heidi Roberts, “in- structed new interviewers to report to their supervisors immedi- ately if anyone spoke to them concerning the Union while they were working.” Admitting this allegation, Gallup argues that such action does not constitute a violation of the Act. Supervisor Heidi Roberts testified that this event occurred on either June 1 or 2. She told that evening’s class of some 12 to 11 As alert opposing counsel cover in Gallup’s reply brief at 36–40. 12 In an unfair labor practice trial, the General Counsel is cast in the role of a prosecutor. Zurn/N.E.P.C.O., 329 NLRB 484, 485 fn. 2 (1999). 20 trainees that there was an organizing drive in progress, and that while they were free to talk about it on nonworking times, if anyone talked to them for or against the Union during their working time then please tell a supervisor. Roberts did not limit the application of her instruction to talk favoring the Un- ion, and she recalls no complaints being registered thereafter from any trainee that someone spoke to them (about the Union). (15:2954–2955) Note that Roberts’ request that they tell a supervisor about someone speaking to them about the Union during “working time” is potentially broader than the allega- tion. The latter specifies that “they” be “working,” whereas Roberts’ actual request specifies “working time”—a concept that includes brief interludes when an employee may be waiting for her computer to complete an operation or even resting briefly before dialing the next survey call. However, the dis- tinction is so slight that I shall treat it as a distinction without a difference. Roberts also testified that only work materials are allowed in the NIS. And (15:2944): They [the new trainees] go through a pretty intense 3-day training, and they need to concentrate just on doing the sur- veys. This is their first time on the phones, so I’d like them to concentrate on just the quality of their surveys and not worry about looking at different things, reading different things on the walls or looking at pictures. (2) Discussion Unfortunately, no witness addressed whether the trainees, especially during their first 3 days (when this incident occurred, per Roberts), could talk about anything besides work during their actual working time (that is, between survey calls). With such intensity of focus on their survey calls, it seems unlikely that they would have had much time, between calls, to do much personal talking. The problem is that there simply is no evi- dence covering the point. To the extent that there was time for two or more trainees to have a short personal conversation, there is no evidence that they were prohibited from talking about the weather, football, a national beauty contest, a popular television show, or a national political race. If, during short interludes between survey calls, the trainees were prohibited from engaging in personal conversation about any nonwork topic, then, of course, the union topic, not being a privileged subject, likewise would be banned. But if other nonwork matters could be discussed, even briefly, then it would be unlawful to ban only the Union as a topic of personal con- versation during working time. Frazier Industrial Co., 328 NLRB 717, 717–719 (1999), enfd. 213 F.3d 750, 755, 759, 761 (D.C. Cir. 2000). Thus, “An employer may lawfully forbid employees to talk about a union during periods when they are supposed to be working, if that prohibition also extends to all other subjects not associated or connected with their work tasks.” Teksid Aluminum Foundry, 311 NLRB 711, 713 (1993). And see the earlier discussion about gag rules and the cases cited, including Opryland Hotel, 323 NLRB 723, 728– 729, 731 (1997). Stoody Co., 320 NLRB 18 (1995), cited by Gallup (Br. at 156), involved a no solicitation and no distribu- tion rule. Moreover, there the judge pointed out that the con- DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1230 duct did not involve a “prohibition of conversation about the Union.” 320 NLRB at 29. Taking the record as it is, and assuming that the trainees in fact had always been banned from any nonwork conversations during working time, would that status defeat the Government’s allegation? Stated differently, with such a universal ban in place, would it be unlawful for a supervisor to tell the trainees to report to a supervisor any attempt by someone (that is, by another trainee or a buddy) to talk to them, or any one of them, about the Union? First, notice that engaging in union talk does not have to be solicitation to join the Union or to sign a union card or even to promote the Union. It simply could be a basic question asked by one trainee of another, “What is a union?” [In Texas, that question would not be so unusual.] That ques- tion would not violate any separate no solicitation rule. Val- mont Industries, 328 NLRB 309, 317–318 (1999) (warning of Fontenot), enfd. on point 244 F.3d 454, 471–472 (5th Cir. 2001). Second, in both its primary and reply briefs, Gallup has too narrow a focus by arguing that Supervisor Roberts acted with complete equality as to the one topic of Union (by refer- ring to talk “for or against” the Union). The proper focus is wider—whether there was equality as to all topics, and whether singling out the Union as a topic constituted unlawful disparity of treatment. Unfortunately, Gallup does not address this proper focus of the issue. Thus, the real question is whether Supervisor Heidi Roberts’ request that the trainees report any effort to talk about the Un- ion during working time was unlawful because the subject mat- ter to be reported was limited to the Union and did not include other examples to reinforce the concept of a universal ban, not simply a ban against talking about a single topic, the Union. By singling out the Union topic, did Gallup, by Supervisor Roberts, thereby violate Section 8(a)(1) of the Act? I conclude that the answer is yes. Such singling out, by stressing only the Union as a topic to be reported, violates Section 8(a)(1) of the Act. See Greenfield Die & Mfg. Corp., 327 NLRB 237, 238 (1998), and Meisner Electric, 316 NLRB 597, 597 (1995), enfd. mem. 83 F.3d 436, 437 (1996). Accordingly, I find merit to complaint paragraph 19. 4. Supervisors Heidi Roberts and Chad Murray—May 21, 1999 a. Introduction Complaint paragraph 11 alleges that about May 21 Gallup, by supervisors Heidi Roberts and Chad Murray, removed posted union flyers from the bulletin board by “Mission Con- trol.” Gallup denies. I treat the supervisors separately. b. Supervisor Heidi Roberts As mentioned earlier respecting the “Strikes!” flyer (GCX 28), interviewer Eileen Juneau testified that she helped distrib- ute copies of that flyer to all the desks on May 21, and that she also posted a copy on the bulletin board by mission control. (3:607–609, 657–659) Respecting the copy that she posted on the bulletin board, Juneau testified that she observed Supervisor Roberts remove it shortly after Juneau posted it. (3:610) Rob- erts denies, explaining (as described earlier) that she left early on May 20 for Connecticut and returned to Houston late on May 24 (15:2951–2952)—as copies of her airline ticket and hotel bill (RX 60) reflect. Arguing that Juneau should be credited, because Juneau per- sonally observed the removal by Roberts, the General Counsel further contends that Juneau’s testimony “offered only an ap- proximate date.” (Br. at 12) Not so. Unlike the “about May 21” date in the allegation, Juneau’s testimony at every point is specific—it was May 21. Not only that, but the handwritten notation on the flyer (GCX 28), apparently recording the date, reads “Friday, 5–21–99.” Juneau perhaps was mistaken regarding whom she observed, for Supervisor Uria-Ruiz asserts that, from May 20 to June 30, when she passed by the mission control bulletin board and saw a new flyer posted, usually in the wrong place, she removed it. But instead of discarding it, she moved it to the “Other” section where, as reflected in the May 27 memo (JX 2), employees may post items when advance permission has been obtained by a supervisor. Uria-Ruiz testified that the Union supporters rarely asked in advance, whereas the antiunion group always asked, and despite the union group’s failure to ask permission before posting, she nevertheless moved several union flyers over to the “Other” section. And, Uria-Ruiz testified, she told this to Jun- eau and Sherri Lee at a meeting on June 4. (1:49–54; 2:335– 336, 349–350) I need not determine what actually happened here. It is enough that the evidence relied on by the General Counsel does not support the allegation. To the extent that Juneau observed someone removing a flyer from the mission control bulletin board on May 21, that someone may have been Supervisor Uria-Ruiz in the process of moving the flyer to the “Other” section of the bulletin board, although I make no finding re- garding the matter. The Government’s attempt, on brief, to go beyond the date of May 21 simply invites speculation as to some other posting date, contrary to the evidence specifying the date of May 21. Accordingly, I shall dismiss complaint para- graph 11 as to Supervisor Heidi Roberts. c. Supervisor Chad Murray Eileen Juneau credibly testified that, on May 24, she posted a copy of an “It’s The Law” (GCX 29) handbill on the mission control bulletin board and placed copies on the desks of the interviewers. About 10 minutes after she posted the copy on the mission control bulletin board, Juneau saw Supervisor Chad Murray remove it, but she had no conversation with him. (3:611–613, 657–658) Supervisor Murray’s “I don’t ever recall doing that” (15:2976) was unpersuasive, and I do not credit him. There is no testimony that Juneau asked a supervisor for permission to post the document on the mission control bulletin board, and Murray did not testify that he removed the document because no permission had been requested or granted. Indeed, there is no evidence specifically rebutting Uria- Ruiz’s credible testimony (1:48; 2:347) that permission has always been required to post any material in the “Other” sec- tion of mission control’s bulletin board. No witness testified, for example, that he or she personally posted something on the mission control bulletin board without obtaining permission from a supervisor and that a supervisor knew about the posting and did nothing or that the posting was there so long that a GALLUP, INC. 1231 presumption arises that management would have known about it. Generic references to posting (by someone) on the mission control bulletin board (6:1038, Lee) do not rebut Uria-Ruiz’s testimony. And postings on the “walls by the bulletin board” (4:681, Snyder) are not postings on the mission control bulletin board. Gallup’s policy requiring permission to post anything on the “Other” section of the mission control bulletin board predated advent of the union organizing campaign, and there is no evi- dence of record showing that management ignored breaches of the policy until union supporters began posting items favoring the Union. On the other hand, there is no testimony that Super- visor Chad Murray removed the posting only because no per- mission had been sought or granted. Given the additional evi- dence shown later respecting complaint paragraph 14 (reflect- ing a purpose more consistent with censorship than with ensur- ing the integrity of management’s control of its property), it appears that Supervisor Chad Murray removed the document as a form of censorship rather than as an expression of manage- ment’s legitimate control of access to its property. As the con- tents of GCX 29 cover protected matters (there is no contention that GCX 29 is unprotected), and as Gallup has approved post- ings of sales and other employee matters, Supervisor Murray in fact discriminated against this posting because, I find, it would tend to support the Union’s side in the organizing campaign. Accordingly, I find merit to that portion of complaint paragraph 11 dealing with the conduct of Supervisor Chad Murray on May 24, which date, I find, is “about May 21.” (Gallup ob- jected to the receipt of GCX 29 on the basis of a variance from the date alleged. 3:612–613) 5. Supervisor Chad Murray a. Introduction Several additional allegations pertain to Supervisor Chad Murray. I summarize the facts of each allegation before dis- cussing the group. b. May 21, 1999 About May 21, complaint paragraph 12 alleges, Gallup, by Supervisor Chad Murray, “removed posted union flyers from the walls of employees’ cubicles.” By its answer, Gallup de- nies, and affirmatively pleads that it continued to apply its pre- existing policy concerning posting and distributions, except that after Mike Lee’s mid-May distribution of a religion-oriented material, Gallup [by JX 2, the May 27 memo from Uria-Ruiz] “relaxed its policy and only required prior notification for the distribution of non-work related literature in its facility.” Inter- viewer Marion Trovato testified in support of this allegation. About May 21, Trovato testified, Lynne Zieler posted some union literature on the outside of her cubicle, of Trovato’s cubi- cle, of Catherine Wagley’s cubicle, and of a vacant cube. Zieler confirms. (5:890) Shortly before 10:30 p.m. that eve- ning, Trovato observed Supervisor Chad Murray come by and remove the dozen or so union papers that Zieler had posted. When Trovato asked Murray what he was doing, he replied, “I’m taking this stuff down.” After Murray had removed it all, he left. (2:367–370, 385–387) Zieler learned of this on return- ing from a break. (5:891) Testifying that he had read complaint paragraph 12, Supervi- sor Murray asserts that he removed copies of the Union’s flyer about strikes (GCX 28) that were lying on vacant desks, and from desks where the interviewer was not present, but “Not that I recall” as to removing any materials from the exterior walls of interviewers’ “desks” [cubes] about May 21. Asked whether he recalled having any discussion with Marion Trovato about his (Murray’s) removing posted union materials, Murray answered, “No.” (15:2977–2978) Murray concedes that when he re- moved papers from a desk in the past that his purpose was al- ways to tidy up the place before a client’s visit or to clean off a vacant desk before a new employee arrived to be seated there. (15:2985–2987) Resolving the factual matters, I credit Marion Trovato and Lynne Zieler. Supervisor Murray’s testimony “Not that I re- call,” and, similarly, his answer to the question asking whether he recalls a conversation with Trovato about removing posted materials (“No,” he does not recall), are feeble efforts com- pared to Trovato’s specific and forceful testimony. Trovato was a persuasive witness as to this, as was Zieler, and I credit them. Supervisor Murray testified unpersuasively, and I do not credit him. Indeed, Murray admits that, before Supervisor Uria-Ruiz issued Gallup’s May 27 memo (JX 2), he believed that Gallup’s policy prohibited the distribution of union materi- als to the interviewers’ desks. (15:2979–2980) His testimony here bears specifically on complaint paragraph 13, and I now turn to that allegation in a moment. As to complaint paragraph 12, I find merit. c. May 21 or 28, 1999 Complaint paragraph 13 alleges that, “about May 21, 1999, or May 28, 1999, Respondent, by Chad Murray, directed em- ployees that they could not distribute union documents and that he did not want them engaging in said activities any longer.” Gallup denies. Sherri Lee testified in support of the allegation. About a week before Gallup’s May 27 memo (JX 2) on dis- tributions and postings, Sherri Lee testified (6:1031; 7:1333, 1334), she assisted the other members of the organizing com- mittee in distributing and posting copies of the “Strikes!” hand- bill (GCX 28). She put copies on interviewers’ desks and on walls. She did not ask any supervisor for permission to do this. (6:1032; 7:1333–1336) Following this action, Supervisor Chad Murray called Lee into his office. Lee took Tony “Money” Anderson (who did not testify) as a witness. In an agitated tone, Murray told Lee that she “could not post” and “could not put that document out.” Murray appeared to be more upset about the contents that he was about the fact of the distribution or posting. (6:1031; 7:1333–1334) The “Strikes!” flyer has, at the top of each corner of the paper, a statement reading, “Don’t Fall For The Boss’s Lies!” Several bulleted points on the paper describe “What the Boss Will Say,” plus “Some Questions for the Boss,” and “What The Boss Won’t Tell You.” Responding to Murray, Lee said that the Union supporters had not put out any flyers until the antiunion group did, and once they did then the Union supporters did likewise. But be- cause she did not want to be insubordinate, she would check with the Union representative and would do as he suggested. (6:1031) DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1232 Advising that he had read complaint paragraph 13, Supervi- sor Murray acknowledges that, before the distribution of Uria- Ruiz’s May 27 memo on postings and distributions (JX 2), on the same Saturday (May 22) that he gathered copies of GCX 28 (15:2997, 2979), he had a conversation with Sherri Lee and Tony Anderson regarding distribution of union materials at work. (15:2979) At the time of this conversation, Murray as- serts, it was his understanding that Gallup’s policy prohibited employees from distributing materials in a work area, such as placing them on the desks of the interviewers. He so told them this (testifying that it was Lee who came in to be there with Anderson), and that they were to confine their distributions to the break areas and nonworking areas. He assertedly told them that he wanted to do this the correct way, and that, “We all want to play by the rules here.” According to Murray, follow- ing the May 27 issuance of Joint Exhibit 2, he adhered to the policy described there by Supervisor Uria-Ruiz, and he had no further discussions with employees about postings or distribu- tions of materials. (15:2979–2981) To the limited extent of the differences, I credit Sherri Lee over Supervisor Murray. As to this matter, Lee testified more persuasively. In fact, Murray admits the essential of the allega- tion, offering his own misunderstanding of Gallup’s asserted policy as his defense, and observing that he did not violate Gallup’s written policy (JX 2) following its issuance on May 27. Respecting the differences, in accordance with Lee’s cred- ited version, I find that Murray spoke in an agitated tone (and not in the calm manner that he that he implies) and in the lan- guage of the account given by Lee. In short, as to the facts, the Government’s evidence is cred- ited as to “about May 21.” The General Counsel cites no evi- dence in support of the “about May 28” allegation, and I find none. Accordingly, I shall dismiss complaint paragraph 13 as to the date of May 28, 1999, and I find merit to the first portion of the allegation, that being the allegation of “about May 21.” d. Last week of May 1999 Complaint paragraph 14 alleges that in the last week of May Gallup, by Supervisor Chad Murray, “removed a union docu- ment from the main bulletin board and told an employee that such postings were impermissible.” Gallup denies. Sherri Lee testified in support. Lee testified that, about the end of May, she and other posted a union flyer, pertaining to the reasons for signing a union card,13 on mission control’s bulletin board. Supervisor Chad Murray removed it. They put up another copy, and Murray removed that one, too. Lee personally saw him do this twice, both within a day of the other occasion. However, Murray would not remove the antiunion posters. After the second re- moval, Murray told Lee and the others not to post it again. Lee told Murray that, although they did not want to be insubordi- nate, they did not think it was fair that he would remove the Union’s posters but not those of the antiunion supporters, and that therefore they would continue to post union material as long as the antiunion people did so. (6:1039–1041) Murray’s 13 No copy of the flyer was identified. Thus, it was not necessarily a copy of a similar flyer in evidence as GCX 31. “I don’t ever recall doing that” (15:2981) was no more persua- sive in tone than the first time. Thus, I find that he spoke his prohibition as described by Lee. As with my findings on the second part of complaint para- graph 11 (Murray component), I also find here that Supervisor Murray was engaging in censorship rather than in an effort to protect management’s legitimate control of its property. In- deed, here Murray left antiunion flyers posted. Even if the antiunion group had received permission to post its antiunion messages on the mission control bulletin board, that distinction was not conveyed by Supervisor Murray to Sherri Lee. Thus, employee Lee reasonably could have perceived that the purpose of the removal, and the leaving of the antiunion posters, was discrimination against the union materials simply because they were perceived by management as favorable to the Union. In short, Gallup acted with disparity toward Sherri Lee and the Union supporters. Accordingly, I find merit to complaint para- graph 14. e. A Saturday in May 1999 “On a Saturday in May,” complaint paragraph 17 alleges, Gallup, by Supervisor Chad Murray, “removed union literature from an interviewer’s desk and instructed employees not to distribute union literature at work.” By its answer, Gallup: . . . partially admits the allegations of Paragraph No. 17 of the Complaint but denies the allegation concerning instructions by Supervisor Chad Murray to employees not to distribute un- ion literature at work. Rather, the Respondent alleges that Supervisor Murray instructed such employees that they could not distribute such literature in works areas but must limit the distribution to [the] kitchen area and the break area. Apparently because of the admissions in the answer, the General Counsel (Brief at 16) cites no supporting testimony, and argues that the case law supports a violation. Supervisor Murray testified that in May he did remove union literature— GCX 28 (the “Strikes!” handbill)—from an interviewer’s desk, and that there was only one such date, that being as he had earlier testified (respecting complaint paragraphs 12 and 13). As this allegation duplicates the posting and distribution of the “Strikes!” flyer (GCX 28) on Saturday, May 22, already covered by complaint paragraphs 12 and 13, I shall dismiss this complaint paragraph 17. 6. Gallup adds “union issue” to its non-harassment policy a. Facts Complaint paragraph 21 alleges, and Gallup admits, that about May 20 Gallup issued the following “Non-Harassment Policy Statement” (JX 1): If you feel you are being harassed or threatened at work about the union issue, contact your supervisor immediately. Har- assment is not tolerated in the workplace. Gallup will follow our normal harassment policy. As one of its affirmative defenses, Gallup pleads that it has adhered to its preexisting nonharassment policy and “did not establish any new procedures for the handling of employee GALLUP, INC. 1233 complaints concerning harassment pertaining to the activities of either pro-union or anti-union employees.” The record reflects that, shortly after the Union opened its organizing drive, Supervisor Uria-Ruiz began receiving com- plaints from some employees that they were being interrupted in their work by union supporters seeking to persuade them to join their ranks. As I noted earlier, as disclosed by Sherri Lee, the organizing committee was surprised to find that many em- ployees were not interested in having union representation. Copies of the written complaints that Uria-Ruiz received are in evidence. (RXs 7, 8, 9, 10, 11) Uria-Ruiz also received a complaint (JX 15) from Janice Rinehart (the first signee on the in-plant organizing letter, RX 5) that someone had shredded the family photo that she had kept in her cubicle, although this complaint came a few days after the posting in issue. Gallup already had a nonharassment policy that prohibited every harassment known to mankind, except harassment over union matters. So, after consulting with higher management, Uria-Ruiz redistributed the existing policy with a new box at the top containing the language (JX 1) quoted above and at- tacked (as facially unlawful) by complaint paragraph 21. (2:292, Uria-Ruiz; 13:2473–2475, Geiger; 15:3056–3058, Uria- Ruiz) There is no evidence that the purpose of adding one more topic to the list was to chill activities for or against the Union. The purpose was to protect employees from harassment by either group. (15:3058, Uria-Ruiz) As Supervisor Uria-Ruiz informed employees by her memo (text in all capitals with one paragraph, quoted here, in bold) of May 26 (JX 4): WE WILL NOT ACCEPT HARASSMENT ON EITHER SIDE! THIS IS GROUNDS FOR TERMINATION! No evidence was presented that Gallup applied the modified no-harassment policy in a manner that would discriminate against union supporters, nor is there any such allegation in the complaint. Complaint paragraph 21 attacks the quoted modifi- cation as unlawful on its face. b. Discussion Notwithstanding the General Counsel’s suggestion (Br. at 27) that employees who feel that they are being harassed or threatened at work by either pro or no-union supporters are not part of a protected class, Section 7 of the Act protects both groups. The General Counsel’s violation theory (Br. at 27) appears to be that the box addition to Gallup’s policy “is noth- ing more than a blatant demand that employees engage in sur- veillance and reporting on the union activities of their fellow employees, and as such is plainly unlawful.” There is no such “blatant demand” here. Employees are not told, for example, to report anyone who asks them to sign a union card or who sim- ply tries to talk to them about the union matter. And contrary to the Government’s asserted theory, Gallup’s modified policy is neutral, thereby protecting all employees. The purpose of the modified policy is preventive, not punitive. An employer need not, and should not, wait until fist fights erupt or stabbings occur in the workplace to act. The policy modification here is calculated to avoid problems. Employees at work have the right to say “No” to coworkers who repeatedly seek to talk with them about any topic, including sports, weather, politics, stock market, religion, sex—or unions. “What part of ‘No!’ don’t you understand?” applies just as much to any one of these topics as it does to another. Workers should be free to ask their employers for assistance against unwelcome verbal aggression well before that aggression esca- lates to the point of verbal assaults, physical threats, or even emotional stress or mental fatigue. “No!” means No! And coworkers who ignore such a “No” and continue to interrupt and bug another worker with unwelcome talk need to know that there can be adverse consequences for such unwelcome verbal harassment. There being no merit to this allegation, I shall dismiss complaint paragraph 21. 7. Gallup locks its copy room a. Introduction Two allegations are involved here. About May 19, com- plaint paragraph 22(a) alleges, Gallup “locked the door to the copy room and prohibited employees from making copies.” Gallup “partially admits” but “denies the allegation concerning the prohibition of employee making any photocopies.” About May 24, complaint paragraph 22(b) alleges, Gallup “distributed a memorandum stating that employees could no longer make copies without approval.” Gallup admits. For an affirmative defense, Gallup pleads that at all material times it adhered to its preexisting policy “concerning the prohi- bition of utilizing the Respondent’s photocopy facilities for non-business related activities and made no exception concern- ing either pro-union or anti-union material.” The memo in question (JX 19), apparently the one also men- tioned in complaint paragraph 22(b), was distributed about May 19. (2:304, Uria-Ruiz; 11:2165; 12:2329–2330, Barlow) Enti- tled “Memo,” not addressed to anyone, and bearing the typed first names of the seven supervisors, the text of the memo clos- ing the copy room reads (JX 19): We make a ton of copies. We make more copies than any good American should. In addition, however, to all of our regular business related copies we have had a great deal of copies made for personal reasons. We have always held to the policy forbidding non-business related use of the copy machine and office supplies. [H]owever, due to consistent disregard for this necessary rule as well as the constant need for repairs and maintenance to the machine, not to mention ever increasing costs associated with the sheer volume of copies we make on a daily basis we sim- ply can not continue to allow free access to the supply room and copy machine. Thus, in an effort to limit the amount of trees and money we burn each month, as well as the amount of wear and tear on the copy machine itself, the Houston office will adopt the same policy as the other six locations in the company. Instead of making copies of your pay sheets each night, we have weekly pay logs that you can fill out with your pay sheet. Payroll will accept these in place of the copies without any trouble at all. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1234 If any of you need to make a business-related copy let any one of the managers know and we will make it for you without delay but again, in order to keep the machine in working order and to cut down the ever skyrocketing costs of wasted paper, ink and other supplies we are left with few options other than to limit the number of copies we make. We apologize for any inconvenience this may cause and thank you for your understanding. b. Facts The copy room at Gallup’s Houston office (one copier until a second arrived when the Bellaire office merged around mid- May) also houses the office supplies and the payroll printer. (11:2166; 12:2328, Barlow) Janice Rinehart reports that, be- fore Gallup locked the copy room, employees not only would make photos of their time sheets and production records, but also of “different things.” (8:1704) Asked whether there had been any restriction on access, Rinehart tells us that there was “not really any restriction” even though there was a sign on the wall above the copier instructing employees to obtain permis- sion to photocopy anything personal. (8:1705) Others confirm the presence of the sign before its mysterious disappearance in about May. (2:312–313, Uria-Ruiz; 2:373, Trovato; 11:2168, 12;2331, Barlow; 12:2415–2416, Collette; 13:2477, Geiger) According to Rinehart, this sign was not honored, and for proof she describes an occasion when she observed one of the executive interviewers14 photocopying an entire recipe book. However, Rinehart does not recall whether any supervisors or managers saw this. In fact, Rinehart cannot recall any instance when a supervisor was present and someone was copying non- work items. (8:1705–1706) Patrick Snyder, however, states that he has been in the copy room when different supervisors were present and personal copying was being done. On one occasion, for example, Su- pervisor Chad Murray was present when an employee made a copy of a newspaper article. In fact, everyone present was discussing the article. “It was a common occurrence. I mean, it was nothing that was made note of. I mean, people used the—I mean, you couldn’t go in and make 500 pages, but if you were just running one or two pages, nothing was ever said.” (4:714, Snyder) On another occasion, a month or two before the start of the union organizing in mid-May, Snyder asserts, Supervisor Stephanie Geiger was present when an employee copied a no- tice to sell a sofa and a loveseat. Geiger, who was standing next to Snyder waiting to use the copier, “didn’t give her per- mission, but she didn’t object to it.” (4:715–716) Geiger, whose first day at the Houston office did not come until May 17 (13:2466–2467), credibly denies ever knowingly allowing an employee to make personal photocopies. Geiger’s denial highlights a credibility problem with the leading nature of the General Counsel’s questions. Thus, when Snyder testifies that he has seen supervisors in the room when personal copies were 14 Executive interviewers are interviewers who work during the day hours telephoning business executives and professionals. (3:536–537, Lambert; JX 23 at 10) being made, the General Counsel asks him to name the supervi- sors whom he saw “observing” the copying. The witness was not asked to describe what the supervisor was doing. (4:714) Factual descriptions by a witness that a supervisor was look- ing at and reading the items being copied, or commenting about the nature of the item being copied, are what justify a conclu- sion that a supervisor observed what was being copied. Other- wise it is just as reasonable to infer that the supervisor, al- though present in the room, was not inspecting what was being copied, but was preoccupied with other thoughts and perhaps was not even looking at the photocopying, much less inspecting what was being copied. In any event, employees have been disciplined for making personal photocopies. Supervisor Douglas Barlow participated in the decision to orally warn Chad Kushner in 1994 (11:2168– 2170), and himself orally warned Shirley Lawler, also about 1994, that if she did it again she would be terminated (11:2170–2171) Those are the two instances that Barlow re- calls. (12:2332) Barlow advises that the copy room remains locked. (11:2173–2174) In the early spring the Houston office was experiencing an abuse of access to the copy room. Office supplies, including bulk items such as toner cartridges and reams of copy paper, were missing, confidential payroll information was found on the copy machine,15 an employee was observed copying an email that Supervisor Uria-Ruiz had sent to the printer in the copy room, paper was getting jammed in the copier, and fre- quently the jammed copies, as Patty Collette advises, were of nonwork items. (2:309–311; 15:3059–3060, Uria-Ruiz; 11:2166–2167; 12:2327, Barlow; 12:2414–2415, Collette) The Xerox technician, Ranzy J. Campbell Jr., suggested three op- tions (a coding device, lock the copies, or lock the copy room). (2:308–309; 15:3060–3061, Uria-Ruiz; 12:2373–2374, Camp- bell) Janice Rinehart suggests that the closing was prompted by the number of copies of the Union flyer that Supervisor Roberts confiscated from Rinehart’s cubicle the day before the closing memo issued. (8:1704) However, this does not follow because, as described earlier, Rinehart and “Money” Anderson confronted Roberts. On that occasion Rinehart, telling Roberts that she had paid for copying the item at Kinko’s, displayed her Kinko’s receipt for the cost of the copying. (8:1699) Respecting the only factual allegation denied, the averment that when Gallup locked the copy room door it also prohibited employees from making any copies, Marion Trovato asserts that, after the room was locked, she was never allowed to make any copies. (2:375) Such testimony is both ambiguous (work copies or copies of personal items?) and inadequate (not an assertion that she asked a supervisor and that her request was rejected). The memo (JX 19), from all the supervisors, an- nouncing the closing expressly states that business-related cop- ies will be made for employees “without delay” if employees ask a supervisor. I find that Trovato’s ambiguous testimony fails to prove that Gallup refused to make any business copies, as promised. Indeed, Supervisor Barlow credibly testified that 15 Thereby illustrating Connor’s Second Law, “If something is con- fidential, it will be left in the copier machine.” A. Bloch, Murphy’s Law, Book three at 60 (1982). GALLUP, INC. 1235 such copies were made, and that he never denied any such re- quest. (12:2231) However, as the same memo (JX 19) states, “. . . the Houston office will adopt the same policy as the other six locations in the company. Instead of making copies of your pay sheets each night, we have weekly pay logs that you can fill out with your pay sheet. Payroll will accept these in place of the copies with- out any trouble at all.” Supervisor Uria-Ruiz testified that the Houston office had been the only one which had allowed copies to be made of the nightly paysheets. (2:304–305; 15:3062– 3063) An example of a “paysheet” appears in the record as GCX 16. Finally, there is no evidence that Gallup used this occasion to discriminate against Union supporters by allowing the anti- union forces to make copies either of their antiunion flyers, of their other personal items, or of their nightly paysheets while keeping the door closed as to the Union supporters. c. Discussion The General Counsel argues (Br. at 22) that closing the copy room, and prohibiting the making of copies of the nightly pay- sheets, worked a hardship on employees by requiring them to hand copy detailed information on their nightly paysheets. Gallup’s purpose in making this change in the “terms and con- ditions of employment” was intended to inhibit union organiza- tional efforts and to penalize employees for exercising their Section 7 rights. In support of this argument, the General Counsel cites Boyer Ford Trucks, 254 NLRB 1389, 1394 (1981). Countering this contention, Gallup argues (Reply Br. at 4–5) that Boyer Ford involved a new rule prohibiting sales employ- ees from duplicating items needed for company business, whereas our case fits under the rule of Yale New Haven Hospi- tal, 309 NLRB 363, 369–370 (1992) (old rule against personal telephone calls, except for emergencies, not always enforced, and manager, returning from a 3-month assignment, renews enforcement to regain efficiency; no disparity in application; no unlawful intent found). The initial question is whether a protected right is involved here. For example, if—admittedly because the employees start organizing for a union—the boss stops smiling at and greeting the employees each morning, such elimination of her morning smile and daily greeting is not unlawful. The law does not require that the boss be congenial. It only requires, under Sec- tion 8(a)(3) of the Act, that he not discriminate respecting hir- ing or tenure of employment or as to “any term or condition of employment.” Coercion under Section 8(a)(1) of the Act, as the General Counsel acknowledges, must threaten a change in, or be a change in, are relevant here, “the terms or conditions of employment.” Thus, if the boss were to threaten that he would stop smiling at the employees in the mornings if they brought a union into the plant or office, there would be no violation, for cancellation of congeniality does not rise to the level of the elimination of a term or condition of employment. Thus, all the evidence about previous access to make some free copies photocopies of personal items (even assuming su- pervisory knowledge or approval of, as Patrick Snyder de- scribes it, a page or two) simply describes, at most, a gift from Gallup. Gifts of company picnics, Christmas turkeys, and such that are not linked to wages, seniority, or work performed, are not included under the statutory clause, “terms and conditions of employment.” See Stone Container Corp., 313 NLRB 336, 337 (1993). Even more so would be occasional free copies of a couple of pages or so from the employer’s copying machine. And even if free photocopies are a class that somehow could be included under the statutory clause in some circumstances, the occasional few copies involved here still would fall outside the coverage under the old “de minimis” legal maxim, “the law does not concern itself with trifles.” Does the addition of the work paper that had been copied make a difference? It did in Boyer Ford Trucks, 254 NLRB 1389, 1394 (1981), only because the rule prohibited the sales- men there from copying items “needed for company business.” As such, the judge described the rule as a “mindless act,” one apparently born of a desire to retaliate. The rule was found to violate Section 8(a)(1) of the Act by imposing a more onerous working condition. That situation does not prevail here because no “mindless” retaliation is involved. First, Gallup had several good business reasons for making its change. As for extra pen and paper work that possibly was a result (the Government’s case relies more on assumption than on evidence; indeed, this theory seems to be an afterthought), the switch to that procedure merely brought the Houston office into conformity with the procedure followed in Gallup’s other offices. Moreover, the change affected eve- ryone, and not simply the union supporters. In any event, I credit Gallup’s witnesses that the purpose was to protect and conserve its property, including confidential papers, and not to retaliate against employees because some were seeking to organize support for the Union. Accordingly, I shall dismiss complaint paragraph 22. 8. CEO Jim Clifton speaks a. Facts On May 26, complaint paragraph 23 alleges, Gallup’s CEO, Jim Clifton, “threatened employees with general reprisals and with loss of their jobs if they supported the Union.” On an- swer, Gallup denies. As its affirmative defense number V, Gallup avers, “All comments made by the Respondent’s Chief Executive Officer Jim Clifton on May 26, 1999 constituted non-coercive expressions of opinion which are protected by 29 U.S.C. § 158(c).” Early in the trial Gallup filed a motion (RX 1), and a sup- porting brief (RX 2), to dismiss complaint paragraph 23 on the basis that CEO Clifton’s remarks are protected by Section 8(c) of the Act. (1:16–18) The General Counsel filed (5:819-821) an opposition. (GCX 40) Initially I postponed ruling (1:25), and later Gallup agreed that my ruling could wait until this decision. (5:821) At this point, the matter simply merges into this decision on the merits. The facts are undisputed. Pursuant to a May 26 memo (GCX 37) from “The Management Staff” announcing a “mandatory” meeting beginning at 7 p.m. in the downstairs conference room to hear CEO Jim Clifton speak about “recent developments and implications regarding the integrity of Gallup,” all employees of the Houston office assembled at the first floor conference DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1236 room as directed. (All tenants of the building are able to use a large area on the first floor as a meeting room. 11:2156–2157, Barlow.) The meeting was videotaped, and a copy of the tape (JX 5) is in evidence. (1:8, 14) A 36-page transcript (JX 6) of the video- tape also is in evidence. (1:9, 14) The parties stipulated (12:2242), and I agree, that the videotape runs for 46 minutes. Because about the first 3 minutes, and part of the last minute, pertain to some preliminary matters (including a request that the employees submit questions written on paper provided, with the promise that every question would be answered later, but that no questions would be taken at the meeting), the actual talk by CEO Clifton, by my count, consumes 42 minutes, and cov- ers 32 pages (pp. 5 to 36) of the transcript. Aside from the notes Clifton had at hand for occasional reference (JX 6 at 5), and as is obvious from the video, Clifton does not read from a prepared text. The transcript is the text that we have of Clifton’s extemporaneous remarks. Because of the extempora- neous nature of his speech, and of Clifton’s conversational style of delivery, the words and thoughts of Clifton’s talk at times lack the smooth flow of a prepared text. Clifton does not reach his main point until page 31 of the transcript. For most of the first 25 pages or so of his talk, Clifton describes examples of difficult business and people situations that he has faced as Gallup’s CEO and how he has handled those situations. He begins by stating that when he heard what the Houston employees were thinking about, “I got very, very frightened. Probably as frightened as I’ve been for a long time about a company issue. Gallup has very, very big issues all the way around the world.” (JX 6 at 5) Despite these other issues, Clifton said that he wanted to come down (to Texas) and to tell the employees about “one of the very big concerns that I have. And that one concern will be what I will address.” But first Clifton launches into his description of several ex- amples of the tough situations that he has faced as CEO. Some of these had to do with unhappy Gallup executives and Gallup analysts regarding internal changes that Clifton had imposed or approved. Clifton threatened to fire them if they persisted in their opposition. Others pertain to customers and how he had handled these disputes, such as in telling Microsoft, “No.” (JX 6 at 22) Perhaps because he was in Texas, Clifton used the phrase “shoot out” to describe some of these confrontations (JX 6 at 10), as well as his expression about needing someone to “draw their guns” (JX 6 at 34) with him and ask the questions he was posing. Clifton makes clear that, to resolve these disputes, he stood firm so as to uphold his view of Gallup’s integrity and reputa- tion. In the most dramatic of these situations, Clifton and the CEO of a group in Taiwan—a group claiming rights to the Gallup name—had to stand before a judge in Taiwan who would determine which CEO was lying. The liar, as deter- mined by the judge, would go (and did go) to prison. Thus, Clifton put his personal freedom on the line in a foreign land. He won. Had the Taiwan group won, Clifton asserted, it would not have told the truth in its polls. That would have devastated Gallup’s reputation in the United States. (JX 6 at 23–25) Clifton describes other stories—all meant to emphasize the integrity of Gallup’s data gathering, the preeminent position of trust that the Gallup Poll has acquired over 60 years because of that integrity, and the independence from special interest groups that the Gallup Poll exercises in order to protect that integrity and to earn that trust. Clifton asserts (JX 6 at 27) that people call the Gallup Poll a “national treasure.” At one point Clifton describes the future of Gallup’s telephone interviewing. Reporting that Gallup is increasing its social audits so much that it is paying for construction of its own building in Wash- ington, DC, Clifton tells the group that “the future of telephone interviewing is very, very strong.” (JX 6 at internal 13–14) Finally, Clifton begins addressing his main point. I quote most of his remarks as to this main, and concluding, point (JX 6 at 31–36): Our biggest selling point is our independence. Our in- dependence can never be compromised or our future is ab- solutely in jeopardy. So let’s just say that we’re working with some large company in Chicago and it’s heavy competition. Compe- tition gets very, very tough in this business. And our competitors say, by the way, here at Sears, one of the things that you might be interested in about the Gallup or- ganization is that their data is collected by the United Steelworkers Union. That changes everything. And there’s no way through that. Now, forget about the busi- ness part, but it’s the perception. You’ve changed Gallup forever the minute that’s an honest answer. Let me take you a place more dangerous, that’s why I came here. We have a very important election coming up. And it’s probably going to be the Texas governor maybe against Al Gore, maybe it will be against Bill Bradley. [Two paragraphs omitted here pertaining to social au- dits that go to the White House.] And, but let’s say that it’s a sticky issue. And here’s the point I want to make with you. The decision that you have affects this and our future more than any decision I’ve ever seen in this company. Maybe with the exception of Chinese control over the Gallup name. [Paragraph omitted about poll on impeaching the President.] Our company, this process, our trust is the highest it’s ever been. Everybody’s scratching their heads, they know there’s somewhat of a disconnect between leadership and the voice of the United States. But imagine that, let’s say Bill Bradley suddenly shows very close to Al Gore. Al Gore is much more—well, let’s say it’s the other way, Al Gore is much more for unions than Bill Bradley. These are the dots that you need to connect. You got to ask yourself, is this what I want to do? Be- cause when they look down they’ll say none of this adds up. And Gallup is always right. And then they say, [Punctuation needed but not added.] say, did you hear, this high score for unions, something that we need to know is that that data’s collected by the United Steelworkers. At that moment our trust is absolutely gone. And you can do anything you want with that thought. But that’s GALLUP, INC. 1237 what happens. So you got to ask yourself in your jobs, am I having trouble, am I uncomfortable? People at Gallup have many of those same feelings. In a lot of companies they do. I would strongly encourage you to work them out yourselves. If you can’t, ask yourself do you want to in perpetuity change a 60 year trust that the Gallup organiza- tion has built between leadership and the citizens of the United States with one vote? But if you vote yes, it’s very powerful. That also makes the national news. [Quotes added.] “Interviewers for Gallup work for the United Steelworkers.” If you do that that changes everything. This is no longer the Gallup Poll, the Chinese might as well [have] beat us in Taipei. I wouldn’t have needed to have gone. But I need somebody to draw their guns right now with me and ask yourselves, do you want to knock this trust out of this 60 year company? WOMAN: No. MR. CLIFTON: Well, that’s why I would encourage you, maybe, maybe just, maybe just in closing, please sit down and divide this issue into two thoughts. Write your questions down. Let’s answer them. Let’s keep working on them. I know there are some people here that have out- standing jobs. I know that there are people here that make a lot of money and I know that there are people here that have managers that have an enormous amount of respect and care for you. I’m one of them that does. Our interviewing with Dr. Gallup before he died, he said somebody ought to look at that company and they said why, Dr. Gallup? And he said, because that company knows the importance of interviewing more than any other company in the world. We feel that way. It’s not quite as good as it is now. I’ve seen the Q 12s, the Q 12 scores, all that. They’re high. They’re much higher than companies all around the world. We don’t see scores like these. Other companies don’t have them. The United Steelworkers, those people that work full time there, they sure don’t. That’s an organization that’s in enormous decline. I would ask you to ask yourselves, is that the group that can help you? Do you need help from that? You’re up here. This is one of the best workplaces. But you see, we’re not satisfied. I don’t blame you. The people that work with me, they’re never satisfied. Work to get it back where it is. Work to get whatever the requirements are, whatever the pay is, just win it back yourselves. I’m behind you a hundred percent. I came here, I dropped everything I had, I wanted to meet with the su- pervisors and I wanted to talk, and I wanted to talk with you. Please vote no. Please vote no with anything that has to do with United Steelworkers. Write no on it. Don’t participate. Because when you put that pen in your hands it’s not just about your job, it’s about whether or not we’re the voice of 260 million Americans. I appreciate—you’re an excellent audience and I ap- preciate you listening very much. And I’m anxious to see your questions. Thank you very much. Good night. Following Clifton’s speech, employees submitted their writ- ten questions. By her covering memo dated June 3 (GCX 39, Rejected) to the Houston interviewers, Jane Miller, vice presi- dent and director of operations, transmitted, by a 25-page at- tachment, a listing of the 92 questions asked with the answers given as to each, in a Q and A format. (All questions were answered, although a handful that were either personal or ir- relevant received a “No response.”) Miller reports that the italicized answers are by CEO Clifton, and that the regular font answers are a collective response by “Your Houston Managers” and “Jean Timmerman and Jane Miller” (and possibly Clifton). The memo continues, with its last paragraph being a statement relevant to the issues here. At trial the General Counsel offered this document (GCX 39, Rejected), plus argument about a one-page memo (RX 12) from the Houston managers, titled, “Back to the Basics,” for the purpose of lending context not only to CEO Clifton’s May 26 speech, but to all other allegations of the complaint. (4:707– 708) The General Counsel further argued that the document, specifically the answers to questions (Q) 29 and Q33, in con- junction with the “Back to the Basics” memo (RX 12), shows animus. (4:709) The memo (RX 12), distributed about June 1 or 2 (2:297– 298, Uria-Ruiz), had been offered by Gallup to show its “course of action” in relation to the complaints received about harassment and work interference from the opposing factions, and I received it on that basis. (2:297–299) Respecting the Q and A document (GCX 39), the General Counsel conceded (4:707) that none of the complaint’s inde- pendent 8(a)(1) allegations attacks any part of the 26-page document. Gallup objected to the receipt of the Q and A document on the basis that there is no independent allegation regarding it and the speech itself is what should be evaluated under complaint paragraph 23 (plus conclusory par. 27), and the Q and A document (coming nearly a week after the speech) is irrelevant. (4:705–706) Asked for an example of the animus reflected by the Q and A document, the General Counsel cited the answers to Q29 and Q33, in conjunction with the “Back to the Basics” memo. (4:709) That memo explains that the recent events have caused the usually productive Houston office “to veer off course somewhat.” All the distractions and discussions, plus some people working fewer hours, have caused employees to turn their attention away from the work at hand and have caused the office to fall behind on some projects. “More importantly,” the second paragraph begins, “our quality has suffered. Recent evaluation scores have been at least one to two points lower than they usually are.” The third (of four) paragraphs begins, “It is time to get back to the basics. We are the Gallup organi- zation, the best and the most trusted name in research in the entire world. It is time for us to refocus our attention on ...” the work at hand. The fourth and last paragraph observes that the next few days will be very important, and the employees are urged to be at work, to do their best work, and to catch up on the projects. The General Counsel made an offer of proof, respecting RX 12, that if permitted to testify, Patrick Snyder would testify that DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1238 RX 12 gave employees [as they interpreted the memo] a mes- sage that their distraction, as the company perceived it, was a result of their union activities, and that such distraction was causing their production and quality to suffer. (4:704) The questions and answers cited by the General Counsel read (GCX 39 at internal 8): Q29. If the union issue goes any further is there a chance that this location will shut down? A. No. The presence or absence of a union has noth- ing to do with an office closing. What is important at any facility is that we maintain our performance, quality, and our client’s needs relating to quality of service. These standards remain and will not change at all if a union is se- lected by employees. The first part of Q33 reads, “What is to stop you from clos- ing this office to keep the union out or if the union wins?” The answer to this portion of Q33 is a repeat of the answer to Q29. (Moreover, Q81 and Q88 are rather similar to the first portion of Q33, and the answers there also duplicate the answer given to Q29. The answer to all four questions are in regular type, not italicized.) Asked how the cited answers to Q29 and Q33 demonstrate animus, the General Counsel responded (4:709) that the “Back to the Basics” memo (RX 12) “states in a veiled manner, Be- cause of the union organizing drive, you guys are letting your productivity and quality fall off.” That (veiled) statement, con- sidered in conjunction with (from the answers to Q29 and Q33), “What is important in any facility is that we maintain our per- formance, quality, and our clients’ needs relating to quality of service,” is what shows animus. (4:709) I sustained the objection and, on request, placed GCX 39 in the Rejected Exhibits folder. (4:709–710) By the Govern- ment’s subsequent offer of proof, the General Counsel prof- fered that, if allowed to testify, Patrick Snyder would testify that he received the “Back to the Basics” memo (RX 12) on June 2, that he understood the content of the memo to be sug- gesting that, because of the union organizational drive, the quality and productivity of employees was being undermined. Moreover, on the very next day, June 3, Snyder received the Q and A document (GCX 39), and in it Q29 and Q33 relate to the possibility of plant closure if a union organizational drive were successful. Part of the answer to those two questions states, “What is important in any facility is that we maintain our performance, quality, and our clients’ needs relating to quality of service.” Snyder would testify, the General Counsel continued, that he understood the answer to those two questions (Q29 and Q33), in conjunction with the “Back to the Basics” memo (RX 12), to suggest to employees that their productivity and quality had dropped off dramatically as a result of the union organizational drive, and when productivity and quality drop off, that plant closure or facility closure is a real possibility, that he under- stood this to be a veiled threat of plant closure, and he under- stood this to demonstrate Gallup’s opposition to the union drive. (4:710–711) After colloquy with the General Counsel, I stated that my ruling remained the same. (4:713) In this connection, one must wonder how Snyder’s vivid imagination, and the Government’s puzzling acceptance of his interpretation of the documents, would blind them both to the first sentence (first two, actually) of the answer to Q29 (that I quoted at trial, 4:709):” No. The presence or absence of a union has nothing to do with an office closing.” My point in describing these documents is to explain that I now think I should have received in evidence the Q and A document (GCX 39) and should have lifted any restriction on the memo (RX 12) so that the parties would be free to argue their points. The General Counsel wants to argue animus, as I have just described. The General Counsel was not asking for any unfair labor practice finding respecting either the Q and A document (GCX 39) or the “Back to the Basics” memo (RX 12). She made her offers only to lend context and to show animus. Of course, Gallup’s position is that the speech should be evaluated, under Section 8(c) of the Act, solely on the basis of the speech itself, and without regard to other events (particu- larly events which come later). It is risky when a judge, at this stage, reverses a ruling and receives documents. This is so because the parties (particularly the opponent) will not have the opportunity to submit additional briefing or to offer any evidence deemed relevant and material should the new ruling work to the prejudice of the opponent (Gallup, here). Although any error I made in rejecting the documents (GCX 39 and in not generally receiving RX 1216) probably would be determined to be harmless, I still am in- clined to receive the documents generally. I do not think this receipt will be prejudicial to Gallup. Accordingly, I now re- ceive GCX 39 (including the covering memo) and RX 12 with but one limitation — this otherwise general receipt is not to make either document available as a target for an unfair labor practice finding. (I do not reverse my rulings rejecting offers of proof by the General Counsel respecting what proffered testi- mony would be about the meaning of GCX 39 and RX 12. The meaning is a matter of law.) Having now received the Q and A document (GCX 39) in evidence (and having moved it from the Rejected Exhibits folder to the folder for the General Counsel’s exhibits), I return to the covering memo and quote Vice President Miller’s last paragraph there: Gallup recognizes that employees have the right to support or not support a union. We respect that right and we will do eve- rything possible to protect that right. However, if it comes to an NLRB-conducted election, we urge all Houston interview- ers to vote no because we do not believe that a union is neces- sary or beneficial to you and your work environment. Be- cause we are so concerned about the negative perception which may develop concerning the independence of the 16 Technically, to carry the Government’s point forward, the General Counsel should have offered a copy of RX 12 under the Government’s next exhibit number for the expanded purpose summarized by the Gen- eral Counsel at trial. It is not for one party to keep the opponent’s exhibit number and to offer it for an expanded purpose. However, our focus was on more important things and I do not fault the Government on that procedural point. GALLUP, INC. 1239 Gallup Poll if Gallup employees are represented by a union, we intend to exercise all of our legal rights to oppose the un- ion. Turning now to the questions and answers (GCX 39), I note that the following (all answers in the original are italicized, thereby indicating a response by CEO Clifton), have some di- rect relevance to Clifton’s speech. (Bold emphasis added.): Q9. Are thee other unions which would make an ac- ceptable one for us to join? A. No. Unions, by nature, have very serious political agendas. They are special interest groups who represent candidates, issues and ongoing agendas. You become part of this [special] interest whether you like it or not through your dues, which support political action committee funds and so on. The United Steelworkers’ primary mission ap- pears to me to be safety in the workplace. At their most recent meeting in Las Vegas, their meeting high point, ac- cording to the published report on the internet, was paying tribute to the families who had members “killed or maimed” on the job as reported by BNA in the Daily La- bor Report. This is a worthy cause; however, it doesn’t apply to Gallup. If workers are being killed and maimed without improvements in safety conditions, this is an im- portant purpose. However, the presence of a union repre- senting our employees would generate the appearance of a lack of independence among many. Q21. The United Steelworkers Union is a “Union.” A union of employed personnel of a specific company to rep- resent specific issues “locally.” If the interviewers are the same and the researchers are the same and all the data goes through the same format then specifically how can the un- ion destroy Gallup’s integrity? The union only enforces federal law contracts. A. I understand that employees have the right to vote for or against a union and support your right. However, I encourage you to vote no because a union destroys the perception of, if not the actual independence of Gallup’s reported results. Unions are perceived as very partial spe- cial interest groups, and so they are. Rightly or wrongly, in my opinion Gallup would be perceived as “biased” by a workplace partnership with a union. Q92. Reactionary speeches are not the greatest counter-measure. Perhaps the “qual-serve” or serv qual is inadequate. Clifton’s appearance will be ineffective unless changes are made in the way interviewer Charging Party’s are handled. You can trust that can’t [?] unin- formed employees will set irrational changes if they feel “powerless” as Clifton was so fond of saying. Union in- volvement is ridiculous but so are people. CREATE AN ALTERNATIVE! A. Yes. I agree with your point more than any other I have read or heard regarding this matter. In defense of my “reactionary” speech, I am more concerned than you will ever know about the perceptions of “truth and independ- ence” regarding The Gallup Organization. We live that perception of trust. The whole world looks to Gallup for the truth. I am very passionate about our integrity and the perceptions of that integrity. That is why I rushed to make a speech containing information that was critical to every- one, the supervisors included. While we are considering “context” documents, recall that early in this decision I quoted from CEO Clifton’s “Welcome” page in the January 1999 edition of the Interviewer Training Manual (ITM) where he states (JX 23 at 2): Gallup has a commitment to quality that no other company in our industry can match. Nothing is more important to us than the accuracy and reliability of our data. And three pages later the training manual (ITM) declares to the new interviewer (JX 23 at 5): The “Gallup” name is the most trusted in market research. b. Discussion To the General Counsel, the message of CEO Clifton’s speech is clear. Clifton himself would discharge managers and risk prison in a foreign land to maintain perceived neutrality and integrity of the Gallup Poll, “and a vote for unionization threatened this perceived neutrality and integrity more than any previous threats.” (Br. at 24) The Government further argues (id.) that it may be safely assumed that “the employees under- stood quite clearly that Clifton was therefore willing to go to similar lengths (discharging employees and violating the law) to prevent the threat which he believed unionization posed to Gallup.” Also, the General Counsel argues (Br. at 25), that Clifton of- fered “not a shred of supporting documentation, simply the bare assertion that the future of the Company and the employees was in jeopardy if unionization took place, thereby automatically compromising Gallup’s perceived neutrality and integrity.” As is clear from its pleading and motion, Gallup contends that CEO Clifton’s remarks are simply opinion speech pro- tected by Section 8(c) of the Act. Respecting the Government’s argument about no documentation, Gallup contends that such was not required because CEO Clifton expressed general opin- ions and did not make predictions as to the “precise effects” that would flow from unionization. And it is respecting the latter, under NLRB v. Gissel Packing Co., 395 U.S. 575, 618 (1969), that documentation must be cited. Some 60 years ago a comic book character called “Plastic Man” was created. In the comics of that day, Plastic Man was depicted as capable of stretching long distances to reach what- ever object, person, or place he desired. An arm and hand could stretch around corners or through windows to grab some- thing or someone, for example.17 The Government’s argument here pictures for me Plastic Man’s arm and hand stretching across a very wide chasm in order to grab a prize. Thus, in order to connect CEO Clifton’s words with the Government’s conclusions, the General Counsel becomes Plastic Man and stretches her arm an impossible distance in order to put the grab on CEO Clifton on the other side of the gorge. A major point of CEO Clifton’s May 26 message is that if Gallup’s independence ever became compromised, then 17 For a brief history of Plastic Man, see http://plasticman.virtualave.net/ DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1240 Gallup’s future would be jeopardized. That is Clifton’s opin- ion, although it seems clear that it is a common-sense conclu- sion based on years of experience in his industry. To that opin- ion Clifton adds another—any association with a union, such as the United Steelworkers Union, would be perceived as com- promising Gallup’s independence, and the solid reputation for trustworthy polls that Gallup had developed over a 60-year span would be lost. Clifton expresses that opinion again to say that if the national headlines were to read, “Interviewers for Gallup work for the United Steelworkers,” that would change everything—it would no longer be the Gallup Poll. He then asks if someone will “draw their guns” with him to, in effect, defend against losing this 60-year trust. Clifton ends his speech by asking the em- ployees (JX 6 at 36) to “Please vote no,” and to “Write no” on anything related to the United Steelworkers so that Gallup can remain “the [trusted] voice of 260 million Americans.” In his speech, Clifton employed some hyperbole to empha- size the connection between the employees, and the Union, and the public perception. Not making the distinction between being represented by the Union for collective bargaining pur- poses, Clifton eliminates the distinction and describes a “per- ception” tied to data “collected by the United Steelworkers Union” (JX 6 at 31, 33), and to news headlines that Gallup interviewers “work for the United Steelworkers” (JX 6 at 34). Such exaggerations, while factually inaccurate, convey no threat. Nor do the colorful phrases of “shoot out” (merely describ- ing confrontations internal to Gallup) and “draw their guns” (essentially asking employees to say “No” to the Union) have any connection to threats about “general reprisals” or a “loss of their jobs” as alleged in complaint paragraph 23. The problem with the Government’s case is that it is based on an impossible stretching. Thus, nothing justifies the stretching made in the Government’s argument (Br. at 24): It may be safely assumed that the employees understood quite clearly that Clifton was therefore willing to go to similar lengths (discharging employees and violating the law) to pre- vent the threat which he believed unionization posed to Gallup.18 This is pure fiction. (In none of the confrontations that he described did Clifton mention that he violated the law in any way. Moreover, while he threatened to fire executives and some Ph.D. analysts because they were not supporting his pro- gram (JX 6 at 8, 10–11), he never said that he had discharged any employees.) The Government would stretch from Clifton’s description of past confrontations, where he has stood firm to protect Gallup’s integrity, to the desired conclusion that em- ployees would reasonably understand that Clifton was really saying that he would violate the law and fire them in order to prevent the unionization of Gallup. Instead of being a reason- able understanding by the employees, this really is Plastic Man stretching far beyond any logical connection. In fact, what 18 The General Counsel’s assumption suggests too much reading of the lines from Samuel T. Coleridge’s Kubla Khan: “And ‘mid this tumult Kubla heard from far, Ancestral voices prophesying war!” CEO Clifton was telling the employees was his opinion that any connection with a union would create the perception that Gallup’s interviewers would be tied to a special interest group, thereby compromising the integrity of their, and Gallup’s, polls. If that perception took place, the Gallup of 60 years would no longer enjoy the public’s trust. Employees could prevent this drastic prospect by saying “No” to the Union. In his speech, CEO Clifton never threatens that he would take action against employees in order to prevent unionization of Gallup. Instead, he urges them to support him and to tell the Union “No.” The “perception” that he describes as, in effect, devastating is something that would be generated by the actions of others, and not by him. Clifton acts to prevent that percep- tion from ever having a chance to develop by coming to speak to the employees and to explain to them why he urges them to tell the Union “No.” I find nothing unlawful about CEO Clifton’s May 26 speech. To the extent that, in evaluating CEO Clifton’s speech, the June 3 Q and A document (GCX 39) and the (about June 2) “Back to the Basics” memo (RX 12) should be considered from the standpoint of context and possible animus, I find no animus present. The “Back to the Basics” memo simply asks employ- ees to refocus on work so that they can catch up on the work projects. In no way does this show animus. An employer is not required to stand back and permit em- ployees to discuss some topic, any topic, at will. Work time is for work. If an employer, as do most, permits personal conver- sations during working time so long as they do not interfere with work, then fine. But that employer is free to draw the line when productivity begins to suffer. Employers are free to say, “Talk during work, if you like, about any topic—politics, sports, religion, pornographic photos posted on cubicles, or unions—but just make sure that such talk does not interfere with productivity.” And if it does interfere with productivity, then employers are free to tell the employees to refocus their attention on reclaiming the expected level of productivity. Of course, employers may not watch productivity drop when talk about other topics gets excessive, and then tell employees to get back to work when the productivity drop is associated with talk about union issues. No disparity has been shown here. Accordingly, I find no animus reflected in the “Back to the Basics” memo. Nor do I find that it adds any particular context for evaluating CEO Clifton’s May 26 speech. Turning now to the Q and A document (GCX 39), I find that to the extent it lends context to CEO Clifton’s speech, the con- text it lends reflects that CEO Clifton’s May 26 speech was lawful. This tone of lawful context begins on the covering one- page memo by Vice President Jane Miller with her last para- graph. In that paragraph (quoted earlier), Miller reassures em- ployees of their protected right to support or oppose a union, and she there advises employees that Gallup, using all its legal rights, will resist unionization in order to prevent “the negative perception which may develop concerning the independence of the Gallup Poll if Gallup employees are represented by a un- ion.” The answers to the questions that follow Miller’s memo reflect this same lawful description of the situation. Finally, I also consider Clifton’s speech in light of my find- ings of contemporaneous unfair labor practices. Although I GALLUP, INC. 1241 have dismissed most of the independent 8(a)(1) allegations (respecting events in May and early June), I have found merit to several (some eight in number). However, these eight inci- dents do not involve threats of discharge or other retaliation nor of office closing, nor even coercive interrogations. Instead, they pertain to overzealous restrictions on the protected activi- ties of the Union supporters. Accordingly, I find that the few May 1999 unfair labor practices that I have found add only a limited “flavor” to the context for CEO Clifton’s speech, and they certainly do not serve to convert a lawful speech into something less. Having now considered CEO Clifton’s May 26 speech (JX 6) both in isolation and in the context of the two roughly con- temporaneous documents (GCX 39; RX 12), and finding no animus in any of the three either singly or collectively, much less the unlawfulness alleged, and also having considered, as part of the context, the several unfair labor practices found as to some eight incidents in May, I shall dismiss complaint para- graph 23. 9. The client visit of August 1999 a. Facts This final allegation of independent 8(a)(1) is in two parts. On August 6, complaint paragraph 24(a) alleges, Gallup “in- structed employees to remove all written materials concerning the Union from their work areas until August 11, 1999.” By its answer, Gallup “partially admits” but “denies that such instruc- tions related only to union material.” (In a moment I address whether paragraph 24(a) is really intended to be only a factual allegation laying the background for paragraph 24(b).) On August 7 and 8, complaint paragraph 24(b) alleges, Gallup “removed all union materials posted in employees’ work areas.” Gallup denies “but admits that on August 9 the Respondent removed anti-union t-shirts from interviewer desks as well as material concerning certain clients of the Respon- dent.” For its affirmative defense number VIII, Gallup avers that at all material times it has adhered to its preexisting policy con- cerning the cleaning of work areas before and during visits of clients of Gallup and “made no exception with respect to either pro-union or anti-union material.” By an undated memo (but marked as distributed on August 5) addressed, “Attention All Interviewers,” Gallup (by the local “Managers”) advised (RX 15, emphasis added): On Monday [August 9] and Tuesday [August 10] of next week the VISA client will be visiting in the office. As with every client visit we are again asking you to clean your area! This means making your desk a professional area. In other words please take down all information that states any clients names, unprofessional pictures (you know the ones in bathing suits) against or pro union materials and any other materials that have Gallup business on them. We really want to have a great visit so please make sure your desk is up to par or we will do it for you—Thanks—Your Managers. PS—You can put it back up on Wed. This notice, prepared by Supervisor Heidi Roberts at Super- visor Uria-Ruiz’s request, was distributed on (Thursday) Au- gust 5. (2:321, Uria-Ruiz; 15:2946–2947, Roberts) Roberts testified that, about the Monday morning of VISA’s visit, she removed union buttons from the desks of Eileen Juneau and Kathy McCarthy, placing the items in the drawers of their cubi- cles. From Tim Wong’s desk Roberts “took down” (meaning, apparently, that the items were posted inside Wong’s cube) “some pictures of ladies in bathing suits,” and she put those photos in the drawer of Wong’s cubicle. (15:2947–2950) Supervisor Uria-Ruiz testified that the visit was by one of her clients. As she started to leave for the airport to pick up the client’s representatives, Uria-Ruiz saw some antiunion T-shirts hanging on the backs of chairs in the executive interviewing section. She gathered them, folded them, and put them in Su- pervisor Doug Barlow’s office until after her client left, and then she returned them to the chairs where she had found them. According to Uria-Ruiz, she acted in accordance with Gallup’s policy of removing any nonbusiness item or postings of a com- petitor of the visiting client. On this visit Uria-Ruiz also re- moved a poster, posted in one of the cubes, showing a female model in a bikini. After the client left, Uria-Ruiz permitted the reposting of the poster showing the bikini-clad model. In preparation for the visit, Uria-Ruiz also asked Bobbie Solis to remove some items in an area near Solis of a competitor of the visiting client, and Solis did so. (Solis confirms. 15:3011) Uria-Ruiz did not remove any prounion material at this time. (Rather than suggesting that she knowingly left up prounion materials, this presumably means that she would have removed any had she seen it.) (2:316–323; 15:3063–3067) Eileen Juneau reports that, while all her union materials were removed from her cubicle (the union materials were posted on the inside and the outside) on this occasion, other nonwork items (not described) remained posted. She observed the same as to other cubicles near her where union supporters had their union materials removed, but the nonwork items, including some satanic items, remained posted. Her neighbors had some nonwork items posted both on the outside and the inside of their cubes. Among the nonwork items allowed to remain posted were photos of scantily clad men and women, and none that she saw was covered. (3:636–642, 662–663, 667–668) Interviewer Glen Lambert, whose posted union literature also was temporarily removed for the client visit in August (3:531, 552), moved from consumer interviewing (night work) to the day shift for executive interviewing on October 1 (3:539). Lambert gave some testimony about posted photos in at least six cubicles “that I work around” of “cheesecake” and beefcake.” (3:534, 552–553, 559) Lambert was not present during the client visit in August, nor during another client visit about mid-October. (3:525, 549; RX 16) During the October client visit Lambert was working days. Clients apparently visit during the evenings, and it is not clear whether they have any occasion to visit the area of executive interviewing. Lambert’s testimony is unclear whether the six cubes with the posted pho- tos of scantily-clad men and women are in his new area or in his old (consumer) area. Moreover, the evidence is not con- nected to any evidence that such photos remained posted during a client visit. Lambert reports that when he came to work the October morning after the client visit the night before (in the consumer DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1242 area), there were nonwork items posted (in the executive inter- viewing area) that had been posted before the visit. (3:537– 538) However, we are not told whether he had arrived at work before the owners and that therefore he could testify that the nonwork items (not described) had not just been reposted after having been removed before the previous evening shift. As Lambert’s testimony is too generalized, I attach no significance to it concerning the October posting. b. Discussion Although complaint paragraph 24(a) can be read as alleging that Gallup’s promulgation of its August 5 directive (RX 15) is either facially unlawful, or unlawfully motivated, or both, on brief (Brief at 20) the Government relies only on disparity of enforcement. That is, the Government does not argue that the rule itself is invalid, but instead argues against only the dispa- rate manner by which Gallup enforced the directive. This is an argument in support of complaint paragraph 24(b). It therefore appears that complaint paragraph 24(a) is intended to be merely a factual allegation to lay the foundation for the action (removal of the posted union materials) described in paragraph 24(b). In its own brief, Gallup addresses complaint paragraph 24 as a disparity allegation. Although complaint paragraph 27 (the conclusory allegation) lists paragraph 24 without carving out and naming only paragraph 24(b) as a violation of Section 8(a)(1), I find that such was the intended allegation. To the extent that I am wrong, and that complaint paragraph 24(a) is intended to allege the issuance of the August 5 memo as an unlawful promulgation, then I hereby dismiss complaint para- graph 24(a) as based on insufficient evidence. Turning now to complaint paragraph 24(b), I note that at trial Gallup referred at one point (3:526) to “special circumstances,” but on brief (Br. at 170; Reply Br. at 54) Gallup argues that there was no disparity in enforcement and that, in any event, the matter was so temporary as to be “de minimis,” meaning, in the full Latin phrase, that “the law does not concern itself with trifles.” B. Garner, A Dictionary of Modern Legal Usage 263 (2d ed. 1995, Oxford Univ. Press). Taking the latter defense first, I see that the cases cited by Gallup involve situations where the employer quickly corrected an isolated situation once it was brought to higher manage- ment’s attention. Those cases do not describe our situation, and I therefore find the de minimis defense inapposite. (Moreover, in one of the cases cited on brief, Xidex Corp., 297 NLRB 110, 111, 116 (1989), the Board reversed the judge’s de minimis finding.) As to disparity, Gallup errs by confining its analysis to whether its Houston management enforced the rule equally as between prounion and antiunion materials. That is only part of the scene that must be compared. The relevant question here is whether the union topic itself, pro and con, was disfavored while other nonwork items (items unrelated to the union issue) were permitted to remain posted. It must be said that Gallup failed to show that its supervisors, or a supervisor designated for the purpose, went through the office and made a good faith effort to carry out its warning: “or we will do it for you.” Aside from the limited actions of Su- pervisors Roberts and Uria-Ruiz, there is no evidence that a supervisor made an inspection (after the limited removals by Roberts and Uria-Ruiz) and found the cubes in compliance with the directive (RX 15). Indeed, although Roberts testified well after Eileen Juneau, Roberts did not rebut Juneau’s testimony that Roberts removed only the union materials and left posted the other nonwork items that included satanic literature and near-pornographic pictures. In light of the evidence (including photos, GCX 26) of the types of almost pornographic photos that are permitted to be posted inside the cubes, Juneau’s credited testimony, and the lack of evidence that Gallup made a general inspection to en- sure compliance with the directive (RX 15), I find that the Government established disparity of treatment that disfavored the materials related to the union issue while permitting other types of nonwork items (including satanic literature and near- pornographic photos) to remain posted. An employer is not required to open its bulletin boards (or walls, doors, or work areas) to the posting of union literature, flyers, and messages. But if the employer permits employees to post other nonwork items in such places, it may not exclude postings about unions. Eaton Technologies, 322 NLRB 848, 853 (1997). Although an employer may show, as an exception to the rule, that “special circumstances” require the removal of union postings, mere identity as a union message, as distin- guished from a posting containing egregious remarks, does not qualify for the exception. Id. at 853–854. And here the direc- tive had no restriction such as “egregious” concerning the union materials, nor did it confine its reach to an excessive number of posted materials or to the posting of such materials on the out- side of cubicles. Also, if we view the situation here as similar to the wearing of union insignia, and compare the visit of a client here to cases where employers wish to limit the wearing of union insignia in areas open to the public, the rule as to special circumstances is the same. That is, mere customer exposure to union insignia alone does not constitute a special circumstance allowing an employer to prohibit the display of union insignia. E & L Transport Co., 331 NLRB 640, 648 (2000); Meyer Waste Sys- tems, 322 NLRB 244, 244 (1996). Stated differently, the pres- ence of union insignia, without more, does not convert a work area of an otherwise acceptable appearance into a zone of un- professional appearance. As the cases report, the “special circumstances” exception can permit an employer, who seeks to present a certain “public image,” for example, to restrict the wearing of union insignia or posting of union materials. Here, however, Gallup does not defend on the basis of special circumstances, and apparently for good reason. For example, Gallup did not demonstrate that it went through the office and made sure that the office was “cleansed” of everything posted other than, perhaps, some qual- ity achievement certificates (such as shown photo 7 of GCX 26, which appear to be awarded to Tim Wong). The special cir- cumstances exception, however, is not available where the motivation is unlawful. E & L Transport Co., supra. Under all the circumstances here, and finding merit to com- plaint paragraph 24(b), I find that Gallup disparately enforced its August 5, 1999 directive that employees remove various nonwork postings, including union-related items, from their GALLUP, INC. 1243 cubicles temporarily during the visit of a client. By such dis- parity of conduct, I find that Gallup violated Section 8(a)(1) of the Act. (Earlier I dismissed complaint paragraph 24(a), to the extent that it alleges an unfair labor practice, on the basis that there is insufficient evidence to support such an allegation.) C. Allegations of Discrimination 1. Introduction As mentioned at the beginning of this decision, complaint paragraph 25 names the following four employees as having been discharged (unlawfully per par. 28) because of (complaint par. 26) their union activities: Lynne Zieler — June 22, 1999 Janice Rinehart — June 24, 1999 Sherri Lee — June 30, 1999 Patrick Snyder — August 6, 1999 Admitting the fact of the discharges, Gallup denies that such violated Section 8(a)(3) of the Act. From its statement of af- firmative defenses in Gallup’s answer to the complaint, Gallup, after giving the grounds for the terminations of Zieler, Rinehart, and Lee, states that each “would have been terminated by the Respondent even if she was not involved in any union activi- ties.” As for the grounds of termination for these first three, Gallup asserts that Zieler “was terminated for using profanity in a tele- phone survey with a survey participant,” and that as to Rinehart and Lee, each was terminated “for falsification of hours worked.” Respecting Snyder, Gallup avers that he was terminated “for making offensive comments to a coworker of a sexual nature and would have been terminated for this event and prior events based upon their cumulative effect by the Respondent even if he was not involved in any union activities.” Following the General Counsel’s resting of the Govern- ment’s case in chief (11:2119), Gallup moved (11:2119) to dismiss complaint paragraph 25 as to Rinehart and Lee. After hearing argument by both Gallup and the General Counsel, and after observing that there was too big a risk involved after 11 days of evidence, I denied the motion. (11:2136–2139) At that point Gallup, as it recognized (11:2140), was put to an elec- tion—it could either rest on its motion to dismiss as to Rinehart and Lee, or it could proceed with its defense as to those two (and as to everything else). If it proceeded with its defense as to Rinehart and Lee, then Gallup would waive its motion to dismiss complaint paragraph 25 as to them. See Alexandria Manor, 317 NLRB 2, 4 fn. 3 (1945). Although most cited cases usually involve motions to dismiss the entire complaint, the waiver rule applies to whatever the motion addresses. In presenting its case in defense, Gallup elected to present its evidence as to Rinehart and Lee. Accordingly, in making my decision as to those two, I base it on all the record, and not simply on the record as of when the General Counsel rested the Government’s case in chief. Price Chopper, 325 NLRB 186, 193 (1997). In the never-ending distress of these unfortunate cases, one wonders whether there will ever be “Harmony and understand- ing, Sympathy and trust abounding.” Perhaps that age will dawn: 19 When the moon is in the seventh house, and Jupiter aligns with Mars. Then peace will guide the planets, And love will steer the stars. 2. Lynne Zieler a. Introduction Zieler’s case is a “whodunit” that would rival some of the bestsellers if it involved a murder, international or corporate intrigue, or a threat to national security. My resolution of the complaint allegation will not solve the mystery. Hired in January 1994 and fired on June 22, 1999 (5:884, 934), Zieler had worked for Gallup for over 5 years (5:937) at the time of her discharge. During her time at Gallup, Zieler never received any discipline. (5:927–928). Moreover, her quality evaluations (GCXs 3, 4; 1:64–71, Uria-Ruiz), averaging a score of 49.89 out of a possible 50 points, reflect that she was just barely shy of perfect. Zieler’s role in the Union’s public organizing was very prominent. In addition to being one of the 20 signers of the Union’s letter of May 16 (RX 5), Zieler was the initial spokesperson for a group of the in-plant organizers that confronted Supervisor Uria-Ruiz on May 14 and asserted that Uria-Ruiz was violating the National Labor Relations Act. (1:37–39, Uria-Ruiz; 5:888–889, Zieler; 6:1026; 7:1323, Lee) A few minutes later, at the mandatory meeting downstairs, Zieler protested “That’s not true” more than once to statements about unions that Uria-Ruiz made to all the employees. (6:995–996) Finally, Zieler was one of four in-plant organizers (thee of the four are among the discriminatees here) who met with Supervisor Uria-Ruiz about posting and distribution on June 4. (GCX 44) I summarized that meeting earlier. Despite Zieler’s prominent contact with Supervisor Uria-Ruiz (some- times in the presence of other supervisors), I must note that Uria-Ruiz was not the discharging supervisor as to any of the alleged discriminatees here. (2:224–225, Uria-Ruiz) The sole ground for Zieler’s discharge (initially on June 22, converted same date to a suspension, pending further investiga- tion, reconfirmed next day, June 23, 1999, as a discharge) is that Gallup found merit to an alleged June 17 telephone com- plaint by an alleged “irate” survey participant (asserted to be Kevin Kline) that, in an attempted telephone survey on June 7, Zieler allegedly asked the reluctant Kline, “What is your prob- lem, asshole?” Supervisor Stephanie Geiger, the discharging supervisor, as- serts that Gallup policy calls for automatic discharge if an in- terviewer uses profanity with a survey respondent (participant). (13:2508) If such a policy is in writing (and Zieler asserts (5:937) that such a memo issued just before she was termi- nated), no copy was introduced into evidence. Even so, Zieler agrees that such is Gallup’s policy, although it is unclear whether that agreement is based on anything other than her own standards and her view of common sense. (5:937–938) Zieler strongly, and persistently, denies ever using profanity with any survey respondent. (5:910, 913, 917–920, 935, 937) Because 19 James Rado and Gerome Ragni (lyrics) and Galt MacDermot (mu- sic). (Aquarius, 1966) DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1244 Zieler would have spoken that June 7 with some “300” survey respondents, she has no way of recalling whether she spoke with a Kevin Kline. (5:917–918) Gallup’s telephone records show that a call was placed, using Zieler’s identification num- ber (IDN) to one “Kevin Kline.” Zieler was at work that day, claimed as work time the period that included the time of the call, and made calls before and after the call to Kline. Further investigation by Geiger and Supervisor Uria-Ruiz during and after the brief suspension reflects that the call was made from Zieler’s telephone extension. Still, sabotage is a possible ex- planation. As a term, “profanity” is used by the witnesses generically. They apply it not only in the sense of abusing the deity, but also using it to cover vulgarities and, presumably, obscenities. Ac- cordingly, in this decision I use the term “profanity” in a ge- neric sense that covers all three concepts. I credit Zieler who appeared to me to testify with a sincere demeanor and persuasive credibility. However, the question is what did Gallup, particularly Supervisor Stephanie Geiger, reasonably believe. If I credit Geiger, does that sink Zieler? If I do not believe that Geiger acted in good faith, will that, plus my believing Zieler, serve to establish the Government’s prima facie case? Speculation (and some confusion) abounds about the tele- phone records, records that include not only those of Gallup but also those of the third party provider, MCI WorldCom. One thing is clear. The Government does not dispute that Zieler may have made the June 7 call in question. What the Govern- ment disputes is that Zieler was rude or vulgar with “Kline” or whoever answered the call. Consequently, the General Counsel argues that the Government’s prima facie case is established by a combination of things. (Br. at 30–31) First, there is all the animus displayed. (Of course, I have dismissed many of those allegations.) Second, one example of asserted disparity. Third, as Zieler credibly denies the accusation (and I credit Zieler), this “utter innocence,” plus her extensive union activities (there is no dispute that Zieler was very prominent in her union activi- ties), plus the antiunion campaign that resulted in “numerous” acts violating Section 8(a)(1) of the Act (I have found far fewer than that which is alleged), then “it is apparent that the invidi- ous purpose for which it [Gallup] invented this alleged miscon- duct was guided by union animus.” And (Br. at 31): Respondent contends that, even if Zieler is not guilty of the conduct alleged, Respondent’s discharge of her was privi- leged as it took place based upon their good faith belief of Zieler’s misconduct. This argument must fail, however, as there is not a shred of evidence to support a good faith belief that Zieler engaged in this misconduct. If the Government prevails here, it will have to be on some- thing other than a finding that “there is not a shred of evidence to support a good faith belief that Zieler engaged in this mis- conduct.” Not only is there a “shred” of evidence, there is, as I have already indicated, a substantial amount of evidence that Zieler made the call and therefore would be the person who spoke the vulgarity. (The General Counsel’s “not a shred” rhetoric makes the Government look silly, and invites me to distrust any point the Government advances, especially when expressed by phrases loaded with hot air rather than based on reason, logic, and record support). Despite such substantial evidence, the question perhaps will be whether the record shows (by disparity, for example) that Gallup seized on the purported incident as a pretext to rid itself of one of the more prominent supporters of the Union. I have not overlooked the Government’s surprise witnesses (on whose testimony the Government also relies to argue in favor of a prima facie case). I say surprise because it does not appear that Gallup had advance knowledge of the gist of their testimony. These witnesses are Kevin Kline and his father, Donald Kline, with the latter testifying first. Residents of Pennsylvania, the Klines flew to Houston and denied that any- one from Gallup had ever been rude to them or called either of them an “asshole.” (2:250, Kline; 2:265–266, K. Kline). Kline (the elder) testified that, about June 1999 (2:249, 253–254, 257), he received one call, and only one call (2:253–255, 258– 259, 262), from Gallup, that the caller, a woman (2:250), asked for Kevin Kline. Kline said that Kevin was not there (Kevin had moved from his parents’ home on May 15, 2:249–250, 265), and the interviewer said that she was calling on behalf of (the hospital). Kline replied that Kevin was not satisfied with his visit because he had been switched from one doctor to an- other. The interviewer asked no further questions. The call lasted no more than a minute. (2:249, 254–255) Neither Kline (2:250, 255–256) nor his son Kevin (2:266) ever called to re- port that the Gallup interviewer had been rude or called him an asshole. At the time of the June call, Kline’s son, Kevin, was 22 years of age. (2:249) The “bombshell” impact of this surprise testimony is dimin- ished by the obvious fact that it came well after Zieler’s dis- charge. The testimony of the Klines, whom I credit, certainly adds to the mystery in the case. With the crediting of the Kli- nes, does that suggest the possibility that the entire incident was faked or “planted” so as to sabotage Zieler? There are cases that involve evidence faked or “planted” so as to provide a fraudulent pretext for discharging an unwanted employee. I have presided in such cases myself, as I cite in the recent case of Parts Depot, Inc., 332 NLRB 670, 720 (2000). b. Sequence of events (1) Per Gallup’s witnesses (a) The complaint Patty Collette wears several hats at Gallup’s Houston office. One of her functions is to answer incoming telephone calls. (12:2414) About 7 p.m. on some date in June, Collette testi- fied, she answered an incoming call from an “irate” male com- plaining about a rude interviewer who had called him. Under standing instructions about such matters, Collette transferred the call to “mission control,” and then paged on the intercom for a manager to pick up the call. Later that evening Supervisor Geiger, on delivering papers for Collette to enter data into the computer, thanked Collette, in (feigned) sarcasm, for sending her the call. Geiger then told Collette, “Well, I think I’m going to have to fire Lynne. He said she cursed at him.” (12:2416– 2420) GALLUP, INC. 1245 Supervisor Stephanie Geiger testified that she took the call that June 17. After identifying herself and asking whether she could help, the caller said that he had received a call for an interviewer who had called him an “asshole” and who then hung up the phone on him. Geiger apologized, and the caller said that he did not want to be called again. To Geiger’s in- quiry about the interviewer’s name and the date of the call, the caller said that it was “Lynne” and that the date was June 7 (10 days earlier). Asked for his name and phone number so that he could be removed from the system and not recalled, he gave his name as Kevin Kline and number as [I do not make that num- ber public]. (13:2478–2479) (Although Geiger omits the phone number from her testimonial description, she has it re- corded in the “Complaint Log”—JX 16.) Based on this call, and after Zieler was terminated,20 Geiger prepared a one-page “Complaint Log” (JX 16) that has four blocks for information. (13:2477–2478) In the top block of six lines, Zieler is named as the interviewer, June 7 is the date of the (survey) call, Kevin “Kline’s” is the name of the (survey) respondent, and Kline’s telephone number, as given above, is listed. The second block is for a description of the complaint. That description reads (JX 16): When the Respondent refused to do the survey, Ms Geiger said, “What is your problem, asshole?” The Respondent called our 1-800# and I spoke with him about the phone call at which time he told me the interviewer’s first name (Lynne) & [that] she called him on Monday the 7th. The third block is for “Action Taken” (13:2478) and reads: Interviewer was terminated after phone history & phone re- cords were pulled. The phone history [JX 7 at 2; 14:2859] re- vealed that Lynne’s id # [IDN] was the only id # that actually spoke w/the respondent. I pulled the phone record [RX 32 at 4, line 172; 13:2492], the contact was made from Lynne Zieler’s station [13:2492–2493; 14:2682, Geiger]. The fourth block simply asks for the name of the (supervi- sor) handling the complaint, and Geiger signed her name there. (13:2478) A moment ago I referenced Geiger’s testimony about when she prepared the foregoing complaint log. I interpret her testi- mony that, as of the initial discharge on June 22, Geiger had before her two documents, one of which was the complaint log (14:2859), to mean that she had the complaint log as completed through blocks one and two, with blocks three and four being the “complaint log” that was not completed until “after (13:2478) the termination. (b) Geiger’s investigation As already indicated, it appears that Geiger’s first action was to fill out the first two blocks of the complaint log. (JX 16) Second, Geiger then investigated calls to Kevin Kline’s tele- phone number by checking the “Phone Record Display” as reflected on the document (JX 7 at 2) in evidence. (13:2480) A disconcerting aspect is that the testimony could have been 20 On cross-examination, Geiger asserts that at the time of the initial termination she had before her two documents, one of which was the complaint log, JX 16. (14:2859) I address this in a moment. more specific. Thus, at trial Geiger testified (13:2480) that she first saw the phone record display document (JX 7 at 2) when it was printed the day before Zieler was fired. By this, Geiger apparently means on Monday, June 21, before the initial dis- charge on June 22, since it is one of the two documents (14:2850) that Geiger had as of the initial discharge. The implication, however, from the record is that Geiger pulled documents up on the computer screen for inspection, perhaps days before the screen, or document, was printed. Presumably, therefore, Geiger first saw the computer screen with the phone record display long before June 21, and that earlier time presumably would have been in the minutes after she had the conversation with the “irate” caller the evening of Thursday, June 17. As she describes, to pull up the phone re- cord display, she simply clicked “F” for “Find,” enters a tele- phone number (such as the one assertedly belonging to “Kevin Kline,”) and up pops the screen that we now see as JX 7 at 2. (13:2483) This screen has additional information not shown on the screen that faces the interviewer. (13:2482–2483) In any event, before the initial (June 22) discharge, Supervi- sor Geiger had the telephone history (another name for the phone record display, 14:2859), from Gallup’s computer re- cords, that showed four calls to Kevin Kline’s telephone num- ber. (JX 7 at 2; 13:2484) The first such call was made on June 7 at 7:34 p.m. and concluded at 7:37 p.m. The interviewer’s IDN is shown as 3886 (which Zieler concedes was her IDN, 5:934; 6:973). The interviewer logged the call status as “104,” a code for “nonspecific callback.” Geiger explains that such a coding means that the number would be called, randomly (by any interviewer), at some later time. (13:2484–2485) The second call, also by Zieler’s IDN, was the following day, June 8, is shown as less than a minute at 4:03 p.m. for code “101,” meaning “no answer.” (13:2485) Call three, placed on June 10 at 1:10 p.m. for 2 minutes, was by IDN 3194, that being the number for Sarah Lakey. (14:2683) The status code of “104” means that a scheduled callback time was made. Geiger acknowledges that Lakey would have spoken with someone at that telephone number because the status code so indicates. (13:2485; 14:2863–2684) Call four, placed on June 15 for 2 minutes, was by IDN 3714, Nancy Sico, who logged the call as a code “20,” meaning a screen failure (JX 31), which in turn means that whoever answered the telephone was not qualified to answer the survey questions. (9:1773; Rinehart; 13:2486, 2662, Geiger) A sepa- rate line, above the calls, has a code “12” followed by the mes- sage, “doesn’t live here[.] left 1800 #.” Geiger asserts that such note would have been inserted by the last caller (Nancy Sico) because it fits only the last call. (13:2488) Geiger inter- prets the entries to mean that the last caller (Sico) determined that Kevin Kline no longer lived there. The 800 number was left, by policy only on request, so that the survey respondent could call Gallup’s 800 number and register any comments. (13:2987) Also on June 22, the day of Zieler’s initial termination, su- pervisor Geiger ran what is called an “intcheck” pertaining to Zieler for June 7. (13:2492; 14:2860–2861) “Intcheck,” or “int check,” is an abbreviation for the term “Interviewer check.” (1:110, Uria-Ruiz) The term is a name for a computer utility DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1246 program by which supervisors can call up a listing of the calls made by an interviewer for a given date. It can be used by supervisors to check the production of interviewers. (10:1904, 1906, Selix; 16:3150, Stipulation) John Selix, a System Ana- lyst with Gallup (at Lincoln, Nebraska) for the last 11 years before he testified (10:1794, 1901, 1910), is the “Father of the Intcheck,” for he created the computer program about 5 years before he testified. (10:1900, 1904–1905) A supervisor at the Houston office may review 10 to 15 intchecks a week. (1:112, Uria-Ruiz) And supervisor Uria-Ruiz, for example, normally does not print hers, but simply reviews it on the computer screen. (1:114) Geiger ran an intcheck (RX 32) as to Zieler for June 7 in or- der to see if her IDN matched the call in question and to estab- lish the time that Zieler was working that June 7, and particu- larly to see if Zieler made phone calls immediately before and after the call at issue. (13:2494–2495) The intcheck shows that she was logged on from 10:03 a.m. to 8:03 p.m.—10 hours. (13:2507) That includes breaks and meal times. This intcheck has 180 lines of calls (apparently 180 separate calls) for Zieler (IDN 3886) that June 7. (RX 32) Line (or call) 172, matching the phone record display (JX 7 at 2) for that June 7, is the call to Kevin Kline’s number, Geiger asserts. (The numbers on the two documents so match.) (13:2492–2493; 14:2862) Calls are shown immediately before and after. (RX 32 at 4; 13:2494) That would show, Geiger asserts (13:2495), that Zieler (pre- sumably) “was sitting at her extension the entire time, [and] that no one else could have come in and used her extension to make a phone call.” The source or sources of the data reflected on JX 7 at 2 (the phone record display) and RX 32 (the intcheck) are not entirely clear. However, the phone record display (JX 7 at 2) appar- ently reflects data extracted from Gallup’s database for the computer screens facing the interviewers. (13:2482–2484) By contrast, the intcheck is derived from a primary document (the “LL12”) which itself appears to be drawn from Gallup’s data- base for its computer generated telephone calls. (10:1796, 1887, 1904, Selix) There may be only one database, but differ- ent computer programs extract different data to serve different purposes. On June 22 Geiger computer generated a third document (a June 22 version of RX 51), but did not save a copy at the time (13:2501), that shows the hours worked by Zieler on June 7, such being 8.30 hours, with 7.40 hours being on the specific project (Abington) that day. The relevance of this, as Geiger explains, was to determine whether Zieler claimed production hours covering the total time shown on the calls made with her IDN. (13:2498–2507) For this purpose, RX 51 falls a bit short in that it does not show the specific hours that Zieler claimed as worked; it merely gives her total of 7.40 hours claimed for that day on the Abington project. (RX 51). However, when that information was compared, on June 22, with the 180 calls shown on the intcheck (RX 32) for the Abington project for that June 7, one could reasonably assume that the 7.40 hours claimed included the 3-minute call placed from 7:34 p.m. to 7:37 p.m. One qualification. The calls shown on the intcheck do not show the actual times of the calls themselves, but only of the time that the number is on the computer screen. (13:2497– 2498, Geiger) As we see shortly, the actual time of the call appears to have been a little over a minute. Although the records do not prove conclusively that Zieler was seated in her cube at the time of the call on line 170 of the intcheck (RX 32 at 4), that would certainly be a reasonable conclusion—as Geiger so concluded (13:2495). During her cross examination. Zieler confirms that she worked a full day that June 7 (it was a memorable day to her because she made money that day), and that her breaks were at times other than the time in issue. 5:935–936; 6:1006–1007) In summary, Geiger’s investigation disclosed that (1) Zieler was working that June 7, 1999; (2) she was ostensibly working at the time in question; (3) Zieler’s IDN is shown for the call in question as well as all other calls on that intcheck; (4) Zieler ostensibly claimed as hours worked time that included the time of the call to the telephone number given by the complaining “irate” caller; (5) Gallup’s records showed that the “irate” com- plainant, giving his name as Kevin Kline, gave a date that calls were made on the Abington project, including a call made to the telephone number given by Kline; and (6) that the “irate” complainant gave the interviewer’s name as Lynne—a name that matched the name of the interviewer, Lynne Zieler, making the call per the IDN and the hours of work claimed as well as Gallup’s phone records. If falling short of steel-trap proof that Zieler was the culprit, at the very least it is rather substantial evidence supporting a good-faith belief that Lynne Zieler made the June 7, 1999, call to the “irate” complainant. But this is only part of the record. Will the rest of the record support the Government’s irrespon- sible accusation (Br. at 31) that Gallup “invented this alleged misconduct”? I say irresponsible for two reasons. First, at trial the Government never attempted to shoulder the factual or legal burden of proving who actually made the call, or of proving that the “irate” caller (assuming that there was such) was an imposter and part of a scheme to get Zieler fired. Second, as I observed earlier, notwithstanding some silly rhetoric (such as “utter innocence,” noted earlier) in the Government’s brief, at least some of the Government’s case is that Gallup seized on the incident as a pretext to cover its true reason of union ani- mus. (c) The initial discharge Gallup’s policy, Geiger testified (13:2508), is that the use of profanity with a survey participant is grounds for immediate termination. Indeed, shortly before leaving Nebraska for Hous- ton, Geiger participated in the May 10 discharge of one Andrea Uithoven (JX 8) for calling a survey participant an “asshole.”21 (13:2509–2510) It was an easier case there because Uithoven admitted that she had done so. (13:2510) [The spelling of Uithoven’s name, not part of the joint cor- rections for transcript volume 13, GCX 1ppp at 1, is question- able. Although it could be with a “V” as Gallup suggests, Brief at 35 fn. 5, I have adopted the “U” as rendered by Geiger (13:2510) and my own reading of her signature on the exhibit, JX 8 at 1.] 21 “What is your problem you asshole.” (JX 8 at 2) GALLUP, INC. 1247 Based on her investigation (and on her knowledge of Gallup’s policy), Geiger and “the rest of the supervisors” de- cided to terminate Zieler. (13:2511) Although Geiger’s testi- monial description is a bit disjointed at this point, she first called her supervisor, Director of Interviewing Jean Timmerman (13:2474–2475), and reported the situation. Timmerman reportedly said that Zieler should be fired. Fol- lowing that call, Geiger met with Supervisors Chad Murray, Charles Davenport, and possibly Jason Becerra, and the group of supervisors decided that Zieler had to be terminated. This was on Monday, June 21, the evening before Zieler was (ini- tially) discharged. (13:2511–2513) Consistent with the version of both sides, Geiger then (at some point before the termination meeting) prepared (13:2573) the termination notice (JX 7 at 1), a one-page form dated June 22. In a box of 17 possible reasons, Geiger checked number 12, “Quality—Poor Quality/Not adhering to company stan- dards—Not Rehireable.” On the lines for “Comments,” Geiger wrote (JX 7 at 1): A call was made by Ms. Zieler on 6/7/99 to a respondent on the Abington POPS study. She spoke to the respondent for 3 minutes on this date. After a subsequent call we received a complaint directly from the respondent on our 1-800 number on June 17th. In this call, the respondent identified an inter- viewer named Lynne that had badgered him relentlessly. He said he wasn’t going to do the survey. She persisted. (This is a violation of hospital interviewing policy.) The respondent was then asked “What is your problem asshole?” This behav- ior is unacceptable and a violation of our mission. The following evening, June 22, supervisor Chad Murray es- corted Zieler to supervisor Geiger’s office. Geiger then in- formed Zieler that a survey participant who, identifying Zieler by (first) name, and specifying the call date as June 7, had complained that in the attempted survey, when he told Zieler that he did not want to do a survey, Zieler then said to him, just before she terminated the call, “What’s your problem, ass- hole?” Zieler replied that she did not remember the call, and asked whether she was being terminated. To this question Gei- ger answered “Yes.” At that point, apparently, Geiger tendered the termination notice to Zieler for her to sign, but Zieler de- clined without offering any explanation in her defense. After asking whether she could leave, Zieler left Geiger’s office. The termination meeting lasted only a few minutes. The termina- tion notice bears the signatures of Geiger and Murray, plus Geiger’s note that Zieler had refused to sign. (13:2510–2511, 2513–2514) (d) The suspension Later that evening of June 22, Zieler, accompanied by Sherri Lee and Janice Rinehart, met with supervisor Geiger. Supervi- sor Uria-Ruiz joined the group and, later, so did supervisor Chad Murray. (13:2514–2515) At this meeting Zieler stated that someone else had logged in with her IDN and made the call. Zieler asked to see the records, and requested a copy, on which Geiger was relying. Geiger said that she would check the next day to see whether she could give Zieler a copy. Uria- Ruiz (2:215) said that the telephone extension needed to be checked, and Geiger told Zieler that she would do so (13:2517). At that point Zieler’s termination was converted to a suspen- sion. (13:2518; 14:2858–2859) Supervisor Geiger concedes (13:2519–2522) that, on June 23 during the suspension stage, four employees (Marion Trovato, Catherine Wagley, Nancy Sico, and Beverly Robinson) came to her and said that they had never heard Lynne Zieler use profan- ity at any time, but that they had heard interviewer Sarah Lakey use profanity on numerous occasions. All four were among those who had signed the Union’s May 16 letter (RX 5) notify- ing Gallup of the in-house organizing committee. Geiger told them that if they heard profanity again they should come tell her so that she could investigate it. [At trial, two interviewers who opposed unionization testified that, in normal office con- versations, Zieler used “the F-word,” would say “shit” all the time, and her normal vocabulary was “consistent with” that of a “longshoreman” (12:2431–2432, Coffey), and that, in such normal conversation, Zieler “cussed like a sailor” (12:2443– 2444, Niehouse). I do not believe either Coffey or Niehouse. Apparent friends, both appeared to be testifying from a bias against Zieler rather than truthfully reporting factual observa- tions, and the excessiveness of their characterizations of Zieler detracts from their credibility.] The follow-up check by super- visors Geiger and Uria-Ruiz disclosed that the telephone call to Kline’s number came from Zieler’s extension, number 259 (Zieler acknowledges that such was her extension, 5:934; 6:978) and lasted for 1.3 minutes. (13:2489–2491; JX 7 at 3) As a result of the allegations against Sarah Lakey, Geiger also checked the record to see whether Lakey had spoken with Kline. The investigation disclosed that no contact was made, Geiger testified. (13:2522; JX 7 at 2) Actually, as Geiger ac- knowledges (13:2522), the phone record (JX 7 at 2) reflects that Lakey (IDN 3194; 13:2485; 14:2863) did reach Kline’s number on June 10 (JX 7 at 2; 13:2523) and that, in speaking with someone at Kline’s number, Lakey was asked to call back later at a specific time. Lakey therefore coded the call as “104,” meaning a scheduled time to call back. (13:2485; 14:2863– 2864, Geiger) From this, Geiger concluded that the only per- son who spoke with “Kline” on June 7 was Lynne Zieler. (13:2525) At trial we learned that MCI’s records do not show that Zieler’s call of June 8 (coded as a no answer) or Lakey’s call of June 10 (coded as a specific callback) reached Kline’s number. These are the second and third calls shown to Kline’s telephone number in June as of the day before the initial discharge. That is, MCI’s records do not show that a connection was made for those two calls. (12:2350–2355; 15:3001, Compagna; RX 46 at 1; RX 63 at 3) Gallup (Br. at 38–40) suggests an explanation that appears to clarify some of the evidence. Thus, the fourth call (that of June 15 by Nancy Sico) that resulted in “screen failure” (meaning, as noted earlier, that no one was present eligible to answer the survey) and the removal of Kline’s phone number from the system because Kevin Kline had moved (13:2486–2489, Gei- ger), squares with the testimony of Donald Kline, Gallup sug- gests (Br. at 40), that in the only conversation he had with the interviewer who called, he told the interviewer (2:254–255, D. Kline), “Kevin’s not here,” that Kevin was not satisfied with DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1248 his visit to the hospital because he waited “for an hour, and they switched him from one doctor to another.” Marking this a “screen failure” and leaving an 800 number “would have made sense,” Gallup suggests. And, to the point, “ . . . it also ex- plains the timing of the complaint call which Geiger received on June 17, 1999 from the person identifying himself as Kevin Kline.” (Br. at 40) Although Gallup does not elaborate on its “explains the tim- ing” suggestion, it appears that Gallup seems to recognize that there is a problem with the fact that the date of June 7 is 10 days earlier than the date the “irate” male called. Later, when I discuss the overall case as to Zieler, I address this matter fur- ther. (e) The final discharge Geiger and Uria-Ruiz then met with Zieler on June 23 in Uria-Ruiz’s office along with supervisor Chad Murray plus Sherri Lee and Janice Rinehart. After reporting the results of the additional investigation, Gallup (apparently by supervisor Geiger) terminated Zieler. (13:2517–2518, Geiger; 5:927; 6:977–979, Zieler) During the meeting, Zieler states, she made the point that she definitely would not have left the phone num- ber (Kline’s) in the system had she in fact called him a vulgar name. (5:926) Geiger’s response, if any, to that logic is not recorded. Indeed, as Zieler there noted, an interviewer can code a call in any way. (5:926) Thus, had she really called someone a vulgar name, all she would have to do, as we have seen earlier, is to code it as a “screen failure” and the number would have been removed from the system. Consistent with this logic, Gallup’s phone record display (JX 7 at 2), one of the records on which Supervisor Geiger relied, reflects that the very next day, June 8, Zieler again dialed (Kline’s) number—a highly unlikely act had she just called him a vulgar name the day before. (She coded that call as a busy signal.) Supervisor Geiger denies that Zieler’s union activities played any part in her discharge. (13:2519) (f) Geiger calls “Kline” Toward the end of June, Supervisor Geiger testified, she telephoned “Kline” to inform him of the results of her investi- gation. Supervisor Jason Becerra was sitting behind Geiger at Mission Control. [As with Geiger (13:2466), Becerra (1:38–39, Uria-Ruiz) is no longer employed at Gallup. Becerra did not testify.] Geiger asked for “Mr. Kline,” and a male voice an- swered “Yes.” Recounting the fact that she had spoken with him earlier about his complaint, she asked for something in writing because the matter would probably turn into a legal issue. “Kline” said that he did not want to be contacted again and, raising his voice, he said that he had told her not to call, that he did not want to be bothered. Saying that it was very important, Geiger asked if he would help. Replying that he did not want to be called again, “Kline” hung up the phone. (13:2527–2528) MCI’s telephone records (RX 46 at 3; RX 63 at 1) show a call to “Kline’s” number on June 28. The call was placed shortly after 4 p.m.22 (12:2356, 2367–2368) The call lasted 2 minutes (1 minute 59 seconds, to be exact). 22 Per the time at the “switch” which in this case is at Philadelphia. (12:2358; 15:3001–3002, Compagna) Kline operates his business from his home. Subcontractors sometimes are in his house. Although the subcontractors “normally” would not answer Kline’s telephone, Kline asserts, he acknowledges that such would be “possible.” (2:260–261) Thus, although not likely, it is “possible” that a subcontractor answered Supervisor Geiger’s telephone call of June 28 and that Geiger spoke with a subcontractor rather than with Kline. Whatever happened in this portion of the mystery, I find that it was not Donald Kline that Geiger spoke with. Moreover, as Gallup considered “Kline” to be an older man (1:147, Uria- Ruiz), it likely was not Kevin Kline who was just 22 at the time (2:249). On brief (Br. at 30), the General Counsel argues that Gei- ger’s account is “utterly incredible” in view of the denials by the Klines, disinterested witnesses who flew in from Pennsyl- vania to testify. Typically, the General Counsel’s head-in-the- sand argument is no help, for it fails even to mention the MCI telephone records, much less suggest how to resolve the dispute in light of those records. (2) Zieler’s additions The prominent difference between Zieler’s account and Gei- ger’s is that Zieler supplies more details. Most of the extra details are not essential. Contrary to Geiger’s description of Zieler’s reaction at the initial discharge meeting (June 22), I credit Zieler. Thus, I find that, on being informed of the details of the complaint, Zieler said, “Absolutely not. Never hap- pened.” Geiger then “whipped out” the termination paper, saying “Well, it did happen, he called; can you sign this; you’re terminated.” (5:910) Saying that she would not sign the paper, Zieler reiterated, “I did not do it and I will not sign anything that says I did. If that’s your final decision, then I guess this conversation is over.” Zieler then left. (5:910, Zieler) A little later that evening, after talking with Union Represen- tative Fennell, Zieler changed her position to a willingness to sign with the qualification, “I did not do it,” for the purpose of obtaining copies of the termination documents. (5:911–913) As noted earlier, the bottom line on this is that Gallup would not agree on the ground that the papers were the private prop- erty of Gallup. At this second meeting, the “suspension” meet- ing, Sherri Lee argued that in the past Gallup had exercised “discretionary justice,” and that some managers had supported the interviewers. (5:914) Janice Rinehart said that anyone could have made the call. Geiger replied no because Zieler was the only person to talk with the complainant that day, that he knows her name, the date of the call, and he asserts that she used profanity. When Supervisor Uria-Ruiz said that no one could use another’s IDN, Sherri Lee reminded her of the occa- sion that Sherri Lee had demonstrated how to confirm who had an IDN by typing in her IDN, for the computer then simply asks, “Are you Sherri Lee, yes or no?” (5:914–915) At trial Zieler, credibly asserting that she never gave her IDN to anyone (6:972, 1013), elaborated on the process of obtaining the name matching an IDN by explaining that such is done when someone forgets her IDN. In such cases, the forgetful interviewer simply types in the code “DAI” (JX 23 at 88; RX 35 at 88) and the computer then discloses the IDNs that Gallup has for the different projects. [The interviewers, including GALLUP, INC. 1249 Zieler, colloquially interpreted these code letters as standing for “Dumb–ass interviewer.” 5:915] Then, by locating the particu- lar project and inserting those IDNs, the person can ascertain the matching interviewer names. That is, the computer brings up the name of the person whose IDN is entered by asking, in the case of Sherri Lee, “Are you Sherri Lee, yes or no?” More- over, interviewers working the same project simply learn, over time, the IDNs of the other interviewers on that project. (5:915; 6:964–965, 1014–1018) As Sarah Lakey worked the same Abington project as Zieler, and even spoke with someone at the number in question here on June 10 (JX 7 at 2; 14:7863), the implication, from the standpoint of this line of argument, is that Sarah Lakey (who did use profanity, as Geiger acknowledges several employees told her on June 23) could have obtained Zieler’s IDN and herself called complainant Kline and ended up calling him an obscenity. Also at the suspension meeting Janice Rinehart kept asking whether Gallup was taking every investigative avenue, and asking why Gallup was taking the stranger’s word over that of Zieler, when there was no tape of the asserted conversation. Indeed, Janice Rinehart asked, how could the interviewers know that such a person really had called. Geiger said that she could pull more records and determine from whose “phone pad” the call was made. Supervisor Chad Murray said, “That’s irrelevant, irrelevant. It’s going to show that you [Zieler] called the number at this time of the day, and the [survey] respondent named you personally by name, [and] you’re the only one that could have talked to him.” (5:919–920) At first glance the final discharge meeting, on June 23, ap- pears to have an odd sequence, for it is during this meeting that Geiger receives an answer to her email to Nebraska that morn- ing asking for the telephone extension number from which the call to the Kline’s number was made on June 7. Thus, on June 23 Glen Kallhoff of the headquarters computer staff emailed Geiger that the extension number was 259 (Zieler concedes that was her extension number, 5:934; 6:978), and that the call lasted 1.3 minutes. (JX 7 at 3) Adding to all the mysteries, by an earlier email that morning, Kallhoff had advised Geiger, “We are not capturing all calls going out of Houston.” (JX 7 at 3) Respecting the four employees that Janice Rinehart took in, one at a time, to report to Supervisor Geiger about Sarah Lakey’s use of profanity, Rinehart credibly testified that Geiger told her, in advance, that she would listen only to reports about incidents in the past, and not about the current incident involv- ing Zieler. (9:1731, 1734, Rinehart) During her own testi- mony, Supervisor Geiger did not rebut this assertion of Rinehart. The explanation for the odd sequence of the meeting (further investigation still in progress) is that Zieler, on her own, in the late morning came to the premises in a renewed effort to obtain copies of the termination papers and also of her personnel file. Learning that most of the supervisors were at a luncheon meet- ing, she waited on the premises until they returned not long after the lunch period. (5:921–922, 956–957) In short, the final meeting began because of Zieler’s presence, not because the additional investigation had been completed. Even so, Su- pervisor Chad Murray’s opinion seems to have been even more solidified, for he remarked about as he had done at the previous meeting, except this time he sounded more final, adding, “and it’s just going to show that you did it, no matter what we inves- tigate.” (5:924, Zieler) At about that point (5:924; 6:976–978, Zieler) it was Murray himself who pulled up a screen on the computer [a copy of which became, it appears, RX 7 at 3] and apparently read from Glen Kallhoff’s email at 1:29 p.m. that date to Geiger reporting that the call in question had last 1.3 minutes and had been placed from extension 259. As noted earlier, Zieler acknowl- edges that her extension number was 259. (5:934; 6:978) The meeting ended with Zieler’s being discharged. (5:927, Zieler) During this final discharge meeting on June 23, as Sherri Lee credibly describes, and respecting the rejection of Zieler’s re- quest for a copy of her termination papers (6:1055) Sherri Lee asserts (6:1055–1056) that, in the past, Gallup gave terminated employees copies of their termination papers. She recalls this specifically as to one Michelle Rinehart who was terminated in either 1997 or 1998. Lee personally saw the papers. I note that this is a point that Lee apparently did not voice at the meeting, but told us at trial. Even so, Gallup did not respond at trial to explain, for example, either that Lee’s testimony was incorrect or that Gallup had since changed its policy before the union matter arose. Although the General Counsel makes no argu- ment based on Lee’s revelation, I find that it is a factor to be considered in the overall analysis of Zieler’s case. The final item of confirmation that would had to have been done after the final discharge meeting was the verification by supervisors Geiger and Uria-Ruiz that extension 259 was the phone that rang in the cube that had been used by Zieler. Gei- ger testified that she and Uria-Ruiz called that extension, as Geiger stood by Zieler’s old cube, and the telephone in Zieler’s old cube rang. (13:2490–2491) Although Uria-Ruiz did not testify that she dialed extension 259 on that occasion, Geiger testified that “They” walked over to the cube and the extension was ringing. (13:2490–2491) By this test, Geiger (assisted by Uria-Ruiz) completed all the investigative checks that Supervi- sor Uria-Ruiz had listed (2:214–215) at the suspension meeting. In crediting Lynne Zieler, I have considered the EEOC charge that she filed in September 1999 alleging gender and age discrimination as well as retaliation for speaking up for employee rights as protected by that statute. (RX 31; 5:944– 946) The filing of the EEOC charge is not inconsistent with her claim here (5:938, 944) that she was fired only for her un- ion activities. (She later stated that it was for all three (four) reasons. 5:945) Because there can be multiple reasons for a discharge, and because one or more reasons may be covered by different statutes, as here, multiple charges covering those sus- pected reasons are not inconsistent. In any event, the theory of the case comes from the General Counsel, not from the Charg- ing Party (the Union here), and certainly not from the alleged discriminatee. All that said, in weighing credibility, I have counted as a negative Zieler’s initial statements that union con- siderations formed Gallup’s “only” reason for firing her. (5:938, 944) Although Zieler corrected that to include the rea- sons alleged in her EEOC charge (5:945), such a switch de- tracts from her credibility. But overall, I find Zieler to be a credible witness with a generally persuasive demeanor. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1250 c. Disparity To show disparity in treatment, the Government points to a November 1998 situation involving interviewer Marion Trovato. Trovato testified that she was reassigned from the hospital “track” on which she was working (JX 26) based on several complaints.23 Her supervisor, Joan Dorwart, told her that one of the complaints was that she had said “goddamn” to the survey participant. Trovato denied it, explaining that she did not use that language. Dorwart said that she had conferred with Supervisor Uria-Ruiz, and that, based on their experience with Trovato and on monitoring tapes of other survey calls, they did not believe that she used impolite language, and for her to be “softer.” (Thus, it is not clear that Dorwart explained anything to Uria-Ruiz other than a complaint of “impolite” language had been registered against Trovato.) However, 2 or 3 weeks later Dorwart came to Trovato and, telling her that there had been five complaints against her in 2 days, said that she would have to be reassigned. Dorwart said that she thought Trovato had been “sabotaged,” but she declined to disclose the name of the person or persons that she suspected. (2:376–378, 389–395) Asked about the Trovato situation, Supervisor Uria-Ruiz tes- tified that, as reported to her, the complaints against Trovato were because of a perceived rudeness flowing from her New York accent during her calls to people in Alabama. (1:54–57; 2:217–218; 15:3067–3069) Trovato’s testimony about the profanity, as opposed to a complaint about rudeness, is diluted somewhat by the testimony of Government witness Catherine Wagley. Now Trovato recalls that Wagley was present when Supervisor Dorwart delivered the news about the profanity complaint. (2:376–377) However, although Wagley did walk in on the conversation, it was the occasion that Trovato was being removed from the track after five alleged complaints in 2 days (not the profanity complaint that was registered 2 or 3 weeks earlier). (2:417–418) Although I credit Trovato, who testified in a persuasive fash- ion, it is questionable whether much, if any, weight can be as- signed to this matter. Not believing the complaint, Supervisor Dorwart told Trovato to be “softer.” But all this was before the five complaints (for “rudeness,” apparently) that got Trovato reassigned. Thus, the disparity is limited in that it pertains to the “discretionary justice” of one Supervisor—Joan Dorwart. Moreover, there is no evidence that anyone in management knew of the nature of the complaint other than Dorwart. Thus, the complaint apparently was lodged directly with Dorwart (2:377), and the nature of the complaint apparently remained with Dorwart. Accordingly, I find that Supervisor Dorwart’s “discretionary justice” favoring Marion Trovato does not pro- vide any basis for finding that Gallup, and particularly Supervi- sor Geiger, failed to exhibit good faith by not extending leni- ency to Lynne Zieler as an act of “discretionary justice.” 23 “Track” is the term commonly used to describe surveys that are continuously worked for a particular client (such as a hospital) of Gallup in order to “track” opinion over a period of time. (1:76–77, Uria-Ruiz; 2:378–379, Trovato) Interviewers generally prefer track work to one time surveys. (Id.) Although not mentioned by the Government on brief as an- other disparity example, evidence was presented at trial that Supervisor Chad Murray condoned, or does condone, inter- viewer Laura Harris’ calling him obscene names in the pres- ence of other employees. (3:512, 544, Lambert; 4:746–747, Snyder) To some extent, Murray concedes, but he emphasizes that Harris, with her habit of rough language, has never chal- lenged his supervisor authority. (15:2983–2985) Whether such acts by Harris constitute improper work behavior is for Gallup to say, but it is clear that such conduct has no relevance to the use of profanity toward survey participants. d. Discussion As previously suggested, on the surface there is substantial evidence that Gallup, by Supervisor Geiger, conducted a rea- sonable investigation even before the initial discharge of Lynne Zieler the early evening of June 22, 1999. The additional in- vestigation reflected that the call of June 7 to “Kevin Kline” was from Zieler’s extension 259. Thus, when the discharge became final the early afternoon of June 23, Gallup’s records showed that the evening of June 7 Zieler had called Kline’s number (on the June 7 that he complained about, and that his complaint named “Lynne”), that Zieler had dialed other num- bers immediately before and after and had claimed hours for a full day (all indicating that Zieler was working in her cube when the call to Kline’s number was made), and that the call to Kline’s number came from Zieler’s extension 259 (as reported by email—actual verification that extension 259 rang in Zieler’s cube was not made until after the final discharge). To overcome this formidable array of defensive evidence, the Government launched a surprise attack. That is, rather than grapple with the defensive “line” by running or passing, the Government simply attempted to float over it in a colorful hot air balloon. Alas, supplied by too much heated gas, the Gov- ernment’s brightly colored balloon simply rose out of the briefs, and was last seen lifting toward the heavens where it can soar forever on inflated rhetoric in that great balloon land in the sky. Back on earth, the rest of us have to deal with reality, and the reality is that the telephone records (Gallup’s and MCI’s) must be acknowledged and addressed. The General Counsel, on brief, argues for a finding of violation almost entirely on the basis that the Government’s witnesses (primarily Zieler and the Klines) should be credited and those of Gallup (primarily Su- pervisor Geiger) should be disbelieved. The General Counsel’s passing reference (Br. at 30) to MCI’s records is that such re- cords, and the testimony of MCI Representative Dyanne Com- pagna, do not demonstrate that “Zieler had a conversation with Donald Kline in which she called him an ‘asshole’ as alleged.” The General Counsel’s placing of the burden on Gallup to prove Zieler’s guilt is a result of the Government’s confused articulation of the correct legal standard to be applied. At dif- ferent points the General Counsel seems to be arguing under the analytical structure of Wright Line, 251 NLRB 1083 (1980), (for example, “pretextual,” Br. at 28), only to end up arguing about the “utter innocence” of Zieler, plus her extensive union activities, giving “rise to an inference that it was those union activities that caused Respondent to discharge her for conduct GALLUP, INC. 1251 which never took place.” At that point (Br. at 31 fn. 15) the General Counsel cites Teledyne Still-Man, 295 NLRB 161 (1989), enfd. 911 F.2d 1214 (6th Cir. 1990), the only case au- thority cited by the General Counsel respecting Zieler’s case. Teledyne Still-Man is inapposite because the analytical focus there was on discharges for strike misconduct—acknowledged discipline for misconduct during protected activities—a con- cept governed by NLRB v. Burnup & Sims, Inc., 379 U.S. 21 (1964). But the cases of Zieler and the others here are the tradi- tional motivational-discharge situations controlled by the ana- lytical structure of Wright Line. That means that it is the Gen- eral Counsel’s burden to prove unlawful motivation, such as by showing pretext by in turn demonstrating either disparity or that Supervisor Geiger did an inadequate investigation and other- wise acted in bad faith. In that connection, the General Coun- sel’s inflated rhetoric about an inference of unlawful motivation based on Zieler’s “utter innocence” plus her union activities simply will not do. Nor will it do for the General Counsel to rest on rhetoric claiming (Br. at 31) that Gallup “invented this alleged misconduct” of Zieler. No amount of overheated rheto- ric, as illusory a crutch as it is, will serve as some MacBethian incantation by the witches, “Eye of newt and toe of frog, Wool of bat and tongue of dog,” to magically produce a prima facie case that Gallup did in fact invent a fraudulent complaint against Zieler. It takes hard evidence, not incantations. So let us start at the beginning—the General Counsel’s prima facie case, or lack thereof. The General Counsel argues that the Government’s prima facie case is established by: 1. Zieler’s exemplary record. 2. Zieler’s very prominent role as one of the in-plant organizers. 3. Animus reflected in the 8(a)(1) allegations. 4. Credibility resolutions favoring Zieler and the Kli- nes. 5. Disparity (the Marion Trovato incident). 6. “Not a shred of evidence to support a good faith be- lief that Zieler engaged in this misconduct.” [Br. at 31] Reformulating the last item from the inapposite legal theory to fit the correct legal standard, I interpret it to read: “Supervi- sor Geiger conducted a cursory investigation or otherwise acted in bad faith.” Items 1 and 2 are undisputed, and item 4 is estab- lished. For the moment, grant (contrary to some of my find- ings) the Government item 3 as to every 8(a)(1) allegation in the complaint. As to item 5, no relevant disparity is shown, as I summarized earlier. In short, the Government’s case comes down to the last item. For the Government to show a prima facie case, even if all of its 8(a)(1) allegations were to be found meritorious, it still must prevail as to reformulated item 6. Without a finding that Supervisor Geiger acted in bad faith, the General Counsel fails. As I have summarized earlier, the un- disputed evidence is that Geiger’s investigation, if not quite perfect, was adequate and reasonable at the very least. That investigation, as of the final discharge on June 23, pointed di- rectly at Zieler as the culprit. Moreover, the General Counsel made no effort to show that Geiger acted in bad faith. Even if I were to find in favor of reformulated number 6 (that Geiger conducted a cursory investigation or otherwise acted in bad faith), and further find that I disbelieve her denial that Zieler’s union activities played any part in her discharge, would that still be enough to find that Zieler’s discharge was, prima facie, unlawful? I think not, for that does not address the telephone records. Thus, in the absence of disparity to show pretext, there must be, I find, a number 7: 7: Fraud. The telephone records (both Gallup’s and MCI’s plus the June 23 email specifying Zieler’s exten- sion 259) are frauds containing planted data, or, such re- cords are legitimate but are rendered untrustworthy, and therefore to be disregarded, by virtue of findings that rele- vant entries are the result of fraud. One more mystery to add to the others is just how the Acting Regional Director decided to include Zieler in complaint para- graph 25. Presumably Region 16 obtained a copy, during the investigation of the charge, of the intcheck (RX 32) and of what ultimately became JX 7 (the termination paper plus the attached phone record display of the calls made and the attached page of the emails on June 23, including the final one from Glen Kall- hoff sealing Zieler’s doom by reporting that the June 7 call to Kline’s telephone number came from (Zieler’s) extension 259.) I say presumably because Gallup cooperated in the investiga- tion of the charges by at least giving affidavits from Supervisor Uria-Ruiz (1:43–44, GCX 2) and from Supervisor Geiger (14:2851, GCX 59) (neither affidavit offered in evidence). Even if Region 16 did not have the June 23 emails, with Zieler’s extension number, but did have copies of the intcheck (RX 32 at 4) and the phone display record (JX 7 at 2), then Region 16, in issuing a complaint that included Zieler, had to have a factual theory that would overcome the documents pointing to Zieler as the culprit. Without that factual theory (and the Trovato situation could not be counted as a disparity to show pretext), the Acting Regional Director apparently issued the complaint relying on the arguments as presented here in the General Counsel’s brief (confused commingling of the legal standards, and no factual theory of liability under the correct legal standard). That being so, the Government should be re- lieved that this is not an EAJA case, for if it were, the Govern- ment possibly would risk being faced with an order to pay Gallup allowable attorney’s fees and expenses as to the Zieler portion of the complaint. Because there are some important similarities between litigation and war, the Government would profit from considering an observation that Napoleon made in penning a letter in September 1806:24 In war, nothing is achieved except by calculation. Everything that is not soundly planned in its details yields no [favorable] result. So, does the Government lose here as to Zieler? Well, there are some troubling items underneath the surface. First, why would Donald Kline (assuming that he was the “irate” caller) wait 10 days to call and complain? Who among us remains “irate” after a 10-day cooling off period? Still insulted, even a tired anger, but “irate”? And with Kline credited, the strong 24 Peter G. Tsouras, editor, The Greenhill Dictionary of Military Quotations 363 (2000). DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1252 suspicion arises that the “irate” caller, assuming that there was one, was an imposter fraudulently planting an accusation for the purpose of sabotaging Zieler. Who prompted the imposter? Did Gallup? Did antiunion employees on their own devise this sabotage scheme, including a method of manipulating the tele- phone records even to show Zieler’s extension 259? If so, what about the phone records of the third party, MCI, that essentially support Supervisor Geiger? Many questions, but no answers in the record. Second, note the nearly identical language involved in An- drea Uithoven’s May 10 discharge (JX 8, “What is your prob- lem you asshole?) just before Supervisor Geiger came to Hous- ton, and the short question with the same vulgar term of which Zieler is accused. Is the nearly identical nature of the two statements mere coincidence, or did it possibly serve to suggest to Geiger, and to Gallup, a convenient fraud to plant in order to get rid of Lynne Zieler? Such a fraud is credible on the surface and difficult to disprove (typical characteristics of many frauds), particularly if planted by someone with access to Gallup’s computers and a knowledge of Gallup’s processes (that is, if planted by an insider). Having credited Zieler and the Klines, and aware that Supervisor Dorwart was of the opin- ion that an interviewer, such as Marion Trovato, could be “sabotaged,” I am of the view that someone or some group “framed” Lynne Zieler, an exemplary 5-year employee, and is getting away with the perfect fraud. Third, why did Supervisor Geiger feel compelled to express (with apparent callous indifference) to Patty Collette, the recep- tionist that June 17, that Geiger thought she would “have to fire Lynne” because an “irate” stranger claimed that Zieler had “cursed at him”? Probably before any investigation at all,25 Supervisor Geiger thought that she would have to fire an exem- plary employee with over 5 years’ of service to Gallup? Does this slip (and it surely was a slip of some kind) unwittingly reveal Geiger’s participation in a secret conspiracy with Gallup (a conspiracy that the evening of June 17 an “irate” male would call and accuse Zieler) to get rid of Zieler because of her prominent union activities? Or does it merely show that, as of June 17, 1999, Geiger was a bit too cocky—even callous— despite being a relatively new and inexperienced supervisor? (Hired in March 1995, Geiger served as an interviewer in Ne- braska for “3.8 years” before becoming a Supervisor there about January 1999. Her first day on the job in Houston was May 17. 13:2466–2467, 2636.) Fourth, why was Supervisor Geiger so eager to “whip out” the termination form at the June 22 initial discharge and, con- tradicting this exemplary employee’s solid denial (“Absolutely not. Never happened. 5:910), almost gleefully inform Zieler, “Well, it did happen, he called; can you sign this; you’re termi- nated.” (5:910) Where was any semblance of managerial con- cern (by Geiger, by her superior Jean Timmerman, or the other participating supervisors, Chad Murray, Charles Davenport, and possibly Jason Becerra) that an exemplary employee was being discharged? Is that the Gallup way, that exemplary em- ployees with over 5 years of service, even if they justifiably are 25 Geiger’s comment focuses on the claim, and does not reflect, as of that early point, that Geiger had yet checked any of the records. to be discharged, may be terminated in a manner exhibiting not even minimal courtesy, but callous indifference? If Gallup has a written policy about treating employees with fairness, cour- tesy, and respect, no copy is in evidence. Fifth, why would Supervisor Geiger, at least during the sus- pension period, not ask Sarah Lakey whether she had spoken with “Kline” on her call of June 10 rather than relying on her interpretation of Gallup’s phone display record? Granted, Lakey may well have said that she could not recall, but the concern here is about good faith, the appearance of fairness, and the lack of any motive to disregard any conflicting facts while proceeding to “nail” Zieler. The failure to check with Lakey in person does not exhibit good faith. Sixth, before Janice Rinehart brought in witnesses on June 23, during the suspension period, to report that they had heard Sarah Lakey, but not Lynne Zieler, using profanity, why would Supervisor Geiger have excluded the current Zieler incident from discussion? Geiger does not tell us. Although there is no evidence that Geiger was not told something about the June incident that may have been of substantial, perhaps critical, importance, it is Geiger’s closed-mind approach that concerns me. I count this as a negative. Seventh, why did Gallup depart from its past practice by re- fusing Zieler’s request for a copy of her termination papers (JX 7, at least pages 1 and 2 if not also the page 3 emails) after Zieler changer her position and agreed to sign under protest? This smacks of kicking Zieler after she already was down. I count this as a negative. Despite these troubling concerns, I am constrained to note that the one item that really damages the Government’s case is the email report from one Glen Kallhoff (JX 7 at 3) that the June 7 call (to Kline’s number) was placed from (Zieler’s) ex- tension 259. (Kallhoff did not testify and we have no details from him on how the database he derived this from picks up this information, or whether that information could be manually entered by someone, even overriding what the computer auto- matically registered. Of course, in the normal situation, some- one such as Supervisor Geiger would have no occasion to doubt the accuracy of such a report.) Tying Zieler to the extension number as the source of the call, given the intcheck lines show- ing that Zieler apparently was in her cube making a series of calls at this time, sealed Zieler’s fate as of the final discharge on June 23. Turn now to grounds 6 and 7. As to reformulated ground number 6 (that Supervisor Geiger conducted a cursory investi- gation or otherwise acted in bad faith), I find that, on the sur- face, Geiger’s investigation, as of the initial discharge on June 22, was reasonably close to adequate even though lacking a tie to Zieler’s extension number. The initial investigation, there- fore, was much better than merely cursory. During the suspen- sion stage, however, I find a negative factor respecting Geiger’s refusal to listen to any information from employees concerning the charge against Zieler. I also have found as a negative factor Gallup’s refusal to give Zieler copies of her termination papers, contrary to past practice. Moreover, Supervisor Geiger’s callous attitude toward a good employee such as Lynne Zieler, and Geiger’s seeming eagerness to dispatch Zieler, plus Geiger’s lack of a satisfactory GALLUP, INC. 1253 demeanor respecting this part of the case, persuades me that Supervisor Geiger was not acting in good faith. This includes the finding that I disbelieve her denial (13:2519) that Zieler’s union activities had anything to do with Zieler’s discharge and her assertion (14:2849) that Zieler would have been discharged regardless of her union activities. Further, I disbelieve Geiger’s testimony about a June 28 call to and conversation with “Kline.” Respecting the disbelief of a negative (the denial), note that the mere disbelief of a negative, without more, does not establish the positive of that which is denied where the burden of proof to show the affirmative is on the party (the General Counsel here) obtaining the disbelief findings. Such disbelief findings merely supplement positive findings. Stated differently, complaint allegations must be established by posi- tive evidence, and disbelief findings of denials, standing alone, will not substitute for the burden of establishing the allegations by affirmative (positive) evidence. Just what does my finding of bad faith by Supervisor Geiger prove? Certainly such bad faith is consistent with a theory of fraud (ground number 7). It also is consistent with a theory that Geiger secretly harbored a personal dislike for Lynne Zieler. However, there is zero evidence of the latter, and perhaps no more than that of the former. In short, the finding of bad faith on the part of Supervisor Geiger is not enough to establish a prima facie case of unlawful motivation. There must be more than a mere inference from Geiger’s bad faith that all (or the relevant data in) the records were doctored on orders from top executives at Gallup. (Yet science infers the existence of a “black hole” from the actions of nearby stars.) Thus, the fact that Geiger’s bad faith actions would be consistent with a mas- ter plan of fraud, dictated by top Gallup executives, does not prove the existence of such fraud. There has to be a support- able finding either that the telephone records (of both Gallup and MCI), plus the email showing the extension number as 259, were fraudulently manipulated so as to indict Zieler, or a sup- portable finding that top executives, either on direct orders from CEO Clifton, or prompted by expressions from Clifton similar to the ones (“shoot out” and “draw their guns”) in his Houston speech of May 26, on their own issued such orders to manipulate. Recall that earlier I dismissed complaint paragraph 23 that (in conjunction with paragraph 27) CEO Clifton’s speech violated Section 8(a)(1) of the Act. In some cases there is direct evidence of the fraud by which the boss directs an executive to falsify records so as to justify the discharge of an employee. Thus, consider this from the recent case of EEOC v. Premier Operator Services, 113 F.Supp.2d 1066, 1071 (N.D. Tex. 2000): Testimony of Defendant’s former management offi- cial, and Vice President, Marvin McCroy, established, by direct evidence, a violation of Title VII’s prohibition against retaliation. Mr. McCroy testified that Eric Brown, owner and President of Defendant, adamantly required that employees acquiesce to the blanket English-only pol- icy or be terminated. McCroy testified that signing the English only memo signified agreement to the policy as a condition to continued employment. Mr. McCroy also tes- tified that the president expected him to be an “enforcer” of the policy, and that he did enforce it. . . . . With regard to some class members, McCroy was or- dered by Brown to write that their terminations were “volun- tary” although they were not. Defendant stipulated that one Charging Party, Fran- cisco Gracia, achieved the third highest call completion rate out of 31 operators during the last pay period before his discharge; and McCroy testified that the other Charg- ing Party, Albert Estrada, was one of Defendant’s top per- formers. Nevertheless, the Defendant’s President ordered that backdated disciplinary notices be added to their per- sonnel files, after the fact, to justify their terminations. In other cases, the evidence, although less direct, is substan- tial, as in the case I cited earlier (and in cases cited there), Parts Depot, Inc., 332 NLRB 670, 720 (2000). Here I find no evidence of fraudulent manipulation of the re- cords (telephone and email) by, or at the direction of, Gallup’s management. True, I am very skeptical of those records. That skepticism is based largely on the lack of complete details about how the database is created, the records generated, and whether such records, or even the database, can be fraudulently manipulated to falsely indict an interviewer. This skepticism feeds a strong suspicion of actual fraudulent manipulation (by persons unknown) that resulted in the discharge of a very good employee, Lynne Zieler. Thus, I find that the General Counsel has failed to establish crucial ground number 7. This failure is fatal to the Government’s case as to Zieler. It was the General Counsel’s burden to prove ground number 7 as part of the Gov- ernment’s prima facie case. Ground number 7 was not some optional choice that the General Counsel could select or de- cline. As it appears (from the arguments in the General Counsel’s brief) that the Acting Regional Director included Lynne Zieler in the complaint without any viable theory (factual or legal) of liability by Gallup concerning Zieler’s discharge, it is clear that the allegation as to Zieler was doomed from the start. Accord- ingly, I reluctantly shall dismiss complaint paragraph 25 as to Lynne Zieler. 3. Janice Rinehart a. Introduction As mentioned in the introduction to the discrimination alle- gations, respecting the June 24 and 30, 1999 discharges of, respectively, Janice Rinehart and Sherri Lee, Gallup alleges, as an affirmative defense as to each, that each was terminated for “falsification of hours worked,” and that both would have been so terminated by Gallup even if there had been no union on the scene. Although I summarize the cases of Rinehart and Lee separately, the basic description of procedures, terminology, and similar matters that I give in a moment apply to both Rinehart and to Lee. Early in this decision, in the statement of the case, I noted that Rinehart and Lee were members of the Union’s in-plant organizing committee and that each had signed the Union’s May 16 letter (RX 5) to Gallup. Rinehart and Lee were among the in-plant organizers who were the most visibly active in DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1254 communicating with, and confronting, the local management, such as confronting Supervisor Uria-Ruiz on May 14, just be- fore the mandatory meeting of that date, when Lynne Zieler handed Uria-Ruiz the one-page union flyer (GCX 49) ad- dressed to managers and supervisors, and advising them, that “You Are Violating The Law! if you . . .” (followed by a list of nine examples of unfair labor practices, including discrimina- tion, threats, and interrogation). (5:888–889, Zieler; 8:1688– 1699; 10:1934, Rinehart) Also tendered to Uria-Ruiz at this May 14 meeting was a copy of the Union’s letter (RX 5) announcing the organizing drive and containing, at the bottom of the page, the signatures of the 20 in-plant organizers. The signatures of Janice Rinehart and Sherri Lee are among the 20 (with Rinehart’s being the first in the first column from the left margin and Lee’s being in the second column). In the minutes of the mandatory meeting that followed, Janice Rinehart is shown, near the middle of the first page, as announcing the date of the first general union meeting. (JX 18 at 1) At trial, when asked by Gallup’s lawyer about Rinehart’s comments, Supervisor Uria-Ruiz states that Rinehart stood and spoke without invitation. (2:282) Then at the June 4 meeting with Supervisors Uria-Ruiz and Chad Murray, Rinehart and Lee, along with two other in-plant organizers (Eileen Juneau and Lynne Zieler), were present for the in-plant committee. Rinehart left about one-third the way into the meeting, after she received an answer to her question of Uria-Ruiz on whether the “Grim Reaper” posting (by employ- ees opposing the Union) had been approved. (GCX 44 at 5) Of those four in-plant organizers in the June 4 meeting, three (Rinehart, Lee, and Zieler) were fired and are three of the four alleged discriminatees here (with Patrick Snyder, not at this meeting, fired on August 6, making the fourth). At trial, Rinehart served as the General Counsel’s designated assistant exempt from the sequestration order. (1:32, 37; 16:3167) Hired September 4, 1996, Rinehart worked for Gallup as a telephone interviewer until she was fired in a meeting with Supervisors Geiger and Barlow on June 24, 1999. (8:1686; 9:1736; 10:1931, Rinehart; 12:2276–2277, 2282, 2299, 2304, Barlow; 13:2677, Geiger) Although interviewers are expected to reach and maintain a quality level of at least 48 out of a pos- sible score of 50 (JX 23 at 19, ITM), Supervisor Barlow con- cedes (12:2400–2401) that not all interviewers do so. Rinehart’s 25 evaluations averaged a score of 49.48 (GCX 14) out of a perfect (1:64) score of 50. (1:181) Such reviews are a measure of quality (JX 23 at 19) and not of hours (1:182, Uria- Ruiz), and the issue as to Rinehart is not quality, but an alleged falsification of hours worked. Even so, the General Counsel argues (1:183) that there is relevance to Rinehart’s very high quality scores because there is a logical connection between the work ethic that achieves a standard of exemplary quality, and the same work ethic that strives to reach, and usually does reach, the highest levels of quantity (whether in production or in hours worked). As we see in a moment, Supervisor Barlow takes the issue a step further through his opinion, stated at trial, that Rinehart cheated (that is, she was dishonest in recording her hours), and there was, in his opinion, a purpose to her cheating. Even to attempt to understand all this, we will need a general understanding of Gallup’s complicated pay system. The only evidence that Rinehart had ever received any prior discipline is a May 20, 1999 written warning (JX 3) for “Har- assment.” The parties did not address this warning at trial (other than, the first morning, to put it into the record by stipu- lation, 1:8, 14), nor on brief. The written comments by Super- visor Chad Murray are that several employees had complained that Rinehart had interrupted them while they worked in order to promote the Union. Rinehart was warned that any “further harassing activity may result in termination.” Signing under protest, Rinehart wrote that the accusations were false and lodged because she was trying to organize a union, a right pro- tected by law. No complaint allegation attacks this warning. b. Overview (1) Termination meeting Gallup discharged Janice Rinehart for allegedly falsifying her work hours for the week of (4 days, actually) Monday, June 14, through Thursday, June 17, 1999. (1:58–60; 13:2676, Gei- ger; RX 53; JX 9) For that week of 4 days (Rinehart generally liked to complete her weeks by Thursday, 11:2104–2105), Rinehart submitted paysheets (GCX 16 at 1–4) showing a total of 40.5 work hours under the column heading for “Actual Hours.” Supervisor Geiger, however, calculated Rinehart’s “actual hours” for those 4 days to be 28.7. (13:2676; RX 53) Although Rinehart was a member of Supervisor Geiger’s team, and Geiger did the investigation and calculated the num- bers, there is no dispute that Supervisor Barlow took the lead at the June 24 termination meeting in Geiger’s office When Bar- low informed Rinehart that the foregoing discrepancy was the reason that she was being terminated, Rinehart exclaimed that the discrepancy was “impossible” because she arrives early, works late, and that Barlow is well aware of her long hours. (13:2678, Geiger) Geiger testified that Barlow (who confirms, 12:2299) responded (13:2678): I have no doubt that you’re here 40 hours a week, but you’re only working 28.7 of those hours. Rinehart asserted that she worked what she had claimed. (9:1740, 1785) Production was not the issue. Barlow told Rinehart that she had produced very well, but the problem was with the hours claimed. (91739, 1784; 11:1984, Rinehart) Barlow acknowledges that he said (12:2300), “Well, I recog- nize that you’re a high producer, but this doesn’t go to your high production. I goes to hours recorded.” Geiger added, “Janice, this is nothing against you personally. You’re a very high producer, but we cannot have you lying on your hours.” (9:1743, 1783, Rinehart) “Well, it didn’t happen,” Rinehart replied, but she was in “total shock” at that point. Thus, when, in response, Geiger offered to show the records to Rinehart, Rinehart declined and left after signing the termination notice (JX 9) that Barlow placed before her. (9:1742–1744, Rinehart) In signing the termination notice, Rinehart added the follow- ing protest (JX 9; 9:1742; 13:2680–2682): I don’t agree with this at all. It is trumped up charges once again because of my union involvement. Based on Gallup’s practice (13:2681, Geiger), Rinehart’s re- quest for a copy of the termination notice was denied by Super- GALLUP, INC. 1255 visors Barlow and Geiger. (9:1742–1743; 13:2681) Filed June 25, 1999, the charge in Case 16–CA–19898–3 was amended to include the discharge of Janice Rinehart. In September Rinehart filed an EEOC charge (RX 36) alleging discrimination based on gender, age, and retaliation. Part I of Rinehart’s statement of particulars on the charge asserts that she was dis- criminated against because of her activities through the “We Care About You Committee.” Recall that such committee is the name adopted by the Union’s in-plant organizing commit- tee. (11:2097) The existence of such companion charges are generally irrelevant to the proceeding before the other federal agency because there may be more than one moving motivation for a discharge. In theory, Gallup could have fired Janice Rinehart because it resented not only her union activities, but also her age and gender and to retaliate against her for exercis- ing her statutorily protected rights, as Rinehart essentially as- serts. (10:1932–1933). Other than saying that she thought she was fired for her union activities (10:1932), Rinehart, unlike Zieler (5:938, 944) and Sherri Lee (7:1304–1305), never asserts that it was the “only” or “sole” reason. Accordingly, I attach no impeachment weight to the EEOC charge or to Rinehart’s testimony about the matter. Respecting his testimony about the admitted desire to keep high producers, Supervisor Barlow stated that such goal has to be balanced with an overriding requirement that Gallup’s em- ployees maintain high ethical standards respecting such mat- ters. (12:2300–2301) Indeed, Gallup’s ITM (the interviewer training manual mentioned earlier in the statement of the case) provides, as the opening sentence under the topic on “Ethics” (JX 23 at 18): “You have been hired because you are ethical!” As part of the “Data Collection Agreement” (JX 23 at 61), ap- parently signed by Janice Rinehart and all other interviewers, each interviewer agrees, in part, to “uphold the high ethical standards by following all survey instructions and by accurately collecting and recording the data as stated by each respondent [survey participant] interviewed.” In Barlow’s opinion, in light of Rinehart’s experience and training, and because the total discrepancy was so large, Rinehart’s individual discrepancies were the result of cheating by Rinehart—a knowing misrepresentation of her hours—done for the purpose of avoiding the work required to maintain her status as a member of the 35-hour team and to avoid being bumped to the lower pay scale of the 25-hour team. (12:2302– 2304) Before moving to the next topic, I pause to note that the record gives no direct explanation of why Supervisor Barlow generally presided (9:1784, Rinehart) at Rinehart’s termination when Rinehart was a member of Supervisor Geiger’s team and when Geiger was over the Adventist Hospital track (10:1926– 1927). Since May 1998, Barlow, as his exclusive duty, has been managing the Executive Interviewers. (11:2142–2143) Geiger testified that she did not want to terminate Rinehart because Rinehart was a very productive interviewer and Geiger liked her personally. (13:2626–2627) Perhaps it could be in- ferred that Barlow generally handled the termination because Geiger perhaps would have been less than comfortable in doing the deed. (2) Gallup’s pay plan (a) General description That brings us to the need for a brief description of Gallup’s pay system. It is undisputed that the telephone interviewers are not paid by the hour. As I shall explain in a moment, they do have an hourly pay rate that affects the pay that they receive, but it is not a salary or the typical wage rate that results, after so many hours worked, in a sum of money. As Rinehart testified, the pay system is very complicated. (11:1980) Sherri Lee describes it as a piece rate system. (6:1191–1192; 7:1276) Actually, the current pay system, chiefly devised by Supervisor Barlow (who is known as the “Father” of this pay plan, 11:2179), was designed to simplify the older plan (RX 4 at 2), and presumably did so. The new plan was installed as a pilot program in Houston in early May 1998, and went nationwide on June 1, 1998. (RX 4 at 2) (b) Two major components Although several, even many, factors affect pay, Gallup’s new (1998) pay system for interviewers has two main compo- nents. (The old system had the same two concepts, but just used different terminology.) First, productivity, and second, time (both terms mine). Under the first or productivity cate- gory (not to be confused with other terms that Gallup has for referring to production), there is no dispute that Gallup pays the telephone interviewers a specified amount per completed tele- phone survey—called “pay per completes,” or PPCs. (The amounts vary according to the nature of the study or project being done for a client.) Thus, should an interviewer be unlucky enough to call a survey participant who wants to “vent,”26 the interviewer must listen, as Janice Rinehart fre- quently had to do on the hospital “track.” (11:1997–1998) Obviously, even a completed survey in that situation takes longer than one which zips through with no venting. And should a participant decide to hang up after having vented his frustrations, then the time has been lost and the interviewer has made no money. (Rinehart occasionally was successful in re- dialing such a person and persuading him or her to finish the survey. 11:2048) Not all is lost by such a call, however, be- cause the time spent on the phone qualifies, as we shall see, as part of the interviewer’s “actual hours.” Focus now on the second, or “time,” category. As already indicated, interviewers are members of one of five teams that are defined in terms generally by the number of weekly hours [not just plain hours, but “actual hours”] that the members commit to work (such as 35 hours, 25 hours, 15 hours, and 10 hours), and also by hourly pay rates for each such team. Thus, at the time relevant here, the Premier and 35-hour teams were at the $11 per hour pay level, the 25-hour team at $9, the 15- hour team at $7, and the 10-hour team at $6. (1:45, 89, Uria- Ruiz; 11:2183, Barlow; RX 4 at 3; GCX at 1, last line in box) Premier Team members commit to 40 hours a week. (1:46, Uria-Ruiz; 13:2531, Geiger) That team has its own topical description in the ITM. (JX 23 at 35) Premier members have some additional perks, such as overtime when desired, and a 26 Janice Rinehart testified that Gallup trains the interviewers to let survey participants “vent.” (11:1997–1998) DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1256 program that provides a percentage of a child’s college tuition. Among the eligibility requirements are career evaluation scores averaging 48 or higher, 13 payroll months on the 35-hour team, and the unanimous approval for membership from all of the location’s managers. (JX 23 at 35) As Supervisor Geiger testi- fied, “typically” the Premier Team consists of “the top produc- ers of the locations.” (13:2531) In late May 1999, as Geiger recalls, the Premier Team had some 12 to 15 members. (13:2531–2532) Janice Rinehart was one of those members. (1:44–45, Uria-Ruiz; 13:2531, Geiger; GCX 16) (c) Pay per completes As interviewers are not paid an hourly wage rate or salary (1:46, Uria-Ruiz), the significance of the team base rate ($11, $9, and such) is that the PPCs assertedly are based on such pay rates. (1:47, 87, Uria-Ruiz; 12:2303, 2395, Barlow; RX 4 at 3—“All PPCs are based on these rates.”) This suggests that such team base rates are used as multipliers (not simply as an- other factor to be considered) in computing the relevant PPCs. As Supervisor Uria-Ruiz explained, members doing a specific survey will receive a higher PPC if they are on a higher level team. (1:47) An example of this is seen in the paysheets in- volved in this case, specifically the sheets of those working on Adventist project (“Startup Code”) W470 on June 16. [As seen on the paysheets, there also is an “Access Code,” that being 6942 for this same project, and that number is used a few times in the record. For this project there also is a project number, as shown on one of the documents. (RX 54; 13:2579, Geiger) Because the intcheck (GCX 54) pertaining to Janice Rinehart uses the startup code (W470 respecting the principal Adventist project here), and cross referencing is to the intcheck, I shall use the startup code, rather than one of the other code numbers, to identify a specific project.] For that June 16 and Adventist W470, Janice Rinehart, a member of the Premier Team, received a PPC of $2.20 (GCX 19 at 1); Lisa Miller, as a member of the 35-hour team, re- ceived the same (GCX at 3); Catherine Wagley, a member of the 25-hour team, received $1.80 (GCX 19 at 6); and 15-hour team member Joy Jones received $1.66 (GCX 19 at 4). The supervisors at each location set the PPCs (RX 4 at 4) based on their consideration of a variety of factors (11:2193–2194, 2196, Barlow), none of which, as set forth in the ITM (JX 23 at 21) or described by Supervisor Barlow (11:2194–2195), is the dollar figure ($11, $9, and such) of a team’s base rate. A moment earlier I said that the base rates of the teams are “assertedly” used as factors in computing the relevant PPCs. I say “assertedly” because, aside from the fact that the team lev- els produce, as shown, different PPC amounts, there is no inde- pendent verification in the record, such as a formula or detailed example, to show just how the $11 hourly rate, for example, rather than some other figure, is applied as a factor in arriving at the relevant PPC. Even if the supervisors do apply the stated dollar amounts in some fashion (even as multipliers), they are free to adjust other factors to reach a desired PPC. Thus, con- sider a comparison of the pay rates with the PPCs for Adventist W470. It is an 18.18-percent drop from $11 an hour to $9 per hour, and the 40-cent reduction in the corresponding PPCs ($2.20 to $1.80) is the same percentage drop. Not so when we compare the change from $9 to $7 (a 22.22-percent drop) with the 14 cent reduction in the corresponding PPCs of $1.80 to $1.66 (a reduction of 7.78 percent). Fortunately, this case does not turn on verification of the PPC formula, and I mention this gap in the record only because the parties appear to accept as gospel that the dollar factor as used (even if as a multiplier) in establishing a PPC is an actual mirror image of the dollar figure of each team’s base rate. (d) Alleged falsification What is involved here, however, is the allegation that Rinehart deliberately falsified the recorded time that she claimed in order, as Barlow believes (as noted above), to keep from being bumped to the lower pay of the 25-hour team. In- deed, most of us would consider a pay cut of 18.18 percent to be a severely adverse event. The correlative moral and eco- nomic importance of this to Gallup, Supervisor Barlow testified (12:2303), is that Gallup was cheated by being charged the higher rate associated with the Premier Team when in fact (ac- cording to Supervisor Geiger’s figures) Rinehart only worked enough hours (“actual hours” is the technical term, as we shall see) to qualify her for the 25-hour team’s 18.18 percent lower rate. Never mind that Rinehart, in obtaining 59 completes on June 16, for example, came within one completed survey of reaching 100 percent of the goal (quota as it is informally called27) of 5 completed surveys (9:1777; 11:2052, Rinehart) per hour (60 completes) for Adventist W470. What mattered was the asserted falsification of claimed “actual hours.” But wait. Did Gallup jump the gun? Return to the training manual (ITM) and the compensation requirements set forth there. Under the heading for “Gallup’s Compensation Policy,” the paragraphs there include the following information (JX 23 at 22): In order to meet your team’s requirement, you must satisfy the following criteria. Your pay will be based on your team’s pay scale if all requirements are met. At no time during your employment will you be paid below minimum wage. A pay period consists of two consecutive weeks. In order to be eligible for 35-hour team pay, you must have worked a minimum of 70 hours for the pay pe- riod, or 75 if it is an extra five-hour work week. . . . . Examples: . . . . 2. If you are on the 35-hour team and work 35 hours the first week and 30 hours the second week, you will have worked a total of 65 hours for the pay period. This does not satisfy the minimum 35-hour team requirement, but does satisfy the 25-hour team requirement. Therefore, you would receive the 25-hour team pay. 27 11:2180, Barlow. Previously it was officially called “quota.” (RX 4 at 2) Witnesses still speak in terms of surveys completed (“com- pletes” or “surveys”) per hour or per shift in reference to whether they have made the “quota” set for the project. GALLUP, INC. 1257 Notice that the Premier Team is not mentioned here, but is covered later under its own topical heading. (JX 23 at 35) Under that topic, aside from the mention that the Premier Team is for those who commit to 40 hours per week, there is no men- tion of the factors set forth above concerning, for example, the 35-hour team. At times in the record it appears that references to a person may be as a member of the Premier Team or the 35- hour team as if membership in those two teams is interchange- able. That is not so by definition in the ITM, but it does sug- gest that, in some respects, certain provisions apply to both. The most obvious example of this is that the pay rate of $11 is the same. And it seems unlikely that members of the Premier Team would be paid on a weekly basis while everyone else is paid every 2 weeks. I therefore find that the general concepts, set forth above respecting the 35-hour team, apply also to the Premier Team. This finding does not resolve what numbers Gallup would apply. It does mean, however, that a member of the Premier Team has 2 weeks to maintain her standing on that team. I so find. Just what number, whether 70 hours, 75 hours, or even 80 hours, is not established in the record. The point here, however, is that a member of the Premier Team would have the next consecutive week to work whatever overtime was necessary to make up any shortfall in the “actual hours” of the first week in order to maintain her standing on the Premier Team. (The parties do not address this point on brief.) What is the significance of the foregoing finding? Recall that, at least as described by Supervisor Barlow, Gallup had two concerns. One, Rinehart lied on her paysheets. Two, Gallup suffered an adverse economic consequence. The termi- nation document (JX 9), prepared by Supervisor Geiger (13:2680), states: Violation of Data Collection Agreement—falsified actual hours which led to inaccurate pay. Under the Data Collection Agreement (not an employment contract, but at least a written understanding that the interview- ers sign when hired) any falsification of data on the “paysheet” will result in termination and the offending employee will be “bumped to minimum wage for the pay period.” (JX 23 at 62) In accordance with my finding above, I further find that Janice Rinehart had the usual 2 weeks to work and make up any shortfall in her hours so as to maintain her standing on the Pre- mier Team. Even if she were to fall to the 35-hour team, the pay rate would still be $11 per hour. I therefore also find that, contrary to Supervisor Barlow’s assertion, Gallup did not suffer an immediate economic loss by Rinehart’s asserted misconduct because Rinehart had an additional week to make up any short- fall in actual hours. Indeed, as Sherri Lee credibly describes, sometimes supervisors even excuse a shortfall, or told the em- ployee to make it up the following week—a “rollover.” (6:1241–1242; 7:1279–1281, 1310, 1312) Although the con- cept was not litigated, a rollover possibly is an informal proce- dure that sometimes would be utilized in place of the formal procedure of “overrides.” Overrides are mentioned in the ITM (JX 23 at 23), in the “Payroll Contract” (GCX 34 at internal 67), and by Sherri Lee (7:1281). Also, as we see when we get to Patrick Snyder’s case, Supervisor Geiger describes the situa- tions when work gets slow and supervisors can grant “exemp- tions” for a specified time from the weekly hours requirement. (13:2688) The termination, however, was based on Supervisor Geiger’s calculations of (and interpretation of) the numbers for the 4-day period of June 14 through June 17, 1999. (13:2676, Geiger) That leaves the one ground of lying on the paysheets (about a matter that eventually would affect pay). That, of course, is a serious and legitimate ground by itself, and the Data Collection Agreement so advises those who sign (and Rinehart presumably signed one). (e) Interviewing costs? I do not overlook Barlow’s testimony that Gallup bills its customers based on “our hours that are put to the projects,” so Gallup therefore needs accurate records to support its billings. (11:2189; 12:2279) In that connection, respecting how Geiger came to check Rinehart’s hours, both Geiger (13:2547) and Barlow (12:2278–2279) tell us that it had to do with investigat- ing the concept of Gallup’s “interviewing costs” on the Advent- ist project. It seems that Tracy Stinson, the project administra- tor (in the Nebraska office, 13:2547), called Geiger and ex- pressed the desire to see Gallup’s profit margin increased on the Adventist project. Thus, Stinson asked Geiger to see if the quota on the project could be set higher and the hours needed to complete the project lowered. As to the latter, Stinson report- edly thought that the project was taking too many interviewer hours for the number of completes being achieved. (13:2547– 2548) Stated differently, Stinson reportedly thought that the ratio of completes per interviewer hours was too low. Follow- ing Geiger’s review of the situation, the quota for the Adventist track was not raised. (14:2869) It seems strange that the supervisors, especially Barlow (12:2278–2279), would be pointing to “interviewing costs” and suggesting (11:2189) that Gallup’s clients are billed (per a “set agreement,” 12:2279) on the basis of the “hours” worked on the projects. Gallup has a variety of incentive pay bonuses that are mentioned in the record (RX 4 at 3–5; JX 23 at 34), with most apparently related directly or indirectly to the attained PPCs. Gallup certainly would have many overhead costs. And Gallup has a “budget” for each project (12:2278). Notwithstanding all this, what we have learned is that Gallup pays its interviewers for completed surveys. For each project that Gallup seeks to do for a customer, one logically would infer that Gallup submits a bid price based on “x” plus “y” and “z,” where “x” is the an- ticipated expense of the completed surveys, “y” is the expense of Gallup’s various overhead items, and “z” is the profit figure that Gallup would include in its bid price. I return to the sub- ject of Tracy Stinson’s call a bit later. (f) Actual hours The concept of “actual hours” is near the center of Rinehart’s case. The new (May 1998) pay plan’s first page (as received in this proceeding, that means RX 4 at 1) is devoted to “Actual Hours.” The contents of that first page bear directly on the issues here. Indeed, Rinehart reports that she kept a copy of that first page push-pinned to the wall of her cube. (10:1954) Now quoting that first page, I do not duplicate the larger print or bolding of the heading or the bolding of the time and deci- mals in the two columns (RX 4 at 1): DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1258 Actual Hours Recording actual hours accurately is extremely important un- der the new pay system. The new pay sheets and payroll en- try process allow you to report your actual hours to the nearest tenth of an hour. Previously you had to round up to the near- est quarter which would either rob you of actual hours or cost you on your capacity. Please keep track closely how much time you are actually spending on the phone. If you com- pletely lose track, have a manager run an intcheck to find out your actual hours correctly. From now on, please report your actual hours according to the following scale. What to report on Time Worked Pay Sheet 1 Hour 1 1 Hour 6 minutes 1.1 1 Hour 12 minutes 1.2 1 Hour 18 minutes 1.3 1 Hour 24 minutes 1.4 1 Hour 30 minutes 1.5 1 Hour 36 minutes 1.6 1 Hour 42 minutes 1.7 1 Hour 48 minutes 1.8 1 Hour 54 minutes 1.9 2 Hours 2 ***Round to the nearest decimal if necessary. To help your capacity, if you work one hour and fifteen minutes then you will report 1.2. If you work one hour and forty- five minutes you will report 1.7. Questions? Please come and see a manager. Under Gallup’s pay system, “actual hours” is more restric- tive than a concept of “work hours.” The latter could include many job-related tasks, and even include short mid-morning or mid-afternoon rest periods or smoke breaks, and surely would include restroom breaks. No so under Gallup’s actual hours, for Gallup’s written description in the January 1999 ITM, deal- ing with the “Nightly Paysheet,” reads in part (JX 23 at 24): Actual Hours: This number is critical in determining how long you actually worked, in order to ensure proper PPC rates. “Actual hours” is the amount of time you actually worked on a project, not counting smoke breaks or any other type of break. [Nearest 6 minutes and decimals explained.] Let’s look at an example: You begin working at 4:00 pm. You take a break at 5:00 pm to make a phone call and have a snack. You begin working again at 5:18 pm and work until 8:00 pm. Your “actual hours” worked would be 3 hours 42 minutes. This would be recorded as 3.7 hours. Likewise, the old “Quota Sheet Review” (RX 45), replaced by the “Paysheet” under the 1998 pay system (RX 4 at 2; 12:2291, Barlow), instructed that “Actual Hours” meant “Time spent on the phone, not including breaks.” (RX 45; 12:2291) The managers’ thumbnail test for actual hours is this: It is lim- ited to “dialing on the phone.” 1:82-83; 15:3071, Uria-Ruiz; 11:2190–2192; 12:2304–2305, Barlow) As Uria-Ruiz explains, when the January 1999 ITM issued, copies were given to new interviewers but not distributed to the current interviewers be- cause the latter already had received copies of the changes, such as the new pay plan (RX 4), that were incorporated into the January 1999 ITM. (15:3070, 3121–3122, 3130) This does not mean that Gallup’s interviewers must take rest- room breaks and smoke breaks entirely on their own time. During the hours that interviewers are logged on to the system, Gallup’s unwritten policy allows interviewers to add .1 hour (6 minutes) per each hour logged in order to cover a variety of nondialing activities such as restroom and smoke breaks, seek- ing out a supervisor for work assistance, any occasional paper- work associated with a survey, the 2 minutes or so needed to switch projects, for “quota fills” (a major topic discussed later), or for similar matters. (1:42–43, 75, 83–84, Uria-Ruiz; 11:2192, 2196, 2199; 12:2283, 2339, 2342, Barlow; 13:2535– 2544, Geiger) As Geiger testified, “Everything that’s not dial- ing goes into the .1.” (14:2776, 2867) That, of course, assigns numerous work-related tasks to the .1 (6 minutes) per hour grace allowance. There is a dispute concerning whether the break policy, or grace allowance, is really .1 per hour (6 minutes), as Gallup’s witnesses assert (Gallup, Br. at 56; Reply at 20), or 10 minutes per hour as the Government’s witnesses claim (General Coun- sel, Br. at 42–43). I shall not dwell on this dispute because it does not appear to have much relevance to Rinehart’s case in view of the large time discrepancy that Gallup attributes to her. Supervisor Geiger testified that she does not know how the concept of a 10-minute break allowance got started. (13:2544; 14:2776) The Government offered testimonial evidence (Janice Rinehart, Sherri Lee, Glen Lambert, and Beverly Robinson) that supervisors Geiger and Heidi Roberts, plus agent (payroll coordinator) Mike Cota, had told employees that the break allowance was 10 minutes per hour. Roberts gives a soft denial for this, and evidences some difficulty with keeping the deci- mals and minutes straight. (15:2960–2961) Geiger denies by describing a specific event (13:2531–2532), and Cota did not testify. Although I generally credit these Government wit- nesses, I resolve the issue in this abbreviated fashion. First, I suspect that the source of the dispute is a misunderstanding about fractions and decimals (somewhat akin to Supervisor Roberts’ problem indicated just above). However, even with the Government’s evidence, it is not clear that Gallup had fol- lowed a well-established practice of 10-minute breaks. Along comes the 1998 pay plan that deals in fractions using reporting increments of .1 hour (6 minutes). It is quite possible that some employees, and even a supervisor or agent, mentally converted the one-tenth of an hour to one-tenth of a hundred, or 10 min- utes. Or the mental mistake could have arisen from the same proc- ess that confused Nancy Sico, a “buddy,” who reports that she tells the new employees (quite likely just a momentary lapse at trial) to record their time to the “nearest sixth of an hour,” or “six minutes.” (3:483–484) That is, perhaps some employees correctly converted the wrong standard of a “sixth of an hour” to 10 minutes. Whether either of these is the origin of the prob- lem or not (and I make no finding in this regard), with one ex- ception I could find (were I to reach that portion of the evi- GALLUP, INC. 1259 dence) that Supervisor Roberts and agent Mike Cota had told some employees that the break allowance was 10 minutes. The exception is Supervisor Geiger, and, in this respect, I find that Janice Rinehart (9:1767–1768) was mistaken. In deal- ing with numbers at trial, Geiger was impressive with her men- tal grasp and usual precision. (13:2636; 14:2753, for example.) At trial Geiger was known as a “math whiz.” (13:2616) All this should be no surprise given that in college Geiger took a 4- year minor in mathematics. (13:2636) This is not to suggest that her arithmetic is always correct, for she did make a few errors (for example, !3:2617, 2649, and RX 53 at June 17). But then, no one is perfect. Although not mentioned in the briefs as to this point, Glen Lambert credibly describes an incident, estimated to have oc- curred about mid-August 1999, when he overheard Supervisor Charles Davenport correct an interviewer, who quoted Supervi- sor Heidi Roberts as having said that breaks were 10 minutes, by telling her that Roberts was wrong and that the allowed break period per hour was 6 or 7 minutes. (3:554, 559) Note that the interviewer doing the quoting did not say when Roberts had said this. It seems likely that Roberts would have said this before the discharges of Janice Rinehart and Sherri Lee, when the hourly break times figured in the terminations. That is, presumably the local management reminded its members after those discharges that the pay system provides for break periods of 6 minutes per hour, expressed as .1 hour, and not 10 minutes per hour which would have to be rounded to a .2 hour, or 12 minutes. In any event, I credit Supervisor Geiger that, at a meeting with the Premier Team about late May 1999 [Geiger also met with the hospital track about the same time, 13:2668], when Janice Rinehart asked whether hourly break periods were 10 minutes, Geiger said that a period of 10 minutes was not used, that the time was .1 hour. (13:2531–2532) Finally, the .1 hour leeway of 6 minutes has to be “sandwiched” within the time that the employee is on dialing time. An interviewer may not add 6 minutes per hour to her actual hours if she never took the breaks. (13:2536–2544; 14:2777, 2783, 2903, Geiger) Thus, because breaks have to be taken or not claimed (13:2538), the break policy is referred to as a “use it or lose it” approach. (11:2109, Rinehart; 13:2543, Geiger) As we already have seen, the singular importance of “actual hours” is that, by determining the hourly team the interviewer is a member of, it becomes the second major factor (of two) that generates an interviewer’s actual pay. Because different teams earn at different rates, their PPCs also differ. Summarizing again, as one of the documents in evidence reflects, a PPC on the Adventist Project W470 in June 1999 earned $2.20 if the interviewer was on the 35-hour (or Premier) team (GCX 19 at 1, Janice Rinehart); $1.80 for member of the 25-hour team (GCX 19 at 6, Catherine Wagley), and $1.66 for those on the 15-hour team (GCX 19 at 4, Joy Jones). Using those numbers, we see that, assuming the same numbers elsewhere, an inter- viewer earning at the rate of $2.20 per PPC would take an 18.18 percent pay cut if he or she were bumped to the 25-hour team. (g) Quotas, trackers, and credibility Turn back now to the concept of quotas. Under the new (May 1998) pay plan (RX 4 at 2), the “Quota rate” is the new PPC. From that one would think that the old “quota” would now be the “completes” expected (at least informally) per “ac- tual hour” of work. And informally that is so. (11:2180, Bar- low) However, the pay plan (RX 4 at 2) provides that the “Ac- tual quota” (ostensibly the old “quota”) now has the name, “production hours.” That term is defined, generally, as “A measurement of the actual amount of work you produce. The total dollar amount earned on a project divided by Team Base = Production Hours.” (RX 4 at 3) By that definition (and by an example accompanying the definition), we see that the defini- tion of the term is a formula that, in effect, states the number of “actual hours” as a number that the number of completes bears to the number of the expected completes (the old quotas) per hour for that study. As an example of this, consider the date of Wednesday, June 16, 1999. For that date Janice Rinehart submitted a nightly paysheet reflecting that she had worked 12 “actual hours” on Adventist W470, with 59 completes at $2.20 for pay of $129.80 and 11.8 production hours. (GCX 16 at 3) With the expected completes (quotas) being five surveys per hour for that project (1:90–91, Uria-Ruiz; 11:2052, Rinehart; 15:3115, Uria-Ruiz), the arithmetic shows that Rinehart needed a total of 60 com- pletes to be at full (100 percent) “quota,” to use the old termi- nology (and as the witnesses still do28). To see the relationship in the ratios of the completes and the hours, let us divide 59 by 60. That yields the number 98.33333. Multiplying that number by the actual hours claimed of 12, and we reach the 11.8 figure for production hours as shown on the document. The relevance here of production (particularly as to the num- ber of completes reached) is that it has some bearing on credi- bility. As already seen, Rinehart’s discharge is based on the allegation that she falsified her hours. But Rinehart asserts that, before the Union, hours were never an issue, and that it was production (making quota) that Gallup had always stressed. (11:1968, 1982) Doubtlessly that was Rinehart’s perception, but even the General Counsel does not go that far and, instead, asserts (Brief at 44) that Gallup placed the “heaviest emphasis” on “production hours” or making quota. Certainly there is much support in the record for the view that Gallup placed heavy emphasis on reaching and exceeding quota (getting as many PPCs as possible). This corporate atti- tude is reflected in Supervisor Geiger’s remarks about a meet- ing that she held, about late May 1999, with, as she vaguely recalls, her Premier Team members. In Geiger’s words, the Premier Team “typically” consists of “the top producers of the location.” (13:2531) [As Geiger held a meeting with her hos- pital “trackers,” including Janice Rinehart, about that same time (13:2668), it is possible that the late May meeting Geiger de- scribes was with the hospital trackers. The point, however, is Geiger’s description of the excellence of the members of the Premier Team.] Notice, however, that in the same breath as her phrase about the “top producers,” Geiger also defines the Pre- 28 For example, Supervisor Uria-Ruiz (1:114; 15:3115–3117); Su- pervisor Geiger (13:2546, 2548; 14:2870); and the ITM (JX 23 at 27). DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1260 mier Team as “a select group of interviewers who work at least 40 hours a week . . .,” thereby placing equal emphasis on the (actual) hours worked as on the amount of production. (13:2531) “Trackers” refer to interviewers who, as Supervisor Barlow describes, also are “top interviewers.” (11:2182) As suggested by the term, “trackers” are interviewers who work of a client’s project that “tracks” opinion, for example, over a period of time, as contrasted with a project that is short term. (1:76–77, Uria-Ruiz; 2:378–379, Trovato; 11:2182, Barlow; JX 23 at 24, third explanatory box at top.) Under most any definition applied by Gallup (whether Pre- mier Team, trackers, or something else), Janice Rinehart was a “top producer.” Hark back to the General Counsel’s argument that quality (the high scores on Rinehart’s evaluations) is rele- vant notwithstanding that Rinehart was fired for falsifying her hours because the work ethic needed to achieve such excellence is inconsistent with the execution of a plan to cheat Rinehart’s employer. The same theory seemingly would apply to Rinehart’s work ethic in maintaining a standard of excellence for production (for making the quota of completed surveys per hour). The relevance may not be direct and compelling, but it does appear to have at least some significance. c. Quota fills Within the term “quota fills,” the first element, the word “quota,” refers to a “targeted sample” of responses that are to be harvested from, normally, a given section (could be geo- graphical, such as by telephone number prefix, or by type of survey participant, or some other targeted group). Once the number of completes (completed surveys) set as the sample for a particular section of the overall study has been obtained, the “quota” or sample for that section has been reached and there is no point in making further calls to that section or area of the project. This status of sample reached, or “filled,” is called “quota filled.” The computers are programmed to announce “Quota filled” and to keep searching for a good number to dial. (1:78–81, Uria-Ruiz; 6:1010–1011, Zieler; 10:1818, 1824– 1826, Selix; 11:2197–2198, Barlow; 13:2601; 14:2897, Geiger) When the “Quota filled” notices begin appearing repeatedly, interviewers need to ask a supervisor to “hide” the numbers (in the computer system) so that the computer will disregard them and the interviewers can then begin on a new section whose sample is open for dialing. (1:81–82, Uria-Ruiz; 10:1826, Selix; 11:2198–2201, Barlow) Unfortunately, a supervisor is not always handy so that very little time will be lost. Indeed, an interviewer can lose 15 to 30 minutes seeking out a supervisor and getting the quota-filled section “hidden” so that the inter- viewer (or her computer) can make calls in a good section. (10:1921–1928; 11:2047, 2107–2109, Rinehart) Although John Selix (the “Father of the Intcheck”) suggests that the inter- viewer loses no time because numbers pass in an electronic “flash” (10:1825–1826), interviewers such as Lynne Zieler (6:1010) and Janice Rinehart (10:1920–1921 for example, de- scribe how the “quota-filled” notices do not simply “flash by,” but stay on the screen 30 to 45 seconds or so while they write down the identifying data to give to a supervisor when asking him or her to hid the numbers. Thus, Zieler would compile lists reaching about 80 numbers or more to be hidden, 6:1010– 1012). To the extent there is some flashing, the interviewers still get a bunch that required time to process. (11:2010–2011, Rinehart) Management witnesses assert that any time an interviewer spends watching the computer screen waiting for a good num- ber as, for example, “quota-filled” announcements scroll by, or who write down the numbers that are listed on the screen, or who take such numbers to a Supervisor to be hidden, may not include such time in their paysheet column for “actual hours.” (11:2199; 12:2283, 2339, Barlow) And according to Supervi- sor Barlow, whenever an interviewer experiences “quota-filled” notices, he or she needs to maintain a log reflecting the time spent logging the numbers and, presumably, looking for a Su- pervisor to hide the numbers, and to subtract the total of such time from the total for actual hours worked that shift. (12:2340–2342) Geiger testified that short times spent watch- ing quota fills, such as for 2 to 3 minutes, are not to be de- ducted because such gaps are covered under the general .1 hour grace allowance per hour. (14:2899–2900) Time lost in excess of the grace allowance of .1 hour per hour would need to be deducted from actual hours. (14:2900, Geiger) d. Janice Rinehart’s recording practice From the time that she began working for Gallup as a tele- phone interviewer (in September 1996), Rinehart testified (10:1915, 1955; 11:1981), she recorded her time the same way. In this respect, she counted as working time (that is, as “actual hours”) all restroom breaks, smoke breaks, and all other short breaks of 10 to 15 minutes or so, including such short trips to the downstairs deli to get a sandwich to eat at her cubicle, and all time she was at her computer working for Gallup perform- ing such work as waiting for quota-filled numbers to pass and for good numbers to “drop” onto the computer screen, any time contacting or looking for a supervisor to hide the quota-filled numbers, any time spent editing an occasional survey form, and any other similar work performed for Gallup. She did not count, as time worked, her 30-minute or so lunch breaks down- stairs, or any similar or longer breaks for shopping trips to nearby stores. (10:1916–1918; 11:1969, 1981–1983, 2002– 2004, 2075–2076, 2079, 2106–2107) Rinehart assertedly fig- ured her time honestly and to the best of her ability. (11:1975, 1982) Before her discharge, Rinehart was never told that she was recording her actual hours erroneously (11:1974), nor was she ever told that actual hours did not properly include time devoted to quota fills (10:1919, 1930; 11:1969). Notice that Gallup’s pay-plan description of “Actual Hours” does not mention quota fills (RX 4 at 1), although it does pro- vide, as Supervisor Geiger observes (13:2667), that interview- ers are to track closely the time that they “are actually spending on the phone.” Rinehart testified that she did not understand this to mean that time on the telephone was the only item of time worked to be included under the term “actual hours.” (10:1953–1954) The pay plan (RX 4 at 1) instructs interview- ers to record their “actual hours” to the nearest tenth of an hour, as Supervisor Geiger also observes. (13:2530) Recall that Rinehart (10:1954) had push-pinned a copy of that that docu- ment (page 1 or RX 4) to the wall of her cube. However, to the GALLUP, INC. 1261 extent that there was any question that actual hours did not include breaks of all kinds, that issue was resolved by the re- vised ITM that issued in January 1999 with the language, quoted earlier, that time spent on smoke breaks or any other kind of break is not to be recorded in the paysheet column for “actual hours.” (JX 23 at 24) Although, as Supervisor Uria- Ruiz testified, and as described earlier, copies of the revised manuals were not distributed to current employees (such as Rinehart), current employees had already received the changes as passouts. Indeed, recall that Rinehart testified that is the manner in which she received a copy of the pay plan. (10:1954) If she did not read such passout, she may have hurt herself by that failure. Clearly she would be charged with knowledge of the passout’s contents. Aside from breaks, what about items of actual work, such as looking for good numbers to drop onto the computer screen, or seeking out a supervisor to have him or her “hide” the quota- filled numbers. Neither the pay plan (RX 4) nor the ITM (JX 23) ever expressly informs employees that such work activities, and any other work activities not part of dialing on the tele- phone, are not properly part of “actual hours.” Moreover, no memo was ever distributed to employees, or posted, so notify- ing them. (12:2304–2305, Barlow) Although Supervisor Bar- low told new employees, when he trained them back in 1992 and 1993, that actual hours meant only when they were dialing on the telephone, he has never told anyone that quota fills are not part of actual hours. (12:2283–2284, 2288, 2292) As we have seen, interviewers can lose valuable time while waiting for quota fill notices to pass and for the computer to find a good number to dial. Should reasonable interviewers interpret the pay plan’s phrase about closely tracking their time “on the phone” as including, or excluding, such work-related matters as handling quota-filled numbers? Is the phrase am- biguous? As for the plan’s caution for interviewers to have an intcheck run if they lose track of their time (RX 4 at 1), Rinehart testified that she never lost track of her time. (10:1954–1955) As for any familiarity with an intcheck, Rinehart testified that the first time that she ever saw one was during pretrial with the General Counsel. (9:1744; 10:1918) Supervisor Geiger’s response to all this testimony about gaps in working on the phone to take restroom or smoke breaks or breaks to look for a supervisor to hide quota-filled numbers is that the .1 hour grace allowance per hour provides coverage for such matters. (14:2776, 2867) If searching for a supervisor takes more than the allotted 6 minutes grace per dialing hour, then the interviewer should subtract the extra time from her actual hours. (14:2900) Actually, if quota fills are consuming more than 2 or 3 minutes, then the interviewer should switch to a different project. (14:2868, 2897–2898, Geiger) Geiger gave no details or examples on the practicality of the switching that she refers to. The more practical-sounding testimony came from the witnesses, such as Lynne Zieler and Janice Rinehart who describe, as reported earlier, the problem with quota fills as being a common problem. I credit their testimony, for they testified persuasively. With it being such a common problem, Geiger’s testimony on this point had all the ring of a bureau- cratic response by someone who never really got involved with trying to solve the problem. For example, Geiger did not tell us how practical it is to switch (and how often), or how much time is required to switch, and whether it is to some project that would not be a hospital track. Geiger did not testify persua- sively in this area. Supervisor Uria-Ruiz suggests that it would take an inter- viewer only a minute or two to switch projects, including walk- ing to a start-up computer to enter the required data and return- ing to her cubicle. (1:43, 74–75, 83–84; 15:3076–3077) Uria- Ruiz’s time estimate is unrealistic. The more persuasive ver- sion is given by Rinehart that it takes 4 to 6 minutes to make a switch. (11:2100–2102) With quota fills being a common problem, especially as projects were in the last week or so of their run, the time spent trying to solve their problems, when deducted from actual hours, would seem to consume much of the hourly grace period. e. Gallup’s 5-hour grace allowance Recall that interviewers are given a grace allowance of .1 hour per “actual hour” worked per week to cover breaks and other time away from dialing. For someone on the 25-hour team, therefore, he or she would have, for the 25 hours, a total grace allowance of 2.5 hours. Except for one fact, applying the same multiplier to the 35-hour team would result in a weekly grace allowance of 3.5 hours (35 hours x .1 hour = 3.5 hours). The fact exception is that the grace allowance is different for members of the top teams. Thus, one perk enjoyed by members of the 35-hour and Pre- mier teams is an enhanced grace allowance of 5 hours per week. (1:44, Uria-Ruiz; 13:2541; 14:2824–2825, 2864–2865, Geiger) This enhanced standard of 5 hours results in an extra 1.5 hours of grace over the 3.5 hours that the these top teams would otherwise receive. As to the members of these top teams, discipline respecting the overreporting of hours is ap- plied by the following standard. If their reported actual hours exceed the correct number (as shown by an intcheck) by no more than 5 hours, they receive only a warning. If it exceeds the correct amount by more than 5 hours, they are terminated, Supervisor Geiger asserts. (13:2534–2535; 14:2846, 2865– 2866) In a pretrial affidavit given to the investigating Board agent, Supervisor Uria-Ruiz states that, even if the discrepancy is less than 5 hours, discharge is still an option. (1:44) In the latter instance, presumably the discharge option is exercised when Gallup determines that the employee falsified her time records, and not exercised simply when an employee enters incorrect times based on an unintended mistake. [Uria-Ruiz’s affidavit report, normally inadmissible hearsay unless offered for admis- sions or impeachment, became admissible evidence here be- cause Uria-Ruiz’s confirmation at trial was elicited by the Gen- eral Counsel who had called Uria-Ruiz as an adverse witness.] As Janice Rinehart was a member of the Premier Team (or the 35-hour team as is reported at a point or two in the record), the applicable grace standard as to her is that of 5 hours per week. At trial the General Counsel argued that Gallup’s “actual hours” necessarily had to include any time working to resolve quota fills. “That is working time. That is part of actual hours.” (10:1868, 1875; 11:2093) The theory of this argument is that quota fills, logically work being done for Gallup, are DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1262 included in the term actual hours because Gallup never specifi- cally told its interviewers that actual hours did not include quota fills. (10:1880) On brief, the General Counsel has aban- doned this as an express argument. Turn now to the event at issue. f. Supervisor Geiger’s investigation (1) Tracy Stinson calls Supervisor Geiger tells us that the whole Janice Rinehart epi- sode began when Tracy Stinson, the project administrator (in the Nebraska office, 13:2547) for the Adventist project, appar- ently on Wednesday, June 23 (13:2546), called Geiger and expressed the desire to see Gallup’s profit margin increased on the Adventist project.29 Thus, Stinson asked Geiger (who set the quota on the Adventist project, 13:2546, Geiger) to see if the quota on the project could be set higher and the hours needed to complete the project lowered. As to the latter, Stinson thought that the project was taking too many inter- viewer hours for the number of completes being achieved. (13:2545–2548; 14:2869, Geiger) Stated differently, Stinson thought that the ratio of completes per interviewer hours was too low. Following Geiger’s review of the situation, the quota for the Adventist track was not raised. (14:2869, Geiger) Strange. The project ended the day before, on the morning of June 22. (11:1989, Rinehart; GCX 16 at 6) What would a reasonably prudent manager do in Tracy Stinson’s position as the project administrator for Adventist W470? No doubt she would make periodic checks on the status of the project. And if at one of several stages, such as that of 20 percent complete, 40 percent, or possibly even at 60 percent, she discovered that the project was not making the desired profit, then she could call Supervisor Geiger and ask her to do a study. But the day after the project had ended? Well, yes, if the purpose was to obtain the result of the study to use for negotiating the next contract. But that purpose is entirely different from the asserted purpose that Tracy Stinson reportedly expressed. Is it reasonable to think that our reasonably prudent project administrator would be calling for a study to squeeze out more profit on Adventist W470 when the hearse was enroute to pick up the body of this project and carry it to the morgue? Was there another reason, a reason not given at trial, for Tracy Stinson’s call? Stinson did not testify, and of course the report of her otherwise hearsay call was offered merely to show why Supervisor Geiger initiated her study. Even so, I am not bound to accept as gospel that this stated purpose of the call (with its strange timing) was what really motivated Gallup in calling Geiger and asking her to do this study. In this connection, recall from the earlier discussion covering the coercion allegations that several members of the Union’s organizing committee had been involved with distributing un- ion flyers and some had met with Supervisor Uria-Ruiz on June 4. (Janice Rinehart was one of those who attended the June 4 meeting, but, as earlier noted, she left early.) It is probable that higher management at Gallup began to question whether any of 29 The specific project is the one named on RX 52—“Adventist IP.” (13:2545–2548; 14:2869–2870) It also is identified by the startup, or study, code number, W470. (GCX 16 at 1; GCX 54 at 1) this union activity was occurring during times that the union supporters were claiming credit for actual hours worked. So, did Gallup have Project Administrator Tracy Stinson call Hous- ton to launch such a study, under the guise of needing to in- crease profit, when the real purpose was to see whether there was an ostensible justification for eliminating any union sup- porter who had failed to exclude from actual hours the time spent on union activity? The record does not have any accounting for Janice Rinehart’s time spent on union activities during working hours the week of June 14 (the relevant week, as we see in a mo- ment). Even so, if the record shows, including by supportable inference, that Stinson’s call was so motivated, it is not neces- sary to show whether Rinehart in fact, the week of June 14, was active for the Union while on the premises, or that at least a significant amount of that union time was claimed as actual hours on her nightly payslip. I shall return to this point later. (2) Geiger’s search and review Following Stinson’s call, Supervisor Geiger began her study. She first pulled up, and printed, the weekly report (RX 52) for the week beginning Monday, June 14, the week previous to the current one. (13:2545–2548; 14:2870) This weekly report covers the nine trackers working “Adventist IP,” or Adventist W470. The document shows, among other items, the number of completes and the number of actual hours being claimed for each day of the week. (13:2548; RX 52) The “red flag” in this weekly report that caught Geiger’s eye was Janice Rinehart’s claim of 12.0 actual hours on Wednes- day, June 16. (13:2551, 2569, 2590; 14:2854, Geiger) Geiger testified that the survey was not even “up” for 12 hours in that it does not come “up” until 9 a.m. “and we stop calling at 8:00 p.m. our time; that’s 11 hours.” Thus, 12 actual hours would have been “pretty unlikely.” Moreover, and as already noted, the project was about to finish and there were not many good numbers to dial. Finally, the 9 a.m. to 8 p.m. calling hours are programmed into the computer. (13:2551–2553, Geiger) As to this latter point, apparently nothing about the computer pro- gram would prevent an interviewer from manually dialing around any computer ending time. Of the nine Adventist “trackers,” only Rinehart was shown on the weekly report (RX 52) as claiming (as to any day) more “actual hours” that Ad- ventist W470 was “up.” (13:2560, Geiger; RX 52) On cross-examination, Geiger admitted that she had stated, in per pretrial affidavit, that the closing time was “about 7 p.m.” (14:2852–2854) The General Counsel argues (Br. at 44) that this 1-hour discrepancy, plus the fact that the relevant intcheck (which I reach shortly) discloses that Rinehart made calls as late as 8:37 p.m., undercuts Geiger’s testimony about a need to investigate respecting Rinehart’s 12 hours. Not so says Geiger because the trackers know that they are not supposed to call survey participants beyond 9 p.m., or 9:30 p.m. (Geiger gives both times, 14:2854–2856), their time, and the survey participants here were on the east coast, meaning that Eastern Time Zone. Thus, to the extent that Janice Rinehart called past either 8 p.m. or 8:30 p.m., Central time, then she violated Gallup policy. 14:2854–2856) In light of the 11 hour “up” GALLUP, INC. 1263 time for the project, the “red flag” for Geiger as to Rinehart’s time was the claim of 12 “actual hours.” (14:2754) I find nothing impeaching in this bit of possible discrepancy in Geiger’s pretrial affidavit. “About” 7 p.m. is a generalized time frame—a concept the Government uses for “about” dates in every complaint issued by the General Counsel across the country. The real point, as made by Geiger at trial, is that the “red flag” that caught Geiger’s eye was the claim of 12 hours, when the project is up for only 11 hours (even if Rinehart did make some manual dials against Gallup policy). (14:2854) Moreover, I find nothing here detracting from Geiger’s credi- bility or tending to show that the stated motivation for the study was false. In light of the substantial difference between Rinehart’s asserted understanding of what it was appropriate for her to claim as actual hours worked, and the opposing evi- dence showing that none of the breaks that Rinehart claimed, nor apparently any of the time worked on quota fills, would qualify as actual hours, it is not surprising that Geiger would find a substantial discrepancy in the actual hours claimed by Janice Rinehart. Having been attracted by the red flag as to Rinehart’s claimed hours on June 16, Geiger then proceeded to review Rinehart’s paysheets (GCX 16) for the week.30 [Nothing on the other days that week, standing alone, would have prompted Geiger to have taken a closer look. (13:2550–2554, 2569, Gei- ger)] From there she pulled up, or printed, an intcheck (GCX 54) of Rinehart’s calls for the week (4 days) of June 14. From these documents, supervisor Geiger then prepared, on a single sheet of paper (RX 53), her handwritten notes for Rinehart’s 4 days that week. Geiger also included notes as to Monday, June 21 as covered by the intcheck, but that date did not figure into the termination decision. Only the calls for the 4 days were used for the termination. (13:2676, Geiger) Listed in five col- umns, Geiger’s hand notes show the day date, project, claimed hours, and actual hours—as based on Geiger’s calculations (13:2555, 2560, 2588, 2590) from the numbers shown by the intcheck. Supervisor Geiger’s calculations, the results of which are reflected on RX 53, are what served as the basis for Rinehart’s termination. (13:2676, Geiger) Supervisor Barlow simply relied on the calculations performed by Geiger when speaking to Rinehart at the termination meeting. (12:2278, 2280–2281, 2377–2378, Barlow) (3) Preliminary terms and considerations To a substantial extent, any review of the line items on the intcheck calls for an interpretation based on the knowledge and experience of what is involved. Thus, some of the analysis and interpretation is objective, and some is subjective. The intcheck involved here (GCX 54) has 26 pages, but pages 1 through most of page 19 are the ones carrying through the first 4 days of the week. In round numbers, the pages have about 50 line items each, or a rough total of about 950 line items for the 4-day week involved here. In her investigation, Supervisor 30 The parties agree that GCX 56 inadvertently duplicates GCX 16, and that it would be preferable to cite GCX 16. (14:2872–2873) Al- though GCX 19 at 1 duplicates GCX 16 at 3, that is not inadvertent because GCX 19, with its pages for other interviewers, has a different purpose. Geiger focused only on obviously (to her) questionable items, such as long periods for an answer machine, and for long stretches of time, usually called gaps, in the transition from one line number to the next (that is, in the transition from one call to the next) in arriving at her conclusions. As addressed at trial, 39 of these line items figured either in Supervisor Geiger’s calculations, in her testimony, or the testimony of Janice Rinehart. (Many of the other 900 or so line items provide some comparison figures for the 39 line items.) Because Geiger was stricter in her analysis at trial than she was in June 1999 (13:2604, 2612, 2625–2626, 2640, 2676, Geiger; 15:3024, Uria-Ruiz), the 39 includes more line items than Geiger disal- lowed originally. These “adjustments” (the term used at trial), or disallowances, are subtractions (13:2595) that Geiger made as she analyzed Janice Rinehart’s 950 or so line items in the intcheck (GCX 54) for Rinehart’s 4-day week of Monday through Thursday, June 14 through 17, 1999. After observing that there were a lot of gaps in the intcheck (GCX 54), and not wanting to spend the time required to check every questionable gap, she focused on the “major” gaps, mainly those of 10 minutes or more. Secondly, Geiger took this lenient approach because she did not want to have to termi- nate Rinehart. First, Rinehart was a “very productive inter- viewer.” Second, Geiger liked Rinehart personally. (13:2626– 2627, Geiger) At trial Geiger listed 32 line items for which she made, or probably made, deductions. As to the other 7, she asserts that she could have but did not, and as to one she was not asked (but Rinehart was, and so it is included in the count of 39). On many of these deductions at trial, Geiger makes a greater deduction than she did originally because, as noted, she originally was more lenient. (Presumably, Gallup’s approach at trial was to show that the original analysis was lenient, a factor militating against a finding of unlawful motivation.) Geiger’s one-page table (RX 53) of handwritten totals re- flecting her analysis of Rinehart’s paysheets (GCX 16) and the intcheck (RX 54) contain just that—totals. References at trial to Geiger’s individual deductions are based on what Geiger estimates that she probably did in light of the totals. This is so because the notes (calculations) that Geiger made in June 1999 on the individual line items apparently were discarded. Geiger made them on Post-it notes and she does not know where they are. (13:2605) Thus, Geiger’s handwritten chart of totals for these 4 days is critical to our review, and I now reproduce the bulk of it, the 4-day week of Monday through Thursday, June 14 through 17, 1999 (RX 53). In this reproduction, I have modified the format somewhat, but not the substance. I have added the fourth column (Hours Logged-on based on Geiger’s testimony) and the last column (Geiger Allows at Trial) to show her stricter approach at trial. The column for hours logged is relevant because, describing the methodology that she followed, Geiger testified that she first calculated the number of hours (to the nearest tenth of an hour per RX 4 at 1) that Rinehart (or one of the other trackers) was logged onto the system each day. She then analyzed the line items of the intcheck (GCX 54). Any deductions were made from the total of logged-on hours, not from the hours claimed. (13:2590–2591) Even so, the logged-on hours also are relevant in relation to the claimed hours. Presumably Gei- DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1264 ger followed the same methodology on June 23, 1999—the day that she made her calculations. (13:2560–2561) The deduction figures that Geiger gave at trial were more restrictive than the ones that she applied in June 1999. (13:2604, 2625–2626, 2640, 2655–2656, 2670, 2676) Also, I have added the project number under the project name. Finally, I show in brackets Geiger’s mistake, admitted at trial (13:2675), respecting the UNL study on June 17. I make no changes in the totals based on the error because the error was not discovered before Rinehart’s discharge. With those modifications, Geiger’s hand chart for Janice Rinehart is as follows (RX 53): Date Project Actual Hours Claimed Hours Logged- on Geiger Allows June ‘99 Geiger Allows at Trial June 14 Ad- ventist W470 11.0 11.5 7.9 7.5 UNL R406 2.0 1.5 .9 .9 Totals 13.0 13.0 8.8 8.4 June 15 Ad- ventist W470 6.8 6.7 5.7 4.9 Cal Fed F170 .5 .7 .3 .3 Totals 7.3 7.4 6.0 5.2 June 16 Ad- ventist W470 12.0 11.4 7.2 6.8 UNL R406 2.0 2.2 2.0 2.0 Totals 14.0 13.6 9.2 8.8 June 17 Ad- ventist W470 4.0 5.3 3.5 2.6 UNL R406 2.0 [1.5] 1.4 1.2 .9 Totals 6.0 [5.5] 6.7 4.7 3.5 Totals For Week 40.3 [39.8] 40.7 28.7 25.9 The Government does not contest Supervisor Geiger’s analy- sis of any specific line item. Instead, the General Counsel sim- ply dismisses the relevance of Geiger’s analysis by arguing (Br. at 41–42) that Gallup’s intcheck system simply leaves out vari- ous work items that all of us would consider as being work for our employer. That argument reflects Rinehart’s opinion that the intcheck is unreliable and incomplete. (11:2030, 2033, 2058) But even if the intcheck were flawed (something dis- puted by Gallup), that would be of no help here to Rinehart and the General Counsel. As Gallup points out (Reply at 22), if it was flawed it was flawed for all employees, and the Govern- ment failed to show any disparate treatment exempting other employees from intchecks while applying the intcheck system to Rinehart. As for the various work items not picked up by the intcheck, that simply is Gallup’s system. As summarized earlier, Gallup pays interviewers for completed surveys based on a team rate. The various teams are determined by weekly hours. Earlier I summarized how someone on the 35-hour team makes more per completed survey than someone on the 25-hour or 15-hour teams. In that respect, “actual hours” are an important compo- nent of Gallup’s pay system. Accordingly, the General Coun- sel’s lament about all the time that interviewers can lose be- cause of “quota fills” may well be correct, but it is irrelevant. The same would apply to the “editing” process even if Rinehart had to fill out papers. Supervisor Geiger, however, credibly testified that the Adventist track had no paperwork because everything to be filled out was on the computer. (13:2607– 2608, 2667) To the extent that any computer editing is not counted by the computer clock, we can all sympathize, but again, that fact is irrelevant. Indeed, the Government should not need to be reminded that this is not a federal Wage and Hour case. (Even if it were, the compensation system here might be deemed a permissible piece rate system.) Nor is this an arbitration proceeding brought to determine what the term “actual hours” means under some collective bargaining agree- ment. The same situation applies to “callbacks,” another of the work items mentioned by the General Counsel. This term de- scribes a process by which an interviewer, having contacted someone, is asked to call back another time (nonspecific) or, better yet, at a specific time. If the telephone number already has been called several times, it will be culled from the list of eligibles when it reaches its planned limit of attempts—its “call design.” (JX 23 at 57) To avoid losing a potential complete to the “call design” computer file, interviewers will make a hand note of the number and time to call, then manually dial the number at the scheduled time. Only if and when the survey participant answers will the interviewer then bring up the com- puter screen. Thus, even though the interviewer gets no “actual hours” credit for her off-screen work on the callback until she brings up the screen for the survey, the interviewer considers that a worthwhile investment on the anticipated chance that the call will result in the monetary reward of a completed survey. (3:494–496, 499–01, Sico) To repeat, the off-screen work is outside Gallup’s compensation system. While these noncredited work items are interesting, Janice Rinehart (testifying some 10 months after her discharge) under- standably could not remember any details of any of the specific line items for the week of June 14, 1999. As to most of the 38 line items that Rinehart was asked about at trial, in explaining whether the extra portion of the gap time (the time beyond, for example, the half minute or so for the dialer to reach and dis- connect from an answering machine), Rinehart replied that she could have been, might have been, may have been, or possibly was or was not handling the other work items. (11:1991–1992, GALLUP, INC. 1265 2016, 2047, 2072, for example) Or, she “probably,” at least on one occasion of a 71-minute gap for a “No answer” that lasted 32 seconds,31 took part of the additional gap time for some lunch. (11:2043–2044) Rinehart agrees that a line item for an answering machine would take no more than one minute. (11:1991, 2021, 2046, 2074) Geiger says that 30 seconds is acceptable (13:2604), and the times reflected on the intcheck (GCX 54) for answering machines certainly support Geiger as to that. Recall Rinehart’s testimony that her short breaks of no more than 15 minutes, even if to pick up a sandwich at the downstairs deli to bring back to her cube, she included in her actual hours worked. That is, for these types of breaks, as for work items off the telephone, Rinehart left her computer on and the screen up. (11:2077–2078) When the screen is up and a number is left on the screen after a call of some kind, as Supervisor Gei- ger credibly reports (13:2662–2663, 2665–2666), the clock continues to run even if the interviewer gets up and goes somewhere. That is, the gap that shows up on the intcheck, instead of being the normal 4 to 5 minutes for, say, a completed survey on Adventist W470, becomes 20 minutes, 30 minutes, even an hour. [Geiger testified that the average time for a com- plete on W470 was 5 minutes. (13:2594, 2611–2613) Rinehart advises that in a “perfect world” the completes could all be done in 4 to 5 minutes or so, but sometimes, for example, the interviewer has to read items four or five times because the survey participant does not understand and the interviewer must not explain. (11:2051–2052)] Leaving the number up and the clock running appears to be contrary to a direct instruction in the ITM concerning a special screen to enter when the inter- viewer leaves his or her computer. (JX 23 at 50) All this leads to the next complaint of the General Counsel (Br. at 42), that being that the intcheck fails to disclose when an interviewer has to spend extra time on a call because, for example, the respon- dent (survey participant) wants to chat. (4) Supervisor Geiger’s interpretations (a) Introduction The 39 (my count) line items, or calls, in question here are disputed in the sense that Janice Rinehart and Supervisor Gei- ger, considered together, were asked about all of them at trial. (Recall that one line item was skipped in questioning Rinehart, and a separate item was skipped as to Geiger.) The actual number of calls for which, at trial, Geiger “adjusted” is 32. And as implied earlier, no one is sure whether all these 32 were the subject of adjustments on June 23, 1999 when Geiger pre- pared her page (RX 53) of calculation totals (13:2560–2561), although it appears that approximately all of them received, overall, a slightly smaller subtraction in June 1999 than Geiger applied at trial in explaining what she would deduct were she doing so as of the trial. In Geiger’s experience, chatty respondents who actually cause an extra long survey are “few and far between.” (13:2611–2613) Indeed, when we inspect the intcheck it is clear that for most of the “adjusted” calls (the ones for which Geiger disallowed the extra time), there are at least several 31 GCX 54 at 11 line 56. (By my count, this is disputed call 21.) more of the same category right there on the intcheck that fit the pattern of time described by Supervisor Geiger. Consider, for example, those calls that reached answering machines. Of the items causing Rinehart the biggest headaches here, answer- ing machines top the list. More specifically, it is the extra time beyond the short period that it takes the computer to connect with and disconnect from an answering machine. (In the trial version of her adjustments, Geiger would deduct 4.0 hours for that category alone.) A completed survey that takes extra time perhaps leaves some room for debate, particularly if there were only one or two on a week’s intcheck. But anything over 1 minute on an answering machine requires no analysis or inter- pretation. That is, the person left the number on the screen and did something else for the rest of that time gap. The extra time is what needs to be explained. The 39 line items, or calls, that are in question begin with, as number 1, GCX 54 at 1 line 20 (11 minutes for an answering machine) and end with, as number 39, GCX 54 at 19 line 90 (another 11-minute call to an answering machine). The last call, by the way, sums up much of the Government’s entire case as to Rinehart. Consider Rinehart’s testimonial explanation, on cross examination, of whether the remaining 10 minutes of the “call” (that is, the remaining 10 minutes of the time that the screen was “up”) was treated by Rinehart as “actual time worked” (a better phrasing would be “actual hours” worked), Rinehart replies yes. Thus (11:2077): A. I would say yes to that because that’s when I — it looks like I was through for the day and I was totaling my time sheet. We had to total it every night, get our totals and take it to a supervisor and have them initial on the pay sheet to turn it in. [Per the ITM, JX 23 at 23, “Pay sheets.”] So yes, I would consider that work time. BY MR. SHULTZ: And would you leave your screen on while you did that? A. Yes. Q. Why? A. Because I was still at work. I was still working. I was getting paperwork taken care of for the day, maybe turning in—since I hadn’t worked Adventist that day [Rinehart is confused here, for her first 4 hours that June 17 were on Adventist W470; GCX 16 at 4; GCX 54 at 16 to 18], I may also have been turning in my patient forms that I had filled out for the day. That was still work time as far as I’m concerned. Respecting the patient forms, Supervisor Geiger credibly ex- plains that on the Adventist project there was no paperwork because all forms were filled out while the interviewer was doing the survey on the computer. (13:2605–2607, 2667) Even as to the editing process (that is, correcting spelling and typographical errors, not substance), that too is done on the computer, as part of actual hours worked, before going to the next call. A specific computer procedure, “Edit mode,” is de- scribed in the ITM for this (JX 23 at 86-87 and 50). The edit mode procedure apparently is what Rinehart, at one point (11:2023), attempted to explain at trial. Thus, there would be no hard-copy patient forms for Adventist. Although UNL R406 (access code 9563), the second project that Rinehart DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1266 worked on that June 17, for 1.5 hours and a single complete (GCX 16 at 4), is not well described in the record, it appears that such project was for some organization or league sponsor- ing a study named “Entrepreneurship Youth Study.” (1:162– 163, Uria-Ruiz) It seems highly unlikely that such a study had anything to do with patient forms. In fact, Rinehart was never specific about the “patient forms,” and I find that she mentioned them generically as a throwback to earlier months and other studies when she and the other interviewers actually did fill out such paperwork. Indeed, Rinehart probably is referring to the old “Patient Trax Fax” forms (RX 47). But by May 1999 that form was on the com- puter for all hospital tracks except Beth Israel. (12:2439, Cof- fey; 15:3074, Uria-Ruiz) Supervisor Geiger’s reference to the form as a paper form on “some surveys” (14:2878–2880) does not conflict with the testimony of Coffey and Geiger. Thus, reinforcing Geiger, Supervisor Uria-Ruiz testified that the Trax Fax is filled out while the survey participant is on the tele- phone. (15:3077–3078) And indeed, Rinehart herself admits that in late May 1999 she found the form on her computer, and it was “prior to that” when she filled out the forms by hand and “I counted it all as work time.” (10:1917) Returning to Rinehart’s answer, quoted above, Rinehart goes on to state that she left her screen up whenever she engaged in something that (she considered) was work time. (11:2077) At that point she adds that the interviewers never turned off their computers unless they switched projects. (11:2077–2078 and 1983) Asked to focus specifically on the screen, Rinehart re- plied (11:2078), “I didn’t have any control over that. The dialer did that. The dialer does that.” But the interviewers do have control over the screen, and in the ITM Gallup instructs its interviewers to follow a specific procedure for notating, on the screen, their departure for such matters as restroom breaks if they leave their computers. (JX 23 at 50, “Leaving Your Com- puter”) Yet when Rinehart took any kind of break, even the long lunch breaks that she assertedly did not claim as actual hours, she left her computer screen on because “We’re not re- quired to turn them off. You left them up.” (11:1983) Well, “up” perhaps, but also Rinehart apparently never made the qualifying entry on the computer screen as instructed in the ITM about “Leaving Your Computer.” Geiger describes the procedure very well at trial, and explains that, although the entries would prevent the clock from running on the telephone number, the time would be shown for the break so that the in- terviewer could know how much time to deduct from the actual hours that she claimed at the end of the day. (14:2784–2789) Supervisor Uria-Ruiz gives similar testimony. (15:3071–3073) Rinehart was an experienced interviewer. In the words of Supervisor Geiger, Rinehart was a “veteran” interviewer. (14:2627) It is difficult to understand how Rinehart could be unaware of this procedural requirement that is clearly stated. For that matter, the instruction in the ITM (JX 23 at 24), on how to fill out the nightly paysheet, also is very clear that breaks of any kind are not to be counted in the number of “ac- tual hours.” As I already have discussed, Gallup’s new (1998) pay plan (RX 4) fails to make clear—as Supervisors Barlow and Geiger do clearly explain (as I summarized earlier)—that “actual hours” means only “dialing time,” that is, time on the telephone. It may be understandable that Rinehart and others possibly could have been confused as to whether such work tasks as those associated with quota fills were or were not properly counted as part of actual hours. Even so, it is puzzling as to how the Government can argue that such work items (some, such as waiting for a number, while sitting at the computer, and others while off the computer), count as actual hours, particu- larly when there is no evidence that Gallup ever has knowingly counted them. Actually, the General Counsel reduces the Gov- ernment’s position, on brief (Br. at 41–42), only to an implicit argument that such tasks are part of “actual hours.” At trial, the General Counsel’s stated position was that work associated with quota fills is part of “actual hours,” or that at least it is Gallup’s burden to show that it clearly explained to the em- ployees that such work was not part of actual hours. (10:1863, 1875, 1880–1882; 11:2093) The General Counsel seems to be arguing an arbitration or Wage-Hour case in the wrong forum. Before turning to some sample line items that illustrate Gei- ger’s adjustments, consider first the format of the intcheck here (GCX 54) for Janice Rinehart’s calls the 4-day week of June 14, 1999. The 19 pages have nine columns. The first column starts a numerical sequence for the calls placed on a specified project. The second is Rinehart’s identification number (9:1744), and the project (or study) code [the start-up code, actually] fills the third column. Next comes the date and then the ending time of the call. From the ending time of the call on the previous line to the ending time of the call on this line con- stitutes a time gap (the gross gap in my terminology). In the sixth column appears the seconds showing the time that the telephone number was on the computer screen. (13:2662– 2664, Geiger) Similarly, as John Selix explains (10:1811–1812) concerning the LL12 master computer log (GCX 55; 10:1795) from which intchecks are derived (10:1887–1888), the seconds reflected simply show the time that the number is, in effect, on the com- puter screen. Stated differently, the seconds shown are not necessarily equivalent to the duration of the telephone connec- tion, and we will see that from testimony given by Dyanne Compagna, the telephone specialist32 from MCI WorldCom. Indeed, recall Supervisor Geiger’s testimony that if an inter- viewer rises and leaves her cube with the screen up and number showing on the screen, the clock will continue to run accumu- lating seconds of time. (13:2662–2663) Also, the gross gap (reflecting minutes) may be substantially greater than the num- ber of seconds. When this happens, it means that the time dif- ference between the gross gap and the segment reflected by the seconds was a period in which no numbers were shown on the computer screen. (13:2654, 2663, Geiger) Although no witness explains the significance of Geiger’s testimony on that point, presumably that period of no screen activity would occur when there are some of the asserted quota- filled numbers that flash by, and perhaps when numbers are “running out” and difficult to obtain. In any event, the record indicates that this period of a blank computer screen could oc- 32 “Network Integrity Systems Specialist,” to be precise. (12:2347) GALLUP, INC. 1267 cur while the interviewer is sitting there waiting for a number “to drop,” as the witnesses phrase it. Return now to the columns of the intcheck. Column seven is the telephone number dialed (with “D” indicating that it was dialed by the autodialer, 11:1996). The eighth column is for the code number assigned to reflect, numerically, the result of the call. A list of such code numbers is in evidence. (JX 31; 9:1775–1776) The last column classifies, in text, the result of the call. (b) Sample deductions The Government does not attack (or even analyze) any spe- cific deduction that Supervisor Geiger made. Instead the Gen- eral Counsel erroneously (although now merely implicitly) makes the irrelevant argument that Gallup’s intcheck system excludes various work items. (This contention is irrelevant because Gallup does not include such items within its definition of “actual hours” worked. That term includes only “dialing time,” or time on the telephone.) I shall describe a few exam- ples of the deductions made by Supervisor Geiger. Notwith- standing the Government’s position of no attack, I have exam- ined all 39 questioned line items, and I have compared their patterns to the patterns of the other calls in the same categories reflected in the intcheck (GCX 54) for that week. I have done so as part of the process of considering the credibility of both Janice Rinehart and Supervisor Geiger. Before reviewing some sample deductions that Geiger made for that week, consider first the day that caught Geiger’s eye—Wednesday, June 16, 1999. For that Wednesday, June 16, Supervisor Geiger testified (13:2592), Rinehart claimed 12.0 “actual hours” for Adventist W470. (GCX 16 at 3; RX 53) (Recall that this was the “red flag” that sparked Geiger’s more detailed review and investiga- tion.) On reviewing the intcheck (GCX 54) for that date, Gei- ger determined that Rinehart was logged-on for 11.4 hours. (13:2592, 2625) After inspecting, at trial, the line items for that date, Geiger “adjusted” the 12 hours claimed by subtracting a total of 4.6 hours (for nine adjustments numbers 20 through 28) from the logged-on time of 11.4 hours, thereby allowing a (cor- rected) total of 6.8 hours. (13:2625–2626) And that does not even include the extra .6 hour canceled when Geiger deter- mined that the logged-on time was just 11.4 hours. Thus, the .6 hour was an immediate, and additional, loss. Of course, if an interviewer were able to tack on the “grace” allowance at the end of the day, then even using the lesser hours of 7.2 would allow at .7 hour grace allowance for the day, thereby bringing the logged-on total up and over the 12 hours claimed. As we have seen, however, the grace allowance has to be taken within the day (“use it or lose it”) and not tacked on at the end. Respecting the second project for that June 16, UNL, recall Geiger’s testimony (13:2628–2629) that she did not adjust the claim there of 2 hours. For the day, therefore, in June 1999 Geiger presumably subtracted .6 hour right “off the bat” be- cause of the logged-on time difference, plus, it appears, an ad- ditional 4.2 hours in line-item deductions, for a total of 4.8 hours deducted from the 12.0 hours claimed. This left (RX 53) an allowed total of 7.2 hours for Adventist W470. [That, plus the 2 hours for UNL, left Rinehart an allowed total of 9.2 hours in June 1999, rather than the 14.0 hours claimed. (RX 53)] For Adventist W470 that Wednesday, June 16, the four big- gest (of nine) line-item cuts are for 1.2 hours for a “No answer” (adjustment 21), .8 hour for a “Respondent unavailable for study” (adjustment 22), .5 hour for an “Answering machine” (adjustment 23), and .7 hour for a “Complete” (adjustment 25), for a total deduction of 3.2 hours just for these four line items. I cover these four deductions in the following summary of some sample deductions from the categories of deductions for the week. Keep in mind that Rinehart, as a “tracker” and a member of the Premier team, received Gallup’s 5-hour grace allowance for such things as breaks, handling quota fills, and tending to other non-dialing activities. Turn now to some sam- ple deductions from the categories of line items for the week. The 39 line-item deductions fall into 12 separate categories, with a few of the groups having little impact here. Considering first the most active group, that of answering machines (the classification placed in the last column of the intcheck), I note that it had 14 of the 39 calls. Of these 14 calls, Supervisor Geiger did not, in June 1999, make any deduction from two 6- minute calls (numbers 31 and 32 of the 3933) on the second project for June 16 — UNL R406. However, Geiger adds that she could have done so. (13:2628–2629) From the remaining 12 calls, consider the two highest gaps. First, number 3 is a 52- minute gap, on June 14, when the seconds (the time the screen was up—the “dialing time” count) were only 30. (GCX 54 at 1 line 51) Rinehart agrees to a 51-minute deduction (11:1995), and at trial, and presumably on June 23, 1999, Geiger made an .8-hour (48 minutes) deduction. (13:2634) The second highest is a 35-minute call (my number 23) on Wednesday, June 16, for which the screen was up 2122 sec- onds—35.4 minutes. Rinehart testified that, for the 34 minutes beyond the 1-minute allowance for the call, she may have been seeking out and working with a supervisor to get quota-filled numbers hidden, or she may have been engaged in other (nondialing) work tasks. (11:2046–2048) Of course, no matter what you, I, or the Government might think appropriate, such non-dialing work activities do not qualify as “actual hours” under Gallup’s pay system. Geiger subtracted .5 hour (30 min- utes) at trial. Although the nearest tenth of an hour per the pay plan (RX 4 at 1) would be .6 hour, Geiger testified that she did not want to be “picky.” (13:2604–2605) As mentioned, for the category of answering machines, at trial Supervisor Geiger deducted a total of 4.0 hours for 12 of the 14 line items. The two biggest deductions (.8 and .5 hour) I just covered. Of the 39 contested (adjusted, corrected) calls, the first and the last are to answering machines (calls 1 and 39). As to those two 11-minute calls, at trial Geiger charged .2 hour each. (13:2633, 2674) In June 1999 she may not have both- ered with these two because each is barely over 10 minutes. Accordingly, I shall not count these two. Other calls that were rounded up were offset by calls when Geiger rounded down. Indeed, Geiger credibly testified that her practice is to round up the first time and down the next in order to seek a balance. (14:2904–2905, 2920–2921) 33 GCX 54 at 16 lines 101 (13:2631) and 111. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1268 I do not count call number 6, a 9-minute call to an answering machine on June 14 (GCX 54 at 3 line 151) for which, at trial, Geiger made a .1 hour deduction. (13:2636–2637) Also, I do not count call number 37, an 11-minute call (GCX 54 at 18 line 2) for which, at trial, Geiger deducted .2 hour. (13:2673) Simi- larly, Geiger may not have counted a 10-minute call (number 38) on June 17 (GCX 54 at line 42), even though at trial (13:2674) she deducted .1 hour. Subtracting all these limited charges totaling .8 hour from the 4.0 hours leaves a deductions total of 3.2 hours that, I find, Supervisor Geiger probably made in June 1999 in preparing her totals page, RX 53. Another category that leaves no room for controversy is that for a “No answer.” There was one of these (call 21, mentioned a few paragraphs earlier), and the gap lasted 71 minutes, al- though the screen was up for only the usual time for such calls—32 seconds. (GCX 54 at 11 line 56) Respecting the extra 70 minutes for this line item, Rinehart concedes that, as the call occurred close to the noon hour, she “probably” took part of that extra time for lunch. (11:2043–2045) Observing that Gallup does not monitor breaks or shopping trips, Geiger suggests that Rinehart may well have spent time outside if the weather was pleasant that day. In any event, at trial Geiger, rounding up and over by 1 minute, deducted 1.2 hours. (13:2599–2602) Because Geiger was being more strict at trial than she was in June 1999, I suspect that at the original time Geiger was a bit more lenient, applying a charge, nor of 1.2, but of 1.1 hours. I so find. A somewhat similar category is that for the description, “Re- spondent is unavailable for study.” Two calls here (my num- bers 22 and 36) are charged. (GCX 54 at 11 line 89 and at 17 line 58) The first is a 54-minute gap, 3234 seconds, on June 16; and the second is a 32-minute gap, 1910 seconds, on June 17. As to the first call, Janice Rinehart candidly reports that she does not know what she was doing beyond the initial con- nection time. (11:2045–2046) Supervisor Geiger tells us that this category of call typically consumes no more than a single minute. Even so, she subtracted only .8 hour rather than a .9. (13:260–2603) This category of calls, Geiger explains, usually means that the person will be away for a substantial time, such as on vacation or out of the country, and will not be available for the period of the study. (13:2666) I add the subtraction of .8 hour to the total. For the second call (number 36) of that category, for 32 min- utes, Geiger subtracts .5 hour. (13:2665) Geiger explains that the extra time could be that the interviewer, with the screen left up, simply walked away from her computer. (13:2665–2666) Rinehart testified that it is “possible” she spoke with someone for the 32 minutes to determine availability, and if not, the rest of the time “Possibly would be and then again it may not be” “actual work time.” (11:2072) Rinehart’s response is not per- suasive, even though I recognize the time disadvantage imposes on her memory. However, I note that the intcheck reflects nine other such calls, each being less than a minute, and the remain- ing five lasting from 1.2 minutes to 3.1 minutes. Although there could be an exceptional situation, nothing indicates that such occurred here. Crediting Supervisor Geiger on this call, I add her .5 hour subtraction to the list. The total subtractions for this category are 1.3 hours. The next category I reach is that for “Screen failure.” Super- visor Geiger advises that this descriptive text means that the former patient does not live there or that no one at the number is qualified to participate in the survey. (13:2486, 2662) There are four calls here (my numbers 19, 29, 30, and 33), with num- bers 29 and 30, of 5 minutes and 25 minutes on June 16 (GCX 54 at 15 lines 41 and 70) receiving no adjustment from Geiger. Typically, Geiger reports (13:2662), screen failures take less than a minute. Call 19 (GCX 54 at 9 line 6) lasted for 21 min- utes, with the seconds listed as 1278, on June 15. At trial Gei- ger, rounding up, subtracted .4 hour. (13:2657–2658) Ram- bling from one code to another, Rinehart’s testimony on the point is that she is “not sure what all the coding stuff means.” (11:2025) An inspection of the intcheck (GCX 54) reveals that, of the other (besides these four) 15 screen failures that week, 12 lasted no more than 60 seconds, two more finished within 70 seconds, and the longest one lasted for 1.9 minutes. The average length of these additional 15 calls is .9 minute. I generally credit Su- pervisor Geiger here. Geiger calculated Rinehart’s logged-on time as 41 minutes, or .7 hour. (13:2657) Oddly, Rinehart claimed only .5 hour. (GCX 16 at 2; RX 53) Subtracting Geiger’s deduction of .4 hour from the .7 logged-on time (and not from the .5 hour claimed), Geiger arrived at the .3 hour that she allowed on her calculation page (RX 53) in June 1999 as well as at trial for Cal Fed F170. (13:2658) Accordingly, I count the .4 hour here as one of the sample deductions. The other screen failure (questioned call number 33), on June 17, had a 70-minute gap and the screen was up for only 216 seconds, or 2.3 minutes rounded to the nearest tenth of an hour. (GCX 54 at 16 line 5) That means that, for over 67 min- utes that the screen was up that June 17 there was no number or numbers on the screen. (13:2654, 2663) Was Rinehart waiting for a number to drop, or had she simply left her cube without properly switching to the “break” screen (per JX 23 at 50)? Rinehart concedes that she would not think that all of the extra time (beyond whatever time it took to handle the call) was de- voted to actual work time. (11:2066–2067) At trial Supervisor Geiger subtracted 1.1 hours, or 66 minutes, leaving 4 minutes for the call. Such 4 minutes exceeds the time that the number was on the screen. (13:2661, 2664) Overall, I find that Super- visor Geiger’s two deductions for screen failures, totaling 1.5 hours, appear to be reasonable subtractions. Adding the foregoing four category totals yields a grand total of 7.7 hours of deductions that Supervisor Geiger probably made originally. Recall Geiger’s testimony that anything over 5 hours meant discharge. (13:2535; 14:2846, 2865–2866) Some testimony about the June 1999 totals tends to be a bit confusing. This concerns whether the totals that Supervisor Geiger allowed in her June 1999 calculations (RX 53) include the allowance for breaktime. Thus, Geiger testified that the total actual hours allowed of 28.7 do include the allowance for breaks. (14:2846–2847) I find that to be true only in the sense that Geiger, as she testified at several points, tended to be lib- eral, or lenient, at trial, and usually subtracting less than she could have. (13:2655–2656, for example.) In that sense, her June 1999 total of 28.7 actual hours (RX 53), as I have stated, GALLUP, INC. 1269 is somewhat greater than (2.8 hours greater than) Geiger’s total of 25.9 actual hours as computed at trial when she was being stricter than she was in June 1999. Under her proper methodology, Geiger would calculate the logged-on time, then go back through the intcheck and strip out all the breaks, leaving just the dialing time (the pure “actual hours”). (13:2536, 2590–2592; 14:2825 as to Sherri Lee) More specifically, as we see from all the calculations, reaching just the dialing time means examining the “screen up” time (as shown by the line item figure for seconds), then analyzing the gap, and “adjusting” that gap so as to eliminate (when appro- priate in the supervisor’s opinion) some or most of the gap, and converting the seconds (when appropriate in the opinion of the supervisor) to whatever number is consistent with (in the su- pervisor’s opinion) the average time required to make that cate- gory of call. As we have seen, that is the process followed in stripping out “all the breaks.” Once all the breaks have been subtracted, the supervisor (Stephanie Geiger, here) then adds the grace allowance—5 hours for trackers such as Janice Rinehart. (In theory, the claimed actual hours should match the actual hours as deter- mined by the supervisor’s review of the intcheck.) If the claimed hours exceed the actual hours shown on the intcheck, yet fall within the grace allowance of 5 hours, the interviewer may be disciplined, generally a warning. If the claimed hours exceed the 5 hours, the asserted penalty is termination. At a couple of points during her cross examination, Supervi- sor Geiger’s testimony appears either garbled or inconsistent with the foregoing. Thus, after answering a question consistent with the foregoing, she then answers yes to the question, “So before adding in break time they would be afforded a five-hour leeway.” (14:2865) The question is confusing, and I think that Geiger misunderstood it. After all, Geiger had just testified, “The five-hour leeway period is to represent the break time that they get during the week.” (14:2865) In short, there is not some extra breaktime added to the 5-hour grace allowance. Beyond the grace allowance, Gallup does not pay (in the sense of qualifying for the higher-paid teams) for breaks. On the next page of the transcript (14:2866) another ques- tion, this time by me, is confusing. Again, the question and the affirmative answer suggest that the 5-hour grace period can be added on top of a week that already contains break periods. Geiger probably meant her affirmative answer in relation to what she had allowed for Janice Rinehart in June 1999, and I so find. Turn back now to the categories. The four I have described above produced, as noted, a total of 7.7 hours that, I find, Su- pervisor Geiger reasonably would have deducted in June 1999 from the 40.7 logged-on hours, resulting in total actual hours allowed of 33. Now adding the 5-hour grace allowance returns the total allowed hours to 38.0—unfortunately, 2.3 hours shy of the 40.3 claimed hours. Stated differently, Rinehart’s claimed actual hours exceeded the allowed actual hours plus the 5-hour grace allowance by 2.3 hours for the week. Under Gallup’s system, Janice Rinehart would be subject to termination be- cause her claimed actual hours exceeded the 5-hour grace al- lowance (and did so by 2.3 hours). Although, at this point, I need not consider any additional categories in light of the foregoing, a quick summary of another category or two that contains big gaps may be useful. The category for “Complete” has a 52-minute gap of 3112 seconds for Adventist W470 on June 16. (Questioned call 25; GCX 54 at 13 line 182) As summarized earlier, the length of time needed to complete a survey on Adventist W470 is a matter of debate between Supervisor Geiger and Janice Rinehart. As mentioned there, Geiger puts the average time at 5 minutes. Rinehart would agree if the calls were made in a “perfect world.” Both witnesses have good arguments. For comparison of times, the intcheck in question (GCX 54) has many com- pletes for that week—183 to be exact. (GCX 16 at 1–4 and 11:2072–2073; GCX 54 at 1–19) The first 100 completes on the intcheck (carrying through the first 2 days, June 14 and 15) consumed 489.1 minutes for an average time (rounded to the nearest .1 minute) of 4.9 minutes—squarely within the time frame described by Supervisor Geiger.34 A visual check of the next 48 completes (leading up to questioned call 25, Rinehart’s 49th complete for the day and complete number 149 for the week) indicates an even lesser time with nearly all being 3 to 6- minute calls. Respecting this questioned, or contested, call 25 of 52 min- utes, Supervisor Geiger deducted .7 hour (42 minutes) from the logged-on time, thereby allowing 10 minutes for the call, or “twice the average time of the survey.” (13:2611, Geiger) Asked (on cross examination) whether this completed survey required 52 minutes, Janice Rinehart replied (11:2049), “It’s a good possibility.” To the question whether the balance of the time, assuming the actual call took just 5 minutes, would have been actual work time, Rinehart answered “No,” but added, in a meandering fashion, that she could have taken 10 minutes or so to have filled out a patient “track form.” (11:2050) Presuma- bly Rinehart was referring to a Patient Trax Fax (RX 47), or PTF, described earlier, that (except for the hospital Beth Israel) has been on the computer for the hospital tracks since May 1999. And (also noted earlier) as Supervisor Uria-Ruiz testi- fied (15:3077–3078), PTFs are completed while the survey respondent is on the telephone. Rinehart later essentially reaf- firms her testimony. (11:2084) Of course, Rinehart appears to be saying that beyond that 15 minutes (a possible 5-minute call and 10 minutes on a form) she would not have been engaging in actual work time. Now it so happens that Gallup requires (11:2079, Rinehart) the interviewers to record, on the back of their payslips, the time duration and the “case ID number” for each of their com- pleted surveys. [For the trial exhibits, the documents were copied in a manner to reproduce these hand notes by Rinehart at the bottom of the face of the payslips.] The computer dis- plays this information during a short period of time after the call is disconnected so that the interviewers can copy the data. The case ID numbers are displayed in a numerical sequence. 34 The twelfth complete, questioned call 2 (GCX 54 at 1 line 42), is part of a 17-minute gap, but the screen was up only 202 seconds. Su- pervisor Geiger gave Rinehart a pass on the gap. (13:2633–2634) I count this call of 202 seconds as part of the first 100. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1270 (9:1781–1783; 11:2080–2082, Rinehart; 13:2618–2623, Gei- ger; 15:3078, Uria-Ruiz) Taking the foregoing information about the hand notes and reviewing Janice Rinehart’s paysheet for Wednesday, June 16, 1999 (GCX 16 at 3), and the intcheck (GCX 54 at 13 line 182), we see that the 52-minute call (contested call 25) corresponds to Rinehart’s hand note, “4.3–1840.” Rinehart’s hand note means that the call, being complete number 49 for the day [and number 149 for the week], was displayed on the computer screen, and copied by Rinehart in her own hand, as lasting 4.3 minutes and as having been assigned case ID number 1840. (11:2084, Rinehart; 15:3082–3085, Uria-Ruiz—respecting counting backwards to reach a given line.) In short, by her own hand on that June 16, Rinehart recorded the duration of the call as 4.3 minutes. Gallup’s position is that the numbers Rinehart wrote by her own hand trap her in her own discrepancy. (Br. at 72) The discrepancy, of course, is that the computer clock kept ticking until the total time for the “screen-up” situation reached 52 minutes. In short, for some reason, Janice Rinehart left the screen up after she completed the 4.3-minute call. There is more. Recall my earlier mention that we would learn from the MCI representative, Dyanne Compagna, about the real duration of the telephone time. For this so-called 52- minute call to this specific telephone number of contested call 25, MCI’s records show a different story. Looking now at the specific MCI record (RX 62 at 2, third line from the top), we see that the call that ended at 6:23 p.m. (“18:23”) on line 182 of the intcheck (GCX 54 at 13), and which on Gallup’s computer had a “screen up” gap of 3112 seconds (51.9 minutes), is shown on MCI’s record, for the same telephone number called,35 a dialing and ringing time of 15 seconds followed by an “off hook” (from answer to end of conversation or answering ma- chine message) duration of 4 minutes 24 seconds (4.4 minutes), ending at 35 seconds after 6:23 p.m. (15:2992, Compagna) Although the General Counsel is correct in stating (Brief at 45) that Gallup did not have or rely on the MCI records as of the discharge (15:3126-3128), Gallup is also correct (Reply at 21) in suggesting that they are relevant because they aid in showing the reasonableness of Supervisor Geiger’s analysis of the time records and in her conclusion that Janice Rinehart falsified her time records. In that respect, the MCI records bear on credibil- ity. Thus, they serve a dual purpose while assisting in the deci- sion process. Note that the 47-minute remainder of this 52-minute “call” occurred at the front end of the 52-minute gap. This is shown by the fact that the previous call ended with a busy signal at 5:31 p.m. (GCX 54 at 13 line 181), and the call after contested call 25 was a 32-second “No answer” ending at 6:24 p.m. (line 183). This would mean that Rinehart was not completing any form on the computer, for the screens for the previous calls were gone, and the next call (the one in issue) was not yet made. The remaining possibilities appear to be that Rinehart, during this 47 minutes, either was working on quota fills, wait- 35 Although MCI’s record (RX 62) does not show from which exten- sion at Gallup the call was placed (15:2997), that fact is immaterial in view of the showing that the telephone numbers and ending times match on both Rinehart’s intcheck and the MCI record. ing for a number “to drop,” or possibly stymied by numbers “running out.” That week, Supervisor Geiger testified, she and the trackers were scrambling trying to get more numbers to dial on Adventist W470. The project was simply running out of numbers. (13:2551, 2553, Geiger) Although Geiger does not address the possibility of numbers running out as to this contested call 25, her general description of the problem indicates that such a problem did not afflict this 47-minute period. Thus, in describing her efforts to overcome that problem, she reports that the interviewer would “log back on,” and “change the minutes,” and “dial through the sample again and again.” (13:2553) Indeed, the interviewers are to switch to a different project, especially if a supervisor cannot be found right away. (13:2601–2602, Geiger) Steps such as those surely would leave a computer trail on the intcheck, and there are no such interruptions on the dialing sequence already de- scribed for Rinehart’s intcheck (GCX 54 at 13) That seems to leave two possibilities. One, Rinehart possibly was seeking out a supervisor to hide quota-filled numbers. Two, Rinehart sim- ply took a 47-minute break. Respecting the first possibility, Supervisor Geiger suggests that when the stage is reached where numbers are running out, quota fills are no longer the problem. This is so because quota fills occur when there is still “sample” (numbers) left. But in this last full week of the pro- ject, the problem was that the sample itself was running out. (13:2600–2601) This process of eliminating the possibilities (so far as this re- cord suggests) indicates that Janice Rinehart simply took a break for these 47 minutes. In this connection, observe that this break period occurred at a normal suppertime starting at 5:31 p.m. (Line 181) Based on Rinehart’s earlier testimony about the meal breaks that she took, this would have been one of her 30-minute or more meal breaks for which, as she testified (10:1917; 11:1983), she did not claim as actual hours. Later on the same page of the intcheck there are two other contested gaps. The first of these (contested call 26 at line 195) is a 26- minute call (beginning at 6:43 p.m.) to an answering machine, and the second (contested call 27 at line 200) is a 15-minute complete, for which Geiger gave Rinehart a pass (13:2616). While Rinehart could have brought something from the down- stairs deli back to her desk for supper as to either of these gaps, the existence of the longer 47-minute break, which fits more with a normal supper hour, suggests that Rinehart simply en- joyed this 47 minutes as her supper break—even though she normally did not take an evening meal break (11:2106). Rinehart’s last call (or “screen up”) that June 16 ended at 10:53 p.m. (GCX 54 at 16 line 111) Recall that Rinehart concedes she “probably” took some of a 71-minute gap (involving a 32- second “No answer”—contested call 21) on June 16 (GCX 54 at 11 line 56) for lunch. In any event, I find that, in June 1999, Supervisor Geiger reasonably could have viewed this 52- minute gap on contested call 25 as excessive for a complete, that her 10-minute allowance (subtracting only .7 hour) for the call was generous, and that her implied conclusion that Rinehart simply took a break was reasonable based on Geiger’s knowledge and experience. Before moving to the next topic, I make one or two quick observations about the rebuttal testimony of Supervisor Uria- GALLUP, INC. 1271 Ruiz concerning testimony by Rinehart (9:1776–1781; 11:1987, 2109–2115) that it was difficult even to meet quota on Advent- ist W470 as of June 16, it being near the end of the project, and especially so with the shorter time that Supervisor Geiger was leaving Rinehart after Geiger’s subtractions at trial. Uria- Ruiz’s testimony is in the nature of begging the question of what was Janice Rinehart doing during the time gaps—working on quota fills and such or taking breaks. To help me resolve that question, Uria-Ruiz should have addressed the testimony by Supervisor Geiger about the scrambling going on that last full week of Adventist W470 because the project was running out of numbers. Uria-Ruiz did not do so. It is useless to take a short span of time, when Rinehart was able to pull in several completes, and then, taking that block of time as something Rinehart could achieve over the rest of the day, argue that Rinehart really could have doubled her production (that would be 10 completes per hour, 11:2115, Rinehart) had she worked all the time instead of, it is implied, taking breaks (15:3116), or as phrased in Gallup’s Reply Brief (at 22), working produc- tively and then coasting with long breaks. Supervisor Uria-Ruiz acknowledges that Janice Rinehart was a “very productive” interviewer. (15:3116) And Gallup’s re- cords reflect that to be the case that 4-day week, with Rinehart pulling in the most completes for Adventist W470 on June 14 and 16 (57 and 59 completes; RX 52 col. “Total CMs”— 13:2549–2550) and Lisa Miller on June 15 and 17 (46 and 30 completes). As Supervisor Geiger suggests, it is in the finan- cial interest of the interviewers to complete surveys as quickly as can be done properly and courteously because the interview- ers are paid for completed surveys. (13:2613) Supervisor Uria-Ruiz’s criticism of Rinehart’s devotion to duty, followed by the complimentary phrase about being a “very productive” interviewer, had all the air of those lines in Scripture (Mark 7:6): This people honors me with their lips, but their hearts are far from me. Our next category is “Illness.” This category has two calls of about an hour each on Adventist W470. (Contested calls 17 and 35) Call 17 is for 57 minutes (3428 seconds) on June 15. The other gap, call 35, is 65 minutes with seconds of 3910 on June 17. (GCX 54 at 9 line 134 and 17 line 43) Supervisor Geiger subtracted .9 hour (54 minutes) from the first (13:2653) and 1.0 hour from the second (13:2664–2665) Geiger testified that the “Illness” code means that the person was sick and did not want to do the survey. (13:2651) Geiger implies that such a call would be very short. There are not many illness calls on the intcheck. I count only six, and four of those are contested (calls 7, 15, 17, and 35) with the gaps running from 13 minutes (call 7) to 65 minutes (call 35). The two non-contested calls (the fourth and fifth illness calls) lasted 64 and 68 seconds on June 16 and 17, respectively (GCX 54 at 13 line 185 and 16 line 12). When asked whether she spoke to the respondent for 57 minutes (call 17), Rinehart replied, “Maybe and maybe not.” (11:2019) MCI’s record (RX 62 at 1, third item from bottom) shows that the actual telephone call, from the start of dialing to the disconnection, lasted exactly 58 seconds—16 seconds for dialing and ringing and 42 seconds that the receiver telephone was “off the hook.” (15:2992, Compagna) This call was not a completed survey where, after the disconnect, Rinehart possibly went into “Edit mode” to correct spelling and typing errors while the computer was still showing a connection with the “screen up.” Supervisor Geiger subtracted .9 hour (13:2652) thereby leaving a generous 3 minutes for this call. A similar result would obtain for the 65-minute gap (call 35) for which Geiger deducted 1.0 hour (13:2664–2665), thereby leaving Rinehart a generous 5 minutes for the call. Notwith- standing Rinehart’s testimony (11:2067–2068, 2071) that the call may or may not have lasted 65 minutes (and that any re- mainder “possibly” was or was not actual work time36), we know from MCI’s record (RX 62 at 2, second to last call) that the June 17 call to this number lasted only 49 seconds. I find that Supervisor Geiger, in June 1999, reasonably would have deducted 1.0 for this “illness” call from the Rinehart’s logged on hours that day. Respecting the “Illness” category as sur- veyed here, that brings the total deductions to 1.9 hours. Turn now to the category for “Specific callback.” Earlier I described that a “specific callback” means that the respondent has agreed to a specific time to be called back to do the inter- view. Interviewers will even work “off screen” in order to convert a specific callback into a complete. The question here focuses on the first stage—how long it typically takes to proc- ess a call in which the respondent agrees to a later call at a spe- cific time to do the interview. Supervisor Geiger states that an acceptable time for that is 2 minutes. (13:2596–2599) In so testifying, Geiger pointed to 2-minute gap (76 seconds) for a specific callback. (GCX 54 at 10 line 9) There are many spe- cific callbacks on the intcheck (112 by my count for the 4 days in issue), and a visual check shows that nearly all (aside from the contested ones here plus about three others of about 4 min- utes to about 5 minutes) are made well within 2 minutes. The category “Specific callback” has five contested calls (numbers 8, 9, 20, 28, and 34). I focus on the two that have the largest gaps, that being call 8, a 70-minute gap of 4230 sec- onds, and call 9, a 54-minute gap of 3162 seconds (GCX 54 at 4 lines 161 and 181), both on June 14. At trial Supervisor Gei- ger deducted 1.1 hours for the 70-minute call and .8 hour for the 54-minute call. (13:2637–2638) Calls 20 and 28 have sub- stantial gaps of 23 and 26 minutes, respectively, and call 34 is a 6-minute gap with a screen up situation of just 84 seconds, but I shall skip these as the larger ones serve the purpose of under- standing what happened. Respecting the 70-minute gap, questioned call number 8, and whether she spoke to the respondent for 70 minutes, Rinehart answered, “Maybe and maybe not,” adding that it was a possi- bility. It happened—“Not real often, but yes, it did happen.” (11:1999–2000) Let us see if it happened this time. Inspecting the telephone number and time on the intcheck (GCX 54 at 4 line 161), and finding a match for the ending time of 5:44 p.m. (“17:44”) on MCI’s record (RX 62 at 1 line item number 6), I see that the time for dialing and ringing consumed 12 seconds and the conversation, or “off hook,” portion another 38 sec- 36 This apparently means, as to the “possibly not” actual worktime, that such time was a break. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1272 onds, for a total actual call time of 50 seconds. Thus, I find, this call was not one of those times when it happened that Janice Rinehart, in obtaining a special callback appointment, took an hour rather than 2 minutes or so. Shift now to contested call number 9, the 52-minute gap (52 minutes and 42 seconds). To the same question as before, Janice Rinehart answered (11:2000), “Maybe and maybe not. It just depended pretty much on the respondents.” MCI’s record shows a telephone number match, and an ending time match of 7:13 p.m., one line below that of the call just discussed. (RX 62 at 1, seventh call) MCI’s line item shows that the dialing and ringing time required 12 seconds, and the “off hook” por- tion another 40 seconds, for a total actual duration time for the call of 52 seconds. Again, therefore, this call was not one of those exceptional ones that would occupy Janice Rinehart for nearly an hour. In light of the foregoing, I find that, in June 1999, Supervisor Geiger reasonably deducted 1.1 hours and .8 hour for calls 8 and 9, leaving a generous allowance of 4 minutes (double the “acceptable” time) for each call. Adding the two calls of 1.1 hours and .8 hour gives a total deduction of 1.9 hours for this category of “Specific callback.” Recapping the total hours probably deducted by Supervisor Geiger in June 1999 for these seven categories reflects as fol- lows: answering machines, 3.2; no answer, 1.1; respondent is unavailable for study, 1.3; screen failure, 1.5; complete, .7; illness, 1.9;, and specific callbacks, 1.9, for a grand total of 11.6 hours. Subtracting these 11.6 hours from Janice Rinehart’s logged-on hours of 40.7 gives us a weekly total of 29.1 hours as the bulk figure for what Geiger probably deducted that June. Although that figure is greater than the 28.7 hours that Geiger actually allowed on that June 23 (RX 53), hours that I have not counted remain in the other 5 categories. Adding, at this point, those remaining hours, as given by Geiger at trial, yields the following hours: Death of respondent, .3; Bad/disconnected number, .0; Nonspecific callback, .3; Non-target, .1; and Span- ish interviewer wanted, .3. These uncounted numbers total 1.0 hours. If that figure of 1.0 hours were subtracted from the 29.1 that I have reached, the total would be 28.1—less than the 28.7 that Geiger actually allowed that June 23, but more than the 25.9 that Supervisor Geiger allowed at trial. Of course, in these seven categories (actually six, not counting answering ma- chines) are a few numbers that Geiger possibly would have included in her deductions that June. Were I to examine the five remaining categories, I suspect that I would conclude that part of the 1.0 probably would not have been among the deduc- tions that Geiger made in June 1999. In short, my mostly complete effort at the numbers has not produced a perfect match. But exact accuracy has not been the goal, as such a goal would be unattainable anyhow. (Supervi- sor Geiger could not remember what she had done, and could only approximate what she did do that June 23, 1999 in the process, at trial, of showing what she would do today.) Instead, as part of the process of resolving credibility, I have sought to determine whether Supervisor Geiger had a reasonable basis, and methodology, for the numbers that she arrived at in June 1999. I discuss this further in a moment. (5) Supervisor Geiger checks others Supervisor Geiger also checked the weekly printout (RX 52) for Adventist W470 as to the other employees besides Janice Rinehart. In doing so, she noticed that Lisa Miller also claimed many hours on Wednesday, June 16—8 hours in fact. (13:2567, 2577; 14:2874) As shown by the printout, the other trackers on the printout had fewer hours on this project than either Rinehart or Miller. Also, no others on the printout claimed more hours than the study was “up.” (13:2560, Gei- ger) In checking Miller’s payslip for the date of June 16 (GCX 19 at 3), and calling up a printout (RX 54) from Gallup’s intra- net database for that date, plus an intcheck (no copy identified or offered in evidence), Geiger determined that Miller had claimed 8.0 hours and had been logged on to the system even longer. Although Miller’s actual time on the computer was just 7.6 hours, the difference was well within the allowed grace period. The extra .4 hour, as part of the grace period, would be added to the 7.6 because Miller was logged on to the system for a sufficient time (duration number not specified on the record) to cover the .4 hour needed. (13:2577–2579, 2585–2588; 14:2783–2876, Geiger) From the record, it appears that Lisa Miller’s hours on Adventist W470 were within the permissible range for that Wednesday, June 16. The General Counsel argues (Br. at 48) that Geiger “conven- iently ignored” a red flag as to Lisa Miller, for a separate Ad- ventist project, on that June 16. The asserted discrepancy shows up when the time of 1 hour 55 minutes on the intranet printout (RX 54) for Adventist W169, or the “Adventist Urgent Care Track,” is compared with Miller’s claimed 3.0 actual hours on that project for that date on her payslip (GCX 19 at 3). [Recall that the intranet page is called up by date and the inter- viewer’s ID number. (13:2585, Geiger) That page shows the projects the interviewer worked that date. For Lisa Miller, this is confirmed by her paysheet, GCX 19 at 3. In short, the intra- net page pulled up will show all projects worked that date, not just the one being investigated.] Geiger admits that she cannot explain the discrepancy because she did not investigate it. Asked why she did not investigate it, Geiger explained (14:2876): That wasn’t the track I was looking at. I was looking at the Adventist in/and out ER37 track. That is the same track, or project, as Adventist W470, as re- flected on Janice Rinehart’s payslips. (GCX 16 at 3) Asked why Geiger decided to look at Janice Rinehart’s hours for the rest of the week, Supervisor Geiger, extending her initial “red flag” testimony as to why she checked Wednesday, June 16 (13:2551, 2569, 2590; 14:2854), advises that it was because of the (large) discrepancy of 4.8 hours for that Wednesday, June 16. (14:2877) At one point Supervisor Geiger (who no longer works at Gallup) momentarily suggests, indirectly, that Advent- ist W470 is the “Pops” track. (13:2561) Not so. “Pops” is another name for Adventist W169 (the one Lisa Miller 37 The track name, garbled in the transcript, is shown as modified by the stipulated (RX 87) correction as to this item. (GCX 1rrr at 3) The other exhibits support this correction. GALLUP, INC. 1273 worked), and Rinehart worked “Pops” on Monday and Tues- day, June 21-22. (GCX 16 at 5–6) Having examined and summarized these points, I find noth- ing developed in the record that points to anything done, or not done, here by Supervisor Geiger that would suggest anything improper respecting the Lisa Miller matter. For example, the General Counsel did not show that it is Gallup policy (or at least Geiger’s practice) to investigate in all directions, as indi- cated by possible discrepancies, whenever the supervisor is checking only one track. As I cautioned the General Counsel at trial, I need to see evidence of how Geiger deviated from that which she ordinarily would have done. (14:2886) No such evidence was offered. The General Counsel also argues (Brief at 49) disparity fa- voring one Russell Lund. About May 1999, Beverly Robinson testified, she and Catherine Wagley went to Payroll Coordina- tor Mike Cota (an admitted agent of Gallup) to see if their pay- roll sheets matched the new payroll tracking system. Cota said that Robinson’s was about 30 minutes high, and Wagley’s high by a bit less than Robinson’s. “Don’t worry,” he told them, “What we’re concerned about are people that are really falsify- ing their hours.” Cota said he could give them an example. He then showed them the paysheet of an employee who claimed 8 hours on a project when the computer showed that he had worked only 4 hours. Robinson could see that the employee’s name was Russell Lund. (Although not controlling, there is no evidence that Lund worked on Geiger’s team.) Robinson testi- fied that Lund is still employed. (3:590–592) Catherine Wagley testified consistent with Robinson. (2:399–401) As mentioned earlier, Cota did not testify. Robinson and Wagley testified credibly. Gallup offered no rebuttal or explanatory evidence and does not address this matter on brief. The May date for the incident suggests that the Cota conver- sation possibly occurred before the union activity became open. Even if the event took place in late May, and even if Lund, not shown to be an open supporter of the Union, was openly neutral respecting the union matter, possible disparity exists because of the prominent activity of Janice Rinehart in supporting the Un- ion. Falsification of hours is exactly what Gallup fired Janice Rinehart for, so the stated basis of her termination meets the test mentioned by Cota. However, and although the General Counsel does not articulate a theory, the situation does show an apparently more favorable treatment for Lund than for that extended to Rinehart. The Government was not required to go further and adduce evidence showing the reason for no termina- tion of Lund. Possibly there is a reason, such as this was a one- time incident, Lund simply was confused about some numbers, and as Lund was on the 35-hour team and his 5-hour grace period for the week, that justified a warning rather than a dis- charge. Whatever the possibilities, the burden shifted to Gallup to offer that explanation. Gallup chose not to contest the issue. That is its right. In the absence of any explanatory evidence rebutting the Government’s evidence, and having found Robin- son and Wagley credible witnesses, I now infer and find a dis- parity of treatment that was applied unfavorably as to Janice Rinehart. I will address this point later when I discuss the Gen- eral Counsel’s prima facie case. The General Counsel mentions another incident as asserted disparity, this one involving Shari Patty. (Br. at 48) As the General Counsel there states, Patty’s records for June 24 (not June 27) reflect a 1-hour shortfall between the 7.8 actual hours claimed on her payslip (RX 56) for the project VNS W562 and the 6.8 hours shown in her hand notes on that same document. Patty was given credit for the claimed 7.8 actual hours. (RX 55) However, as Gallup correctly observes (Br. at 77-79; Re- ply at 18–20), Supervisor Geiger credibly explained that Patty apparently had thought it unnecessary to show the 1.0 hour that Patty had spent, at Geiger’s request, assisting Geiger resolve sample issues concerning the survey. Added to the 6.8 hours, the missing 1.0 hour brings the total to a match with the 7.8 actual hours claimed. Crediting Geiger, I therefore find no disparity in the Shari Patty example. For the Government’s final example of disparity in treat- ment, the General Counsel (Br. at 47) cites the example of Tommy Nowlin, fired on April 24, 1999 for falsification of hours. (GCX 5; RX 12) Some months earlier, when Nowlin was a 7-week employee, he was not fired for similar conduct, but instead received further training in how to report his hours. This was in accordance with an understanding among the su- pervisors to give new (through 7 weeks) employees further education in the pay system. Janice Rinehart, of course, was not a new employee, and I find that such additional training for a new employee does not constitute any disparity compared to no additional training opportunity for Janice Rinehart in June 1999. Nowlin was discharged that April because reports that he was bragging about falsifying his time reports triggered an investigation that discovered such falsification and he was fired the next day. (1:103–107, Uria-Ruiz; GCX 5) The General Counsel’s argument that Rinehart was not bragging about falsi- fying suffers from some disconnection of logic and relevance. The bragging merely served as the triggering event for the in- vestigation. Our triggering event here was Stinson’s call that led to the red flag of June 16. Additionally as to Nowlin the General Counsel impliedly ar- gues disparity from a single line of the exhibit that states, “Tommy was chronically short working the hours that were required of him.” (GCX 3 at 3) The disparity implied here is that Gallup gave Nowlin a pass (not extended to Janice Rinehart) on the requirement that he meet his team’s 15-hour requirement (1:104, Uria-Ruiz), for nothing was done until it coincided with his bragging and subsequent discharge. The problem is that the language in the quoted line is ambiguous and could just as well mean that, in the investigation into the report on the bragging, Gallup discovered that, as well as falsi- fying his hours on the occasion or occasions reported, Nowlin had been doing it for some time (or at least not working the required hours for some time, even if not falsified). Because of the ambiguity, I am unable to count Nowlin’s case as an exam- ple of disparity. g. The Government’s prima facie case Has the General Counsel proved, prima facie, that a moving reason for Geiger’s June 24, 1994 discharge of Janice Rinehart was Rinehart’s union activities? Aside from relying on the Russell Lund disparity to show a prima facie case, and animus DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1274 from the 8(a)(1) allegations (a majority of which I have dis- missed), the General Counsel (Brief at 49-51) argues that Janice Rinehart had never changed “her method”38 of reporting hours worked, and that in the past Gallup was satisfied with Rinehart’s performance because of her high productivity. But, Gallup “began to ask questions about these [reporting] practices only after she became a lead union adherent.” The latter point really goes to the heart of the case, which is whether the call from Project Administrator Tracy Stinson was unlawfully motivated by Gallup. Stinson requested a study of the Adventist project, and Supervisor Geiger printed a copy (RX 52) of the data on that project—Adventist W470. (13:2545–2548; 14:2869–2870) Recall that earlier I indicated that Stinson’s call would have to be discussed later. Later is now. Earlier I indicated that Stinson’s call came after Adventist W470 had just ended, and certainly that is what the testimony states, especially that of Supervisor Geiger when she describes the numbers “running out.” (13:2551, 2553, 2560) However, could it be that Geiger, in speaking about the project ending that last full week, simply meant that it was ending for that month and that it would begin anew in July? The evidence does not answer this question directly, and possibly not indi- rectly. There is one item that I did not mention earlier. On the first page of the sealed exhibit (GCX 13), line item ten is for Access Code 6942,39 “Adventist IP 10/98–12/99.” No one tells us what the time-frame reference means. Could it mean that the study operates, without break, for 15 months from October 1998 through December 1999? Does it mean the contract cov- ers that period, but that the study will be done in separate monthly segments? The first possibility is not correct because we know that in June 1999 the project ended—the question here being whether that was totally or just for 1 month. It therefore appears possible that Adventist W470 may have been a 15-month contractual study providing for monthly study results. If so, then the study was scheduled to resume in July, then August, and so on through 1999. But even if that is so, since the quota was set for the 15-month contract at, as we have seen, five completed surveys per hour (and see sealed GCX 13 at 1), presumably all that Stinson’s call to Supervisor Geiger, for a study on raising the quota, could do would be to collect data either for a renewal contract possibly about to be negoti- ated, or for a proposal to reopen the existing contract. The parties did not address this question at trial, nor have they done so on brief. In any event, my earlier observation still holds. The nature of Stinson’s call asked about doing something to increase the profit margin. The implication of Geiger’s account is that Gallup assertedly wanted more profit now, not in some future contract. Yet after Geiger’s study, no change was made in the quota. Suddenly, after Rinehart’s discharge, Gallup ap- parently had no further need to change the quota. 38 “Her” method is correct. Unfortunately for Rinehart, Gallup’s policy reflects the old saying of an iron-willed boss or parent—“My way or the highway.” 39 Adventist W470. (GCX 16 at 1) The access code is for Gallup’s internal use. (1:75, Uria-Ruiz) But that gives us another question. Did Rinehart’s discharge resolve the issue because that got rid of a strong union sup- porter, or because it eliminated a person who was improperly inflating the number of actual hours, thereby hurting profits? Actually, as the record does not have the profit numbers, either before or after Rinehart’s discharge, I am unable to evaluate the claim of low profits. But as I wrote earlier, I am not required to accept the asserted purpose of Stinson’s call as gospel. In fact I do not. Although some ambiguity exists over the length of the Adventist W470 contract, and whether the project ended per- manently in late June or merely for that month, the stated pur- pose of the call (with no evidence offered as to any increase in hours or drop in the profit on Adventist W470), its timing, and the result of no change in the quota still appear to suggest a purpose different from that which was stated. I can, and do, infer that a different motive was behind Stinson’s call. The next question is whether the surrounding circumstances justify my drawing an inference that the other motive was the unlawful one of ridding Gallup of a Union thorn. The surrounding circumstances include a handful of 8(a)(1) findings, the strong and open union activities of Janice Rinehart, the timing, and the disparity respecting Russell Lund. As to the timing, because the motivation appears to have come from Gallup’s headquarters, presumably the matter took more than a few hours to discuss and to plan. And the organizing pot was still boiling in June (or at least “simmering,” and the Union had already started filing charges), and even into August, fol- lowing CEO Clifton’s late May visit and speech. Nevertheless, although I suspect that is what happened, and that the purpose was the unlawful one, does it rest on speculation as to what happened in Nebraska rather than on a proper inference that such is what happened? Can the other elements (some 8(a)(1), strong and open union activities, and disparity) support such an inference? Are the inferences equal? That is, logically one inference (no improper plan discussed or hatched in Nebraska) is just as strong as the other (the facts logically call for an inference that the top management in Nebraska set out to get rid of the most active union supporters, and Janice Rinehart fit that description, and a possible pretext existed as to her hours, so all that was needed was an excuse to examine her hours)? In Lozano v. Lozano, 52 S.W.3d 141 (Tex. 2001), the Texas Supreme Court (in the majority opinion) thoughtfully articulated the rule about equal inferences as follows: Properly applied, the equal inference rule is but a species of the no evidence rule, emphasizing that when the circumstan- tial evidence is so slight that any plausible inference is purely a guess, it is in legal effect no evidence. But circumstantial evidence is not legally insufficient merely because more than one reasonable inference may be drawn from it. If circum- stantial evidence will support more than one reasonable infer- ence, it is for the jury [trier of fact] to decide which is more reasonable, subject only to review by the trial court and the court of appeals to assure that such evidence is factually suffi- cient. All that is interesting, and although knowing the personal source of any action is satisfying, I find it unnecessary for me GALLUP, INC. 1275 to determine the personal source of any discriminatory motive flowing from Nebraska. All that I need address is the applica- tion of the action, not the personal source. The corporate source is before us all—it is Gallup. And it was Gallup that applied the action. Accordingly, I need only analyze the action to determine whether the Government has demonstrated, prima facie, that a moving reason for the discharge of Janice Rinehart was her open40 and strong support of the Union. The Government has, from me, a handful of 8(a)(1) findings. That is one factor indicating unlawful opposition to unioniza- tion. Another factor is the open and strong support of Rinehart for the Union. Also, union activity was still “simmering,” and the union already had filed charges, when Rinehart was fired. Then there is the example of disparity of treatment when Gallup did not fire Russell Lund in May, yet did fire Rinehart one month later for the same type of asserted misconduct. This combination of factors, I find, generates a prima facie case— not a strong one perhaps, but at least one sufficient to establish a violation unless rebutted. There is another factor, not developed earlier. While perhaps not itself a marker of discriminatory motive (because there is no showing of a departure from past practice), this factor at least provides a supportive atmosphere for the above factors. I refer to the circumstances of Rinehart’s discharge. First, de- spite the fact that Janice Rinehart admittedly was a high pro- ducer, and that normally any management would prefer to re- tain such a productive employee, Gallup did not first call Rinehart in and, as part of a joint and cordial inquiry, ask her how she had arrived at her figures when Geiger had reached numbers substantially different. Supervisor Barlow asserts that, despite the normal preference to retain, the requirement of good ethical job performance trumped Rinehart’s high-producer card. (12:2300–2301) Indeed, Barlow even claims that, in his opin- ion, Rinehart was deliberately cheating because the discrepancy was so large. (12:2302–2304) Although I agree that discharge could be the ultimate result from the foregoing process, what strikes me about Gallup’s procedure is the appearance of an improper motive. After all, appearances count. Indeed, as to federal judges there is a stat- ute that addresses the appearance of impropriety. Recently one federal judge, after telling a newspaper reporter that the case before her was more complex than some other case, found her- self recused by the First Circuit Court of Appeals on the basis that her statement could create, in the mind of a reasonable person, the appearance of partiality. In Re Boston’s Children First, 244 F.3d 164, 171 (1st Cir. 2001). In our case, Gallup’s approach, taken with someone that Gallup admittedly would want to retain absent the discrepancy, was accusatory, not one of perplexed concern. Respecting this topic, recall Barlow’s statement that he had no doubt Rinehart was present 40 hours but that she had worked only 28.7 hours, to which Rinehart said that such (im- plying the latter figure) was incorrect. (13:2678–2679, Geiger) At that point, Geiger acknowledges, Rinehart asked whether 40 Such as putting her signature in the number one spot on the list of 20 names that the Union’s letter (RX 5) of May 16 to Gallup identified as the in-plant organizing committee. she was being fired, and both Barlow and Geiger said yes. Then, Geiger tells us (13:2679), Rinehart turned to Geiger and asked (direct address pronouns substituted) (13:2679): Why didn’t you talk to me about it? Not answering Rinehart’s piercing question, Geiger testified that she said that if Rinehart wanted Geiger to explain how she had arrived at the numbers she could do so, that she was talking with Rinehart about it now, that nothing could be done, it was not personal, and that they had to let her go. (13:2679) As mentioned earlier, Rinehart was in “total shock” at this point. (9:1744) And it was at this point, Supervisor Geiger reports, that the termination notice (JX 9) was presented to Rinehart. (13:2679) Upset and shaking, Rinehart had to sit to read it. (13:2680) This description leads to the conclusion that the nature of the termination meeting was implemental. There was no cordial effort to discover, through friendly cooperation, how manage- ment and a “very productive” (15:3116, Uria-Ruiz) interviewer could be so far apart on their numbers. Supervisor Barlow assumes, based on the size of the gap, that Rinehart was cheat- ing. But as we have seen, any misunderstanding about how to treat quota fills could well lead to a large difference in the numbers. The plain fact is, Rinehart’s termination has all the appearance of a rush to judgment—get the numbers, and get her out of here! Recall my earlier finding that Gallup “jumped the gun” re- specting the termination in the sense that Rinehart had a second week to make up any shortfall in her hours. Thus, if Gallup’s motive had been to determine whether it could retain this high producer, it simply could have conferred with her in a nonaccu- satory atmosphere to ascertain whether, if Gallup’s hours were correct, Janice Rinehart was laboring under an incorrect under- standing of how to report her hours, or whether Rinehart in fact was cheating. If the former, then Gallup could have retained this high producer by correcting her misunderstanding and assuring her that she could make up the shortfall in hours the following week by working overtime. Simple. But the simple was not followed. Instead, double-teaming Rinehart in the emotionally-charged atmosphere of a confronta- tional and implemental setting, Gallup gave Rinehart the “bum’s rush” out the door. For me, this accusatory approach toward a high producer reflects an unlawful motive because it is so contrary to common sense and normal business practice. At the very least, the nature of this episode lends strong support to the markers of discrimination listed earlier. Respecting my view of Gallup’s “rush to judgment” con- cerning Janice Rinehart, consider this. The course Gallup fol- lowed of viewing, printing, and analyzing computer printouts, and then accusing and terminating a “very productive” inter- viewer, took the route both of complexity and ambiguity with- out the one additional (but solid and common sense) step of visual observation. Even if this additional step would have added another day or two to the investigation, it would have been well worth it. Visual observation by Supervisor Geiger in just walking by Rinehart’s cube, especially on seeing Rinehart away from her cube, would have given Geiger eyeball proof of whether Rinehart was leaving her screen up while away from DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1276 her cube. (It also may have confirmed that Rinehart was sitting there waiting for a number to drop.) How strange that Supervi- sor Geiger did not chose to include the simplest and most direct procedure of all, and the one method that would provide eyeball proof of what Janice Rinehart was doing (good or bad)—just walk around and look! Gallup had another way of monitoring Rinehart’s work to determine whether, in fact (without all the ambiguity from the computer printouts), Rinehart was cheating. Indeed, Gallup has a “Support Team” that does it every day (or night) as part of Gallup’s regular business operation. Not only is it not a secret, it is part of the training manual. Right there on page 19 of the ITM. (JX 23) It’s called “Taping and Monitoring,” and it is part of Gallup’s quality control system. It is used in conjunc- tion with the quality evaluations that the supervisors give the interviewers. As noted on page 19, in the paragraph on evalua- tions, Appendix D (JX 23 at 95–98) has a two-page form that the support team or the supervisors use for the various topics in reaching total scores for four general areas. The scores for those four general areas (Attitude, Probing, Interviewing tech- niques, and Coding, with their total of 37 topics for the group of four) are transferred to the evaluation log that we already have seen in, for example, Janice Rinehart’s case (GCX 14, average score of 49.48 for 25 evaluations). The support team tape records and monitors interviewers nightly. (JX 23 at 19) Obviously not every call is tape re- corded and monitored, for the system presumably relies on spot-checking the calls. Aside from the main business use of the system, the process also gets into disciplinary matters (1:134, Uria-Ruiz), such as for “falsifying” (5:939, Zieler) or “cheating” (15:3103, Lucario). A couple of the witnesses in this case were, or are, members of the support team: Patty Collette (12:2414) and Tristan Lucario (12:2264, Barlow; 15:3102–3103, 3107–3108, Lucario). The point is, if Gallup had truly been interested in determin- ing whether it could retain a high producer, it certainly had at least two direct methods of observing and monitoring just what Janice Rinehart was doing—direct visual observation by Su- pervisor Geiger, and taping and monitoring by the support team. That process may have taken another day or two, or even three, before Supervisor Geiger conferred with Rinehart, but the ambiguity and debate would have been eliminated. There is no relevant evidence as to Gallup’s past practice in this regard. (Tommy Nowlin’s example does not qualify.) Perhaps Gallup always reacts in the more indirect, complicated, and accusatory manner (collecting and analyzing computer records and then accusing and terminating without the benefit of eyeball confirmation) rather than in an approach geared to retaining, if possible, high producers who may have reported incorrect figures based on some misunderstanding of Gallup’s complicated compensation system. That is, perhaps Gallup supervisors just never go and look. Accordingly, in the absence of evidence of a departure from past practice, I treat Gallup’s failure to include the more direct and common sense avenues (going and looking, and use of the support team) of obtaining verifying evidence as further support (although not themselves markers) for the markers of discrimination that I listed earlier. Turn now to Gallup’s rebuttal. h. Gallup’s rebuttal For its rebuttal of any potential prima facie case, Gallup ad- duced evidence of other employees terminated for falsifying their hours. The General Counsel does not contest that basis as a proper ground for discharge (1:184), but in effect on brief contends that Gallup cannot overcome the disparity example of Russell Lund. Leslie Bannister (not a Houston employee, 1:10) was termi- nated February 5, 1998 for falsifying quota sheets. “Previous warning [apparently January 21, 1998 per JX 25 at 2 for “in- consistent track work”] was that any addition problems with hours would result in termination.” (JX 25 at 1) Although there is no sponsoring witness (received by stipulation, 1:10, 14) and no testimony interpreting certain notes and attached pages (apparently from an intcheck), Gallup apparently relies on this exhibit simply to show (11:2120) an example of an em- ployee discharged for falsification of hours. For that limited purpose, there is no contest. If anything, the exhibit addition- ally may serve to support the General Counsel’s case in that it possibly shows some disparity of treatment (by Supervisor Catherine Langan), because of the January 21, 1998 warning she gave rather than discharging Bannister then. No such cour- tesy was extended to Janice Rinehart in Houston. (Even so, the General Counsel does not cite or rely on the Bannister case.) Darren Washington, also not a Houston employee (1:12), was terminated (“Falsifying hours”) February 3, 1999 for claiming on his paysheet that he had worked 4 hours on January 30 on a certain Busch survey “when he hadn’t worked on it at all.” (JX 30) Presumably that would be a glaring difference and wholly unrelated to what is properly classified as “actual hours.” I find Washington’s case irrelevant to Rinehart’s. Of no value (beyond the basic fact of discharge for falsifying hours) are the examples of the March 1999 discharges of Jafar Amin and Jessica Mangum. (JX 10, 11; 1:9, 14) There is no sponsoring witness. Mangum’s exhibit has for comments, “Worked 32 Hrs 32 min, claimed 4 Hrs 45 min,” but there are no details that show methodology. Yury Kremenetsky, not a Houston employee (1:12), was fired March 26, 1999 for “Falsi- fying actual hours.” (JX 29) The manager’s written comments tell us that Kremenetsky (JX 29): Falsified payroll for 1 straight week. Left auto dialer running while going to lunch & leaving building. Took hour. Leaving the autodialer running while leaving the building for lunch appears to be much worse than leaving a screen up for an hour with no dials, the latter attributed to Janice Rinehart. Be- yond the remote possibility of repeated mistakes, it clearly would be cheating under Gallup’s pay system. Apparently all that Gallup intends in the Kremenetsky situation is to show the generic category for the discharge. (Br. at 64–65, 73) Vanessa Villareal was terminated by Supervisor Uria-Ruiz on October 26, 1998 for “Falsification of hours.” (RX 68) Villareal claimed 5 hours on her paysheet when she worked only 2 hours 19 minutes. Uria-Ruiz determined this by running an intcheck. No evidence tells us what triggered Uria-Ruiz’s investigation. When Uria-Ruiz called Villareal in, Villareal admitted the matter. Uria-Ruiz then terminated her, and Vil- lareal signed the termination form without qualification. GALLUP, INC. 1277 (15:3088–3090, 3122; RX 68) Although the intcheck method- ology is relevant, we have no details as to the nature of the time gaps. Moreover, as Villareal admitted her action (it is not es- tablished whether Villareal admitted simply to her entries, de- claring a misunderstanding as to the proper procedure, or whether she admitted knowingly doing something improper), her case has little relevance for us because Janice Rinehart denies the allegation of falsification. Payroll Coordinator Mike Cota checks about 10 percent of the paysheets against an intcheck. (11:2202–2204, Barlow) Barlow testified that, based on reports to him from Cota’s spot checks, he has initiated his own investigation. Based on just such a tip from Cota (11:2204; 12:2381, 2385, 2410), Supervi- sor Barlow, on February 8, 1999, terminated Jack Burroughs for falsifying his hours (RX 39) and Ronald Sage for falsifying his pay (RX 40). As to Burroughs, Barlow observed that Burroughs was claiming a large number of hours based on a low number of dials. Barlow then inspected Burroughs’ pay- sheet and the intcheck. Burroughs paysheet claimed work of 5.5 hours on a Lexis-Nexis study, but the intcheck showed only 1 hour 44 minutes. (Indeed, one gap, from line 3 to line 4, exceeds 3 hours following a “screen failure.”) Barlow termi- nated Burroughs. (11:2205–2207; 12:2376–2378; 2396–2397, Barlow; 16:3173, Uria-Ruiz) Burroughs was a high producer, but Sage was not. (12:2396, Barlow) As to Sage, he claimed that he had worked 5 hours on an IBM study and that he had obtained two completes on the study. But Barlow’s investigation revealed that Sage had writ- ten case ID numbers on his paysheet that were stolen from an- other interviewer, and that Sage really had worked only 32 minutes and had obtained zero completes. Barlow terminated Sage. (11:2219–2222; 12:2375–2376, Barlow; 16:3173, Uria- Ruiz) The General Counsel attempts to distinguish the cases of both Burroughs and Sage, the former because the low number of dials (22) for a claimed 5.5 hours made it “easy to see that he had lied.” Similarly, it was easy to see that Sage had lied be- cause of the fabricated case ID numbers and zero completed surveys. (Br. at 47) I disagree with the significance of the Government’s argument. After a tip from Cota, Supervisor Barlow launched his own investigation, apparently using about the same methodology as that followed by Supervisor Geiger as to Janice Rinehart. Moreover, the implication from the record is that Burroughs and Sage, separately, were called in, con- fronted with the results, and terminated. Of course, nothing about Sage’s case shows an honest mistake. However, even as to Janice Rinehart, both Barlow and Geiger were of the view that she had lied. I therefore find relevance to Sage’s case. Burroughs made his inflated claim when he was in overtime status, and the base rate for overtime pay is the total hourly rate made after averaging the first 40 hours. By inflating his hours, Burroughs was running up his pay at the higher overtime rate. Supervisor Barlow was of the opinion that Burroughs did this deliberately. (12:2286–2287, 2396–2397) Recall that Barlow is of the same opinion as to Janice Rinehart’s “cheating.” Ac- cordingly, I find the Burroughs case to be fully relevant regard- less of whether it was easy or difficult to see that he was cheat- ing. On July 1, a few days after Janice Rinehart’s termination, Supervisor Charles Davenport fired Angela Lund for falsifying “actual hours on hourly project which resulted in inaccurate pay—claimed 25.5 hours for the week of 6/15–6/21 & actually worked 13.3 hours as was documented by using the internet [intranet] daily production reports.” (JX 14; 1:9, 14; 11:2121) Supervisor Geiger appears to have signed the termination no- tice as a witness. Indeed, the exhibit itself (JX 14), reflecting different handwriting and ink even on photocopies, plus a com- parison with Rinehart’s termination notice (JX 9), discloses that Geiger filled out almost everything on Lund’s termination no- tice. (The General Counsel does not address the fact or argue that any particular significance should attach to such fact.) The record gives no details of the methodology followed by Davenport, in what way the “reports” (it is not clear whether an intcheck was used) disclosed the asserted discrepancy, or what Lund’s defense, if any, was. (The exhibit reflects that Lund refused to sign.) Although the evidence is insufficient to make a finding, a reference by Supervisor Uria-Ruiz (16:3174) to Angela Lund and “her husband” suggests the possibility that Angela Lund is the wife of the Russell Lund who is mentioned earlier. Respecting the papers for the “past-practice” terminations, I have noted that most do not enlighten us concerning how quota fills, for example, were handled, respecting the paysheets, be- fore June 1999. What these past examples do show, however, is that, when investigating and then terminating for falsification of hours, Gallup (1) investigates by computer records, (2) con- fronts, (3) accuses, and (4) with one exception (Russell Lund) discharges. No effort is made (except for new employees, and that perhaps was limited to a certain period when it was thought by Houston’s supervisors that extra training could be needed) to salvage high producers. Supervisor Barlow explains this tough approach even toward high producers on the basis of the high ethical standards that Gallup expects of interviewers. (12:2300–2301) Geiger explains it this way (13:2627–2628): JUDGE LINTON: Why didn’t you go to her and say: Look, we’ve got all these discrepancies here; do you have any explanation for them? THE WITNESS: Well, in my experience of being an in- terviewer, when you’re off by that much time, there isn’t an explanation, and it really wouldn’t have mattered what she said. She was off 4.8 hours on one project [on one day, Wednesday, June 16] and obviously when I did that [calculation; RX 53] throughout the whole week, there was a difference of 12 hours. I could have gone to her and talked to her [before Rinehart was called in for the con- frontation, accusation, and termination] but the result would have been the same. It’s a company policy and in order to be fair to everyone, no matter how much I like her, I would have go ahead and terminate her. JUDGE LINTON: You’re saying you couldn’t conceive of any business reason that she would have that would make any business sense. THE WITNESS: Right. She was a veteran interviewer, she knew—she knows how to fill out her paysheet. [That is the heart of the issue. If Rinehart knew not to count DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1278 time on quota fills, for example, as actual hours, then she was cheating when she did so.] You know, if it was sim- ply a case of misunderstanding, that would be a different matter, but she was only logged on for 11.4 hours [on that Wednesday] and she claimed 12. You know, she wasn’t even in—she wasn’t even there for the amount of time that she claimed, so there really isn’t a feasible explanation as to why you would claim more time than you were logged on to a project, let alone be off by 4.8 hours—well, more than that [possibly a reference to her generous allowances to Rinehart], but to be off by 4.8 hours on one project on one day. JUDGE LINTON: Okay. Thank you. Encompassing the same view, Supervisor Geiger testified that, even assuming that she had been motivated in part to ter- minate Janice Rinehart because of her union activities (a matter she denies, 13:2677), that based on her calculations, the result would have been the same, “no matter what.” (14:2849–2850). While I accept that testimony, I do so on the recognition that no followup question was asked. Thus, aside from Geiger’s calcu- lations, did she get a call from higher management that sug- gested to her that higher management wanted her to check Janice Rinehart’s hours for the purpose of finding a basis to eliminate a union agitator? We will never know. As to Gallup the company, I find that the force of its rebuttal overcomes the prima facie case of the Government. Had Stinson’s call clearly been shown to have been based on a phony reason, the result could well be different. But the facts as to Stinson’s call, although the suspicion is strong that it was based on an unlawful motive (a motive surely emanating from above Stinson), such a suspicion is based too much on infer- ence and speculation to support a finding of fact. I say that a different result possibly could have resulted because a finding of a tainted purpose for Stinson’s call that launched the review would have added weight to the sole disparity case of Russell Lund. As to that bit of disparity, it is slight and isolated when compared with Gallup’s normal steamroller approach when it finds evidence of falsification of hours. (And for far fewer hours than pertain to Janice Rinehart.) Indeed, no effort is made to save a high producer, such as Jack Burroughs who was a member of Supervisor Barlow’s executive team. Instead, it’s Boom! And out the door they go. Accordingly, notwithstanding my finding that the General Counsel established a prima facie case of a violation, I further find that, once Gallup investigated and found what reasonably appeared to Supervisor Geiger (and also to Supervisor Barlow) to be major falsifications over the course of Janice Rinehart’s 4-day week, June 14-17, 1999, Gallup lawfully was permitted to discharge Janice Rinehart. Wright Line, 251 NLRB 1083, 1089 (1980),41 enfd. 662 F.2d 899 (1st Cir. 1981), cert. denied 455 U.S. 989 (1982), approved NLRB v. Transportation Man- agement Corp., 462 U.S. 393 (1983) (as clarified by Office of Workers’ Compensation Programs v. Greenwich Collieries, 41 Note that, similar to here, Wright Line involved a discharge on the basis of falsifying a timesheet of jobs done. Id. at 1089–1090. But the Government’s prima facie case was much stronger there, and the com- pany’s rebuttal much weaker than here. 512 U.S. 267, 276–278 (1994)). Finding that Gallup has car- ried the burden of persuasion on its affirmative defense, I shall dismiss complaint paragraph 25 at to Janice Rinehart. i. Rinehart’s request for a witness (1) Facts In this orphan location, we have a topic generated by a Feb- ruary 2, 2001 memo (not part of the record) faxed from me to all counsel (including the Union’s lawyer) in which I asked that they submit supplemental briefs responding to a series of ques- tions that I posed based on the Board’s July 10, 2000 decision in Epilepsy Foundation of Northeastern Ohio, 331 NLRB 676. I phrased the predicate with these lines: Recall that when Supervisor Barlow came for Ms. Rinehart, Rinehart (alone of the four) made “numerous” (unsuccessful) requests for a witness to accompany her. (9:1736–1738, Rinehart; 12:2277, Barlow) From 1985 to last July, such a request in a nonunion setting was of no effect. As you know, that changed with Epilepsy Foundation of Northeast Ohio, 331 NLRB [676] (July 10, 2000). The General Counsel and Gallup submitted supplemental briefs stating and arguing their positions on the questions that I had raised, and I have carefully considered the briefs. There is no charge or complaint allegation in our case, either before or after July 2000, nor any motion to amend either, to add an alle- gation that Gallup violated the Act by denying Janice Rinehart’s June 24 request for a witness. Moreover, in its prin- cipal brief, dated September 21, 2000, the General Counsel did not raise the issue. The Government’s position is that Epilepsy applies retroac- tively to Rinehart’s request for a witness, that the General Counsel has not waived the application, that Section 10(b) of the Act does not bar consideration of the issue, that the June 24, 1999 interview of Rinehart by Gallup was investigatory, and that there should be a full remedial order. Gallup opposes each point, either expressly or impliedly. Both parties expressed a preference that I address the matter here and that the record not be reopened. There is no dispute that, on the day of Janice Rinehart’s June 24 discharge, Supervisor Geiger asked Rinehart to come to her office. Rinehart asked whether she was in trouble. (9:1736– 1737) Geiger acknowledges that she replied, “Yes.” (13:2677) Saying that she would be right back, Rinehart went to look for a witness, for she believed that Gallup was “trying to fire all of us.” (9:1737) Recall that Lynne Zieler had been terminated just 2 days earlier. Having some difficulty locating an interviewer who was not on a survey, Rinehart was standing by Sherri Lee’s cubicle when Supervisor Barlow came and informed her that he needed to talk with her. (91737–1738) To Rinehart’s question of whether she was in trouble, Barlow acknowledges that he said, “Yes, there is a concern. Come to our office.” Barlow admits that Rinehart then repeatedly requested that she be able to have a witness come with her. To these “numerous” requests, Su- pervisor Barlow replied, “No. I just need to talk to you.” (12:2277, Barlow) Eventually they arrived at Geiger’s office. GALLUP, INC. 1279 Present in Geiger’s office for the meeting were Supervisors Geiger and Barlow and Janice Rinehart. (Nervous, Rinehart remained standing. 9:1738) Barlow began the meeting by asking Rinehart whether she felt that she had accurately re- corded her hours on the paysheets to the best of her ability over the course of the previous week. (9:1738–1739, Rinehart; 12:2278, Barlow; 13:2677–2678, Geiger) Rinehart initially may have said that she assumed so (9:1739; 11:1975), but she eventually said, “Yes” (9:1740), and Barlow (12:2278) and Geiger (13:2677–2678) recall that Rinehart said, “Yes.” In any event, Barlow proceeded to report some of the numbers that Geiger had reached by which Rinehart was off, in particular the 4.8 hours for Wednesday, June 16 and the weekly total claimed for actual hours of 40.3 against Geiger’s 28.7. Rinehart said that was not true, that she had worked the hours claimed. (9:1740, 1779, 1785; 11:1983–1985, Rinehart; 12:2278, 2281– 2282, Barlow; 13:2678, Geiger) Supervisor Barlow said that Rinehart had received enough training that she should have known what she was doing, and that because of the discrepancy in the hours that they were going to have to let her go. From there the meeting proceeded to the termination notice previ- ously discussed. (9:1741–1742, Rinehart; 12:2282, 2299, Bar- low; 13:2679, Geiger) (2) Conclusions Regardless of whether the Board’s decision in Epilepsy would otherwise apply retroactively to this case, and the Board in that indicates that it would, I find, in agreement with Gallup, that the Union and the General Counsel have waived whatever Epilepsy rights might have applied here. The Board’s decision in Epilepsy issued on July 10, 2000. Aside from some reason- able time to become aware of Epilepsy, neither the Union nor the General Counsel ever raised the case as applicable to Janice Rinehart’s situation. Feeling compelled to “at least solicit” their positions, as I faxed counsel on February 2 of this year, I raised the matter and asked for the positions of the parties. The parties earlier had filed their principal briefs in this case on September 23, 2000—over 2 months after Epilepsy and over 4 months before my faxed memo of February 2 of this year. If this were a limitations matter under Section 10(b) of the Act, then the 6-month limitations period expired on January 10, 2001. It cannot be said that the matter was fully litigated, for there was no legal notice or issue to litigate before the Board’s decision in Epilepsy. In Epilepsy, one theory there of the Gen- eral Counsel was liability under NLRB v. Weingarten, 420 U.S. 251 (1975). See Epilepsy, supra, 331 at 705 (respecting Arnis Borgs). Thus, the General Counsel there was proceeding with Weingarten as an alternative theory of liability. Not so here. In short, I find that any Epilepsy rights here have been waived. Finally, even should it be determined that the General Coun- sel is correct up to this point, I further find that the General Counsel cannot prevail under Epilepsy as to Janice Rinehart because the meeting of June 24, 1999, was not investigative in nature. It was confrontational, explanatory, and implemental. Supervisor Barlow’s question asking Rinehart whether she thought that she had filled out her paysheets correctly was merely a way of identifying the underlying problem. (Granted, in theory Rinehart could have answered, “No. I admit that I falsified my time. Please show me mercy.” That possibility is extremely remote, at least where, as here, Barlow proceeded directly into his explanation and there was no effort made to obtain an admission.) After Rinehart’s affirmative response, Barlow summarized the hourly shortage that Supervisor Geiger had discovered, and then Barlow terminated Rinehart. Neither Barlow nor Geiger ever asked Rinehart to admit to falsification or even to explain how she could have arrived at her figures. Certainly the meeting was longer than it would have taken Barlow or Geiger simply to have handed Rinehart a “pink slip,” but for the reasons the Board mentions in Baton Rouge Water Works Co., 246 NLRB 995, 998 (1979), the explanation, even with the confrontational atmosphere that puts the employee into a “state of shock” (9:1744, Rinehart), is much preferred to no explanation at all. Also, the termination notice (JX 9) that was given to Rinehart to sign says nothing about admitting to the stated basis for the discharge. To the extent that an unqualified signature might be used as an admission, no attempt was made to prevent Rinehart from adding her qualification that, in effect, the discharge ground was a lie made in retaliation for her union activities. (And this is so even though Gallup’s policy, that prohibited tendering a copy of what Rinehart had just signed, shows no class whatsoever.) Because the nature of the June 24, 1999 meeting was merely to confront, to explain, and then to inform Janice Rinehart that she was terminated, rather than to investigate the matter, Janice Rinehart had no right to the presence of a witness at the meet- ing under the cases of NLRB v. Weingarten, 420 U.S. 251 (1975) and Epilepsy Foundation of Northeast Ohio, supra. I so find. Baton Rouge Water Works Co., supra. As there is no pending allegation or motion for me to dismiss, I shall proceed to the next topic. 4. Sherri Lee a. Introduction Recall from my introduction to Janice Rinehart’s case that the description of the general procedures applies to both Rinehart and to Sherri Lee. As with Rinehart, Gallup’s af- firmative defense is that Lee was terminated for “falsification of hours worked.” From that earlier introduction, and from the much earlier description of other events, recall also that Sherri Lee was openly very active in her support of the Union. There is no dispute concerning Lee’s visible support of the Union. Even so, Lee has a knack for delivering seeming inconsisten- cies. For example, at one point (7:1304) Lee reports that she was very “visual” in the office, yet later (8:1674) she tells us that during the organizing campaign she tried to be as “invisible as possible.” As the record reflects, during the organizing campaign, and despite any efforts by her to be the contrary, Lee was quite visible, vocal, and assertive. Hired May 10, 1995, Sherri Lee was fired by Gallup on June 30, 1999. (6:1022) Although the quality scores for Lynne Zieler and Janice Rinehart were very high, Lee’s was a perfect 50. (GCX 15; 1:181–186, Uria-Ruiz) Lee’s quantity also was very high, for she was on the 35-hour team (6:1197, 1240; 7:1277), and she was a hospital “tracker” (6:1217). She also served as a “buddy” or mentor for the new employees. (6:1064, 1223; 7:1304) DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1280 Lee was a member of Supervisor Charles Davenport’s team. (2:230, Uria-Ruiz) About 2 p.m. (12:2312, Barlow) on Wednesday, June 30, 1999, Davenport called Lee into his of- fice. Present, as all agree, were Lee, Davenport, and Supervisor Douglas Barlow. The conversation was brief. Davenport asked Lee how accurate her reported hours were for the week of June 20 through June 27. Lee replied that she always recorded her starting time for the day and for each project, her breaktimes, and her ending times. According to their records, including an intcheck that had been generated, Davenport replied, Lee had claimed 35 hours but had worked only 29.5 hours. Lee then offered to get her paysheets (her copies) to show them. Barlow said that was not necessary because they had already looked at her (original) paysheets plus the intcheck. Lee said that there was no way her time could be off by that much. Davenport responded that Lee was terminated for falsification of hours. Lee then ran out the door screaming that she had been fired. Catching up to her, Barlow told her that she could quietly col- lect her belongings, or they would be sent to her. She said she would do so quietly. As she packed at her cube she tried to show Barlow her time records, but he rejected the offer. It was clear to Lee that the minds of the two managers were closed. Later that afternoon Lee returned to Davenport’s office. Again Barlow was present. Lee signed her termination paper (JX 13), writing the statement, “This termination was unjust. I have kept actual hours recorded on my time sheets.” Lee’s request for a copy of the one-page termination paper was refused. (6:1057–1059; 7:1301, 1354; 8:1635–1636, Lee; 12:2310– 2317, 2398–2399, Barlow) (The foregoing is a composite de- scription that I find to be approximately correct. I have rejected portions of the accounts of both Lee and Barlow, as impliedly indicated by my description. For example, Lee reports that she, in effect, made a speech on this occasion. Even though Lee tended to make speeches at trial, I do not believe that she did so here. Contrary to Barlow’s claim that he did not have a conver- sation with Lee after the termination meeting, I find, as de- scribed by Lee, that he did. Supervisor Davenport did not tes- tify.) At trial Lee concedes that her personal copies of the pay- sheets (GCX 47) have the same totals (aside from one error) that appear on the originals (GCX 18) that she submitted to Gallup, and that neither Barlow nor Davenport knew about the detailed times that she had recorded. (7:1409–1411) To the extent that Lee may have made it clear to Barlow, at Lee’s cube, that she wanted to show him starting and ending times, it seems clear that Barlow, serving mainly as a witness, viewed the situation as closed by the discharge decision which pre- ceded the termination itself. Barlow asserts that Lee’s union activities were not a factor in her termination. He had to qualify this, however, because he is not the manager who made the decision, nor was he present whenever or wherever the decision was made. I fact, he never tells us, if he knows, just who made the decision. Barlow’s function was that of a witness to the termination, although he was informed by supervisors Davenport and Geiger as to the numbers developed when Lee’s paysheets were measured against the intcheck. The Union never came up in his discus- sions with Davenport and Geiger, Barlow reports. (12:2314– 2316) The fact is, no one tells us who made the decision to dis- charge Sherri Lee. We are told what triggered the investigation that led to the intcheck that led to the discharge. As intimated a moment ago, Supervisor Geiger was involved. Indeed, it is Geiger who launched the investigation, generated the intcheck, and prepared an analysis of the hours just as she had done re- specting Janice Rinehart. When she had completed her analy- sis, Geiger turned all the papers over to Supervisor Davenport. (14:2850, 2906, Geiger; JX 28) In a moment I describe the conflicting versions of why Geiger began her investigation, but first an overview of the different approaches. b. Overview At trial the General Counsel did not call Supervisor Geiger as an adverse witness and ascertain from her the details of her involvement in Sherri Lee’s discharge. As Geiger gave at least one pretrial affidavit (GCX 59, not offered; 14:2851), presuma- bly she reported at least a summary of the details. Instead, the General Counsel elected to present Sherri Lee’s version of her actual hours (for the week in issue) as reflected in seven pages of tables, that being the second-corrected (actually the third) version of GCX 46 (7:1254–1259), which contains Lee’s ver- sion of the correct numbers as based on her analysis of the intcheck (JX 17).42 Lee then compared that result with the numbers in her hand copy (GCX 47) of the original paysheets (GCX 18) that she had submitted for the week in issue. (6:1065; 7:1254–1263, 1412–1413) The main difference be- tween Lee’s hand copy (GCX 47) and the original (GCX 18) submitted to Gallup is that, on her hand copy, Lee inscribed the starting and stopping times, as shown by the computer, for her various work entries. The procedure that Lee, aided by the General Counsel (6:1186), followed in preparing her tables (GCX 46) was to add the actual clock hours and then to strip out all gaps and breaks in order to arrive at just the “pure” time. (7:1397–1403, 1412– 1413) Then Lee and the General Counsel added any “buddy” time, plus the asserted 10-minute break for each hour (except breaks not applied to buddy hours, 7:1387), plus 5 minutes per project worked for the startup time required to switch projects. (6:1113, 1126–1127; 7:1402–1404) Early on Gallup objected that what Lee’s chart, or tables, really showed was Lee’s time on the premises, a very different concept from “actual hours” worked. (6:1109–1110) Indeed, recall from the summary of Janice Rinehart’s case that Gallup’s compensation system pays only for time on the telephone. It does not pay for startup min- utes, time devoted to quota fills, revising off the computer, or similar matters. (Buddy time counts toward actual hours worked, but buddies are paid their hourly team rate for their buddy time. 3:547, 556–557, Lambert; 7:1288, 1378, 1379, Lee) Recall also that the ITM suggests that time in Edit mode (JX 23 at 86–87), to correct typographical errors, is still part of the “phone time.” This apparently is what Supervisor Geiger was referring to when she testified that edit time to correct 42 Originally JX 17 had a cover letter of transmittal. (1:116) Later, in part because the page numbering was complicated by the letter, the parties removed the cover letter from the exhibit. (6:1103) GALLUP, INC. 1281 spelling and such is counted on the computer clock as part of actual hours. (14:2877–2878) Here, Sherri Lee candidly tells us, for example, that she in- cluded time devoted to the patient “Trax Fax” as part of actual hours even if took 30 minutes off the clock to complete the form. (6:1249–1250; 7:1271–1272, 1308) She included that time in her actual hours because it was still “doing work for the company.” (7:1273) The testimony on this point started with patient “Trax Fax” forms, and merged into “editing” the sur- veys. (7:1271–1275) Although there is a difference between the two forms, Lee’s testimony here certainly seems to include the patient Trax Fax forms as part of her editing process, as other testimony by her confirms. (8:1599–1600, 1681) As mentioned in the summary of Janice Rinehart’s case, all hospi- tal tracks except Beth Israel (Lee’s main hospital track, 7:1267) were on the computers by May 1999. (15:3074, Uria-Ruiz) Although no total time is given in the record, the one item most complained about by both Lee and Rinehart is the time required to handle quota fills, including looking for a supervi- sor. This could take, Lee testified (7:1268), 15 to 20 minutes just to find a supervisor. All this time Lee would consider as work time because she was “taking care of work for the com- pany.” (7:1268) Lee implies, I find, that she included that time in her report of “actual” hours. At the front end of the quota fills problem the interviewer, seated at her computer, waits (as the automatic dialer, perhaps not visible to the interviewer, scrolls through number after number seeking out one that remains good) until a good num- ber “drops” onto her computer screen and the automatic dialer begins calling it. This waiting period frequently would be 5 minutes or much longer. Lee included this time as part of her actual hours. (6:1213–1216, 1246–1247; 7:1415–1417; 8:1496–1501) In short, and as she bitterly complained many times at trial, Lee’s position (largely adopted by the General Counsel) is that Gallup’s intcheck system is flawed because it does not include several work items, such as those just mentioned. (6:1215, 1248; 7:1416; 8:1497, for example.) But as already noted dur- ing the discussion of Janice Rinehart’s case, these complaints describe conditions of employment imposed by Gallup. Inter- viewers are free to accept, free to leave, or free to ask a union for help. They are not free to record matters on their paysheets as part of actual hours when those matters are not included within Gallup’s pay plan (RX 4). That brings us to the falsification issue. During the summary of Janice Rinehart’s case, I described how the different hourly teams are paid, and listed an example of the different pay for members of the different teams who obtained completes on the same project, with the lower teams making less than the 35- hour team. This part of Gallup’s pay system not only is basic, but it is easily understood. Under this system, an interviewer who pads her actual hours so as to maintain her membership on the 5-hour team, damages Gallup financially. This is so be- cause a person who fails to meet her 35 hours (70 hours over the 2-week pay period) can be bumped to a team that pays less per complete. (JX 23 at 22, example 2; GCX 34 at 66, opening paragraph and example 2) In short, any such fraud could cheat Gallup out of money that it would not have to pay in future weeks if Gallup elected to bump the interviewer, correctly re- porting fewer than 35 hours, to a lower team. In this connection, and bearing on credibility, I find that, as of the trial, Sherri Lee was well aware of the financial advan- tage attached to being a member of the 35-hour team. (Indeed, in addition to the greater pay rate assigned to the team, Lee was aware that 35-hour team members qualify for participation in Gallup’s 401(k) and insurance programs. 7:1311–1312) Not- withstanding Lee’s knowledge on this point, when I (struggling at the time to understand whether Gallup would suffer anything more adverse than garbled numbers if Lee submitted incorrect figures), Lee coyly, I find, suggested to me that Gallup would not suffer any adverse economic consequence because the in- terviewers are paid on their completed surveys (true), and not on their hours (misleading and therefore false). (7:1274–1283) Thus (7:1278–1279): JUDGE LINTON: Well, if you had gotten more money than you were entitled to, that would hurt the company. THE WITNESS: Yes, sir. JUDGE LINTON: Now, how does this hurt the company that you reported more hours, but it doesn’t have anything to do with the pay? THE WITNESS: Well, that’s the premise of this case is we’re saying that it doesn’t hurt the company; that it did not happen, and they used it only because I was—used that information because we were trying to form a union. They wanted to get rid of me. That’s the only reason—basis of this particular situation. JUDGE LINTON: All right. Well— THE WITNESS: That’s why they said I falsified my time. JUDGE LINTON: All right. Well, I was trying to find out—is the company hurt financially aside from the fact that, well, this messes up their records? THE WITNESS: No, sir. JUDGE LINTON: All right. Then the item is messed up is not the money but their records are different from what your records that you turned in on hours worked? THE WITNESS: Yes, sir. JUDGE LINTON: All right. Well, we’ll see what the company says. If that’s the same view they have of the situation. Moments later (following several questions by the General Counsel that were not calculated to elicit an answer to my ques- tion) Gallup’s counsel had a chance to remind me (7:1283) that on the first day of the trial (5 months earlier) Supervisor Uria- Ruiz had described (1:47, 87) the adverse economic conse- quence of lower pay if bumped to a lower-hour team. Rather quickly after counsel’s statement, and in response to a direct question (that should have been asked sooner) by the General Counsel of whether she would earn less money if demoted to the 25-hour team, Lee answered, “Yes, I would.” (7:1284) I consider Sherri Lee’s answers to my questions as testimonial dissembling. Return now to the tables (GCX 46) that Lee and the General Counsel prepared based on the intcheck (JX 17). The approach taken in these tables is to track, on a daily basis for the week in DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1282 issue, Lee’s work chronologically rather than by project. Thus, when at times Lee’s work on, for example, Beth Israel W693, was interrupted and she had to switch to, say, CSAT F040, and later that shift return to W693, Lee’s tables follows her course of work. (See that specific example, GCX 46 at 5.) By con- trast, the intcheck (JX 17) is arranged by date and then by pro- ject (account) listed alphabetically by the letter prefix of the code number for the project. Thus, MCI A909 will be listed before CSAT’s F040 and Beth Israel’s W693 because the letter “A” comes before the letters “F” or “W.” And even though Lee’s first project on a given day may have been Beth Israel (as on Friday, June 25; GCX 46 at 5), the intcheck puts MCI first (JX 17 at 17) because “A” comes first, even though Lee worked MCA last on that shift (GCX 46 at 5). Because of the difference in sequence, no doubt the Gov- ernment could find Lee’s tables helpful in understanding things, or, as phrased by the General Counsel (8:1564), helpful “for ease of following.” Nevertheless, as I there observed (8:1563– 1564), I will need to follow the steps taken by Gallup. This is so because, as the issue is Gallup’s motive, I have to take the same route followed by Gallup in order to determine whether there is, or is not, something about the route followed that bears on motive. Thus, if the numbers that Supervisor Geiger arrived at were proved to be wrong because of the route taken, and the Government’s route were to result in the correct numbers, the only relevance the Government’s route would have would be whether it sheds any light on whether Geiger’s choice of a bad route was linked to a bad motive. Accordingly, I find Lee’s tables (GCX 46) of little value here for two reasons. First, the numbers are skewered because they are infected with Lee’s (and the General Counsel’s) version of what Gallup’s pay system should include, rather than what work Gallup’s pay plan (RX 4) actually pays for. Second, and as just discussed, I need to follow the investigative steps taken by Supervisor Geiger. Turning back to the intcheck (JX 17) for a moment, I note that the document consists of 27 pages (only about half of the last page is used), covering the week in issue, with most of the pages having 57 line items each. About 5 pages cannot be counted because the line items for the date of Tuesday, June 22, 1999, appear in doublets. A corrected intcheck would occupy about 22 pages, yielding a rough total of about 1250 line items for the week. Turn now to Supervisor Geiger’s investigation. c. Supervisor Geiger’s investigation (1) The triggering event (a) Sherri Lee’s comment The questions are easy. What prompted Supervisor Geiger to get involved? Was it a call from Nebraska? Was Geiger told to find any excuse to investigate the hours of Sherri Lee? Was it a personal vendetta unrelated to the organizing campaign? Or was it something else, something far less conspiratorial and much more mundane? The Government’s generalized conten- tion is that it was pure animus that drove Gallup to pick apart the hours of both Janice Rinehart and Sherri Lee in an effort to discredit them and to discharge them. (Br. at 49) As to Lee, therefore, the General Counsel impliedly argues that whatever Gallup offers as the prompting event is nothing but a pretext. For its part, Gallup relies on the testimonial description given by (former) Supervisor Stephanie Geiger who provides the details of how she, as someone not Sherri Lee’s immediate supervisor, came to investigate Lee’s hours for the week of June 21–27, 1999. Recall that Janice Rinehart was terminated on Thursday, June 24. Supervisor Barlow puts the time of Rinehart’s termi- nation meeting at, roughly, between 3 p.m. and 4 p.m. (12:2276–2277) On her tables (GCX 46 at 4) Lee shows, as the fourth “account” treated, a gap “between surveys” of 2 hours 21 minutes between 18:17 to 20:38 (6:17 p.m. to 8:38 p.m.). Re- minding us that this was the date of Rinehart’s termination, Lee reports that it was during this 141-minute gap when she (dura- tion not specified) was “in with the managers trying to find out why” Rinehart had been terminated. (6:1137–1138) Supervisor Geiger reports that the triggering event came about in this manner. On that June 24, but at some point after Rinehart had been terminated, Geiger was approached by Sherri Lee just outside Geiger’s office. Lee commented to Geiger that if the managers were going to be checking records, then they should check the records of Sarah Lakey and Shari Patty be- cause they were always walking around and taking breaks. (14:2705–2706, 2809, 2834) Alerted by Gallup’s counsel (re- sponding to my question on the cause of the investigation) dur- ing her cross examination to a generalized version of this con- tention (8:1518–1520), on redirect examination Lee strongly denies ever complaining about the reporting of anyone’s hours at Gallup to any supervisor, and especially as to Supervisor Geiger. Respecting Geiger, Lee claims that Lee and others had caught Geiger lying, that they did not trust her, and that she certainly would not have gone to Supervisor Geiger about this. (8:1674–1678) As Gallup stated at trial (8:1520), and the supervisors testi- fied (2:228–230, Uria-Ruiz; 12:2316, Barlow; 14:2724–2725, Geiger), Gallup’s practice is to check the records of both the person accused and the accuser. That is what Geiger did here. (14:2724–2725; JX 28) Lee’s complaint is the only thing that caused Geiger to check Lee’s hours. (14:2834, 2892) At this point I observe that the evidence clearly is consistent with a complaint having been lodged by Sherri Lee. Lee was at work that June 24 and she reports that, for some duration of time between 6:17 and 8:38 that evening, she met with the managers in an effort to ascertain the basis for Rinehart’s dis- charge some 2 hours or so earlier. On the other hand, there is no evidence disputing Lee’s assertion that she had never previ- ously complained to management that an employee may be falsely reporting hours worked. Turn now to Geiger’s exami- nation of the hours worked by Sarah Lakey and Shari Patty on June 24. (b) Sarah Lakey When Geiger came to work the following afternoon (June 25), she pulled up the screen for Lakey’s production sheets. Reflecting the hours claimed for Lakey’s pay, this screen showed that Lakey had claimed only 2.6 actual hours for June 24. Geiger then pulled up the intranet screen showing Lakey’s time and projects that she had worked on June 24, the day on GALLUP, INC. 1283 which Lee had complained. The intranet screen showed that Lakey had been logged-on to six projects on June 24 for a total of 7 hours 57 minutes (7.9 hours). As Geiger testified (14:2713), the two screens reflect that Lakey took some lengthy breaks. Geiger did not print out copies of the two screens on June 25, and did not do so until November 1 on the eve of trial. Geiger asserts, however, that the data remains on the computer for many months and that the November printouts (RX 48) correctly reflect what she observed on June 25. Overruling the General Counsel’s objection, I received the two-page document (RX 48) in evidence. Based on these screens (showing more hours logged than claimed), Geiger saw no need to investigate further. The General Counsel offered no evidence of any claim by Lakey of more hours than she was actually logged on. Al- though Geiger did not look very far, I see no disparity in Gei- ger’s handling of Lakey’s records. (c) Shari Patty The issue as to Shari Patty concerns an apparent 1-hour shortage in time worked (6.8 hours on VNS) on June 24 com- pared with the time claimed (7.8 hours; RXs 55, 56) that was bridged, as described by Supervisor Geiger, by Patty’s having spent an hour working with Geiger on June 24 helping to re- solve survey sample issues (client had labeled departments incorrectly) with the VNS study that Patty had been working on.43 (14:2718–2719, 2885, 2890) As Geiger concedes (14:2720–2722, 2891), none of the records mentions this meet- ing. Geiger testified that, in checking Patty’s records on June 25, she, at that time, remembered the meeting of the day before. (14:2721) The General Counsel argues that Geiger simply “concocted a story that she had an independent recollection, approximately one year later” for the purpose, the General Counsel implies, of covering for Patty, while doing extensive “reaching” to investi- gate active union adherents. (Br. at 48–49) Countering, Gallup observes that the independent recollection occurred on June 25, 1999, and therefore was not something concocted “one year later” (that is, when testifying in April 2000). (Reply at 18–20) Shari Patty did not testify. So the issue boils down to a credibility resolution. Observ- ing that Supervisor Geiger is not contradicted by any witness or record on this point, that her account is plausible and is consis- tent with the records, that I have credited her in other areas, and that she testified persuasively here, I credit her account of the one-hour meeting with Shari Patty on June 24. (2) The “red flag” Having run computer checks on June 24 (on Janice Rinehart) and June 25 (Sarah Lakey and Shari Patty), Supervisor Geiger reached Sherri Lee on June 26 and started checking her hours of the previous day, June 25. (14:2725–2726) At this point the testimonial description of what sequence Geiger followed be- comes a bit garbled. Geiger’s account could have been more cohesive had the lawyers asked her to give the sequence rather than handing her documents and then asking questions about 43 For example, calls were made to male interview respondents ask- ing how they had enjoyed their stay on the OB/GYN unit in the hospital maternity department. (14:2719, 2890 Geiger) the documents. (And no effort was made on cross examination to obtain the sequence.) Of course, it could well be that, as of the trial, Geiger no longer could recall in detail the sequence that she had followed. In any event, I find that Geiger proceeded as follows. Con- trary to her elicited version suggesting that she first called up Gallup’s intranet screen (that was the first document, RX 57, presented to her in this series, 14:2726), I find that, for her first step, Geiger called up Lee’s production reports (RX 58) that would show June 25. For that date, five entries are shown (14:2733), with three being projects. The first item, a project, is MCI Residential. (RX 58 at 1) Other records (GCX 18 at 6; RX 57) show that the project, or study, number is A909.44 The production report (RX 58) showed that Lee had claimed 2.7 (actual) hours worked for that date on MCI A909. Geiger’s second step, I find, was to call up Gallup’s intranet. Entries on the intranet, even though for one date, are not listed in chrono- logical order. (14:2730) Although we now know that the intranet screen or screens had several entries for that date for Lee (14:2730, 2739–2740), it is clear that Geiger (following up on the first entry on Lee’s production report, RX 58) was look- ing only for the MCI entry. Finding that entry, Geiger immedi- ately recognized that she would have to do an intcheck for the entire week because the logged-on time was only 2.1 hours, or .6 hour short of the claimed amount. That discrepancy served as the “red flag” that caused the further investigation with the intcheck. (14:2732–2733) At that point Geiger did not check other intranet entries for that date because she knew that she was going to run an intcheck. (14:2739–2740) The date of June 26 being a Saturday, and with Gallup’s week running Mondays through Sundays, and as she was not eager to start analyzing an intcheck anyway, Geiger decided to wait until Monday, June 28, to print out an intcheck so that it would cover the full week. (14:2733–2735) When that Mon- day arrived, Geiger probably, as step three, printed the intranet document that we have as RX 57, showing entries for Lee’s projects, with hand notes in the margin showing the dates for June 21 through June 27. For some reason that Geiger does not recall, other than just overlooking the other intranet entries (14:2729–2730), MCI A909 is the only entry shown for June 25. The data is accumulated for a printout document off the intranet by copying the entries and then pasting them into a Microsoft Excel document, and that is what Geiger did to create RX 57. (14:14:2730, 2738–2739) One possible explanation for the single entry on RX 57 for June 25 is that Geiger may have created the document (it bears no date) on that Saturday in or- der to copy and paste the MCI entry that she was tracking. (14:2739) When Monday came and Geiger pasted the other dates on the document, she forgot to copy and paste the other entries from the intranet for June 25. Regardless of what the real sequence was, the important document is the intcheck (JX 17) because that is the document, and not the intranet document (RX 57), on which Lee’s discharge is based. (14:2740, Geiger) 44 Because Geiger was searching via the computer, it is highly unlikely that she first, or even second, pulled Lee’s original paysheet (GCX 18 at 6) because that is a paper record. Moreover, Lee’s pay- sheet lists the two other studies before MCI A909. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1284 Before turning to the intcheck, I take us on a short diversion. I will make it short because no party raised the question at trial or on brief. The question is whether Geiger’s procedure here is consistent with that which she followed in Shari Patty’s case. I find the short answer to be yes. Patty was logged-on (telephone time) for 6.3 hours on the VNS project as shown by the intranet printout (RX 55 at 2). That would be .5 hour short of the 6.8 hour claimed on VNS. This gap did not raise a “red flag” for Geiger because, as she explains, Geiger assumes the addition of .1 hour per hour worked on the project as the allowed grace period for breaks. That would take Patty’s time on the project over the 6.8 hours claimed. (14:2885–2886) But that addition would not help Lee here because Lee’s logged-on time for the MCI A909 project barely exceeded 2 hours, meaning the addition of no more than .2 hour for breaks. Even if the .2 were added, that would bring the logged-on total to just 2.3 hours, still .4 hour short of the claimed time. When the gap exceeds .1 hour, Geiger considers that to be “signifi- cant.” Anything over the .1 hour is, in effect, a “red flag” that triggers further investigation. (14:2887–2888) The only re- maining question is whether Lee could, in effect, borrow time from other projects that day, or whether the determination is made on a project basis. Although the record contains a lot of testimony about the grace time for breaks, some of it is general- ized. In any event, Supervisor Geiger testified that the time is figured on a project basis. (13:2537–2538; 14:2777, 2783, 2903) I find no disparity as to this matter. (3) Supervisor Geiger’s chart At step four Geiger ran an intcheck (JX 17) for the week of June 21. (14:2731, 2734, 2739) As mentioned, this was printed on Monday, June 28. (14:2735) Using the intcheck (JX 17) to compare with the actual hours claimed by Lee in her paysheets (GCX 18) for the Week of June 21, Geiger, as she did in Janice Rinehart’s case, prepared a one-page handwritten chart (RX 59) showing the date, project, actual hours claimed, and the hours that Geiger would allow. (14:2741–2742) As I did in Rinehart’s case, I now reproduce, in the same modified format that I used earlier, the chart (RX 59) that Geiger pre- pared to show the results of her study of Sherri Lee’s hours. None of Geiger’s figures are changed, but I have added addi- tional data. Note, however, that some of the logged-on times were not given at trial. I have supplied numbers, the ones marked by an asterisk, based either on my own computation or on the fact that Supervisor Geiger made no deduction. As to the latter, at trial Gallup simply moved on to the next project without having Geiger calculate the logged-on hours. With this explanation, I display Supervisor Geiger’s enhanced chart (RX 59): Date Project Actual Hours Claimed Hours Logged- on Geiger Allows June ‘99 Geiger Allows at Trial June 21 MCI A894 1.5 1.3 1.1 1.0 MCI A895 1.2 1.3* 1.3 1.3 CSAT F040 1.0 1.0* 1.0 1.0 Isuzu R528 .7 .7* .7 .7 Totals 4.4 4.3* 4.1 4.0 June 22 MCI A909 2.5 3.2 2.1 2.1 CSAT F040 3.5 3 8 3.1 3.0 Totals 6.0 7.0 5.2 5.1 June 23 MCI A909 3.1 3.0 2.6 2.3 Beth Is. W693 2.1 6.2 1.5 1.1 Totals 5.2 8.2 4.1 3.4 June 24 MCI A909 1.1 1.2* 1.2 1.2 CSAT F040 1.3 1.2 1.1 .8 Beth Is. W693 1.1 1.1 1.0 1.0 Totals 3.5 3.5* 3.3 3.0 June 25 MCI A909 2.7 3.2 2.1 2.0 Beth Is. W693 3.2 7.0 2.4 1.9 CSAT F040 2.2 4.6 2.0 1.3 Totals 8.1 14.8 6.5 5.2 June 26 MCI A909 .2 .1* .0 .0 CSAT F040 2.0 2.0* 2.0 2.0 Isuzu R528 2.9 2.9 2.7 2.5 Beth Is. W693 .8 6.1 .5 .6 Totals 5.9 11.1* 3.2 5.1 June 27 Beth Is. W693 2.3 2.2 2.1 2.0 Totals For Week 35.4 51.1* 30.5 27.8 As a note, it is undisputed that Lee also had 3.5 hours of buddy time for the week. (GCX 18) For our purposes here, the 3.5 hours buddy time is not relevant because buddy time, al- though added to actual hours, is not computed as is actual hours because it is straight time paid at the team hourly rate. Also, GALLUP, INC. 1285 while buddy time is considered in determining whether the interviewer completed the required 35 hours per week for the 35-hour team, for example, the issue here is whether Lee falsi- fied her hours, not whether she worked enough hours to remain on the 35-hour team. (4) Supervisor Geiger’s adjustments As an initial point, recall from the earlier “Overview” that at trial I alerted the General Counsel I would need, in the decision process, to follow the documentary path taken by Supervisor Geiger. After all, as I there observed, Geiger did not have be- fore her the Lee-Gant tables (GCX 46) that were prepared after the discharge. (8:1563–1564) Despite that alert, on brief the General Counsel provides no review of the intcheck45 (JX 17) used by Geiger and no analysis of a single adjustment made by Geiger. Instead, relying globally on the Lee-Gant tables (GCX 46; a document I have found to be of little value, in part be- cause it is infected with the false assumptions by Lee and the Government as to what constitutes compensable work items), and on Lee’s annotated copy (GCX 47) of her paysheets (an- other document that Geiger did not have before her), the Gen- eral Counsel simply argues (Br. at 46) that Lee had some 38 hours or so of actual hours for the week. In any event (Br. at 50), Lee’s hours are within the 5-hour grace period allowed for discrepancies with an intcheck, and therefore Lee “should have suffered nothing more than a warning.” By simply citing the Lee-Gant tables, without showing on brief how they are not infected with the viruses that I have mentioned, much less of- fering a critique of the intcheck times and Supervisor Geiger’s adjustments based on some of those times, the General Counsel has simply abandoned the field of battle. Turn now to Supervisor Geiger’s adjustments—44 by my count. Recall from my discussions, in Janice Rinehart’s case, under “Preliminary terms and considerations” and under Super- visor Geiger’s “interpretations,” that knowledge and experience are basic to Supervisor Geiger’s review and analysis. However, while some of the analysis can rely on objective factors, a large component of the analysis consists of Geiger’s subjective as- sessment (based on her knowledge and experience) of each line item that she examined. Even if there were no union involved here, and this were a kind of arbitration, or an audit by some standards committee of the polling industry, there would be a problem with the standard to be applied as to each line item category—such as completes, specific callbacks, and any cate- gory of calls where Sherri Lee, or any interviewer, speaks with someone at the number called. Who is to say how long each call should take? Gallup, of course, but has Gallup ever done so? Assume for the moment that both Sherri Lee and Supervisor Geiger are in good faith about one of the completed calls. Sherri Lee may tell us that the call took 30 minutes. As Lee no doubt is the intelligent and articulate person that she appears to be, she could well sound convincing. But just as convincing, and just as intelligent and articulate (without all the talk), would 45 Other than to criticize the intcheck as not including several work tasks and as being inadequate for the purpose of ascertaining why a given call might consume more time than average for the category. (Brief at 41–42) be Supervisor Geiger who might tell us that the typical time for a complete on that study is 10 minutes. Each could be in good faith because, as Geiger acknowledges in Rinehart’s case (13:2612), there can be some long-winded survey participants. As I mention in summarizing Janice Rinehart’s case, in Gei- ger’s experience such surveys are “few and far between,” and after “a certain point” the interviewer simply must end the call. (13:2613) Perhaps we should keep in mind, however, that in her days as an interviewer, Geiger apparently was a high achiever. This is so for she tells us that it was a bad day for her when she attained no more than the day’s quota on her projects. (13:2545) Such a superior background might well foster ele- vated expectations of others, particularly in the first months of Geiger’s experience as a supervisor. And if that were not enough, add to Geiger’s superior abili- ties a bit of self-assuredness to help Geiger set the standards and expectations to which she might hold interviewers. That combination could translate into a real difference when Super- visor Geiger evaluates the numbers. What we need is an inde- pendent check of some kind. As Gallup apparently does not exercise direct visual observation, or monitoring, as part of an investigation, we have to look elsewhere. MCI’s call records serve in that capacity to the extent that they were subpoenaed and offered as evidence. Turn now to see what these records of MCI tell us about a few of the calls made by Sherri Lee and about some of the deductions that Geiger made. (5) Sample deductions Start with the first project for that Monday, June 21,1999, that being MCI International A894. (GCX 18 at 1; RX 57; JX 17 at 1) This call reflects some of the central differences be- tween the parties in Sherri Lee’s case. The intcheck (JX 17 at 1 line 1) shows that Lee’s first call ended at 15:37 (3:37 p.m.) and took 250 seconds (4 minutes 10 seconds, or 4.2 minutes) for a non-specific callback. In this category of call, the person answering, either soon or late, indicates a willingness to do the survey at a later time or date, but without setting a specific time. The length of the call varies, Lee asserts, with the person reached, and can be affected by such factors as whether a child answers the telephone initially, whether there is a language problem, and by other factors. (7:1435–1437) Recall that the intcheck does not show the start of a call, or when the screen goes up. That is supplied by the process of moving from the call on the line before (when the interviewer is in the middle of a project). When it is between projects, we (or the witness) find the close of the last project. If it is an inter- ruption of a project, then the time showing the closest match is found where the interviewer pulled out to work on a different project before returning. We are lucky here because A894 was a short project of 54 line items, all on the first page of JX 17, and showing an ending time of 16:57, or 4:57 p.m. Thus, the intcheck shows that the study was “up” for 1 hour 20 minutes, or stated differently, that Lee was logged on for 1.3 hours. (14:2744, Geiger) Lee disagrees. Although Lee agrees that her work that day on the study ended at 4:57 p.m., because that is the exact time she took off the computer and wrote in her notes (7:1419; GCX 47 at 1), she disagrees with the starting time. According to her notes, the DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1286 starting time, which she would take from the computer screen, was 3:28 p.m. (15:28), or 9 minutes earlier than the completion time of the first call (screen “up” for 250 seconds) shown on the intcheck. (6:1195, 6:1212–1213, 1215; 7:1416–1417) Dur- ing the first part of that 9 minutes (before the number dropped and she began her 250-second non-specific callback), Lee spent nearly 5 minutes sitting in front of her computer monitor, wear- ing her headset, and waiting for the computer to reach someone so that she could talk to the person. (6:1213–1214; 7:1417) As to these first 5 minutes of waiting on this first call, Lee makes it very clear that she counted that as part of the time that she claimed as actual hours. Thus, by Mr. Shultz on voir dire (6:1214): Q. And you consider that actual time worked? A. If I’m sitting with the headphone on my ears wait- ing on someone to speak to me with a number, I — what else could you call it, sir? Q. And do you get paid for that time? A. Not according to [Joint] Exhibit 17. I’m not up moving around the room. I’m not up talking to anyone. I’m sitting at my desk trying to work, and I’m not getting any surveys. I certainly would consider that time worked. Moments later Lee reaffirms that she would include in her claim of actual hours any time that she would be sitting in front of her computer, wearing her headphone, and waiting to talk to a survey participant. (6:1216) But as we saw much earlier in the summary of Gallup’s pay plan (RX 4), Gallup does not count such time waiting, even while wearing a headphone, as time eligible for time claimed as actual hours. Only “dialing time” is eligible for a claim of actual hours under Gallup’s pay system. Employees who do not like this part of Gallup’s pay system are not at liberty to exercise self help to “amend” Gallup’s pay system on their own initiative. As Lee’s time for this project was 1 hour 29 minutes (3:28 p.m. to 4:57 p.m.), she claimed 1.5 hours for her time and sub- mitted her claim for that on her paysheet for the date. (GCX 18 at 1; GCX 47 at 1; 6:1195; 7:1420) But Supervisor Geiger, figuring the logged-on time as begin- ning with the completion time, of the 250-second call, of 15:37 to the study’s end at 16:57, or 1 hour 20 minutes, allowed only 1.3 hours as the logged-on time. (14:2744) Now why did Gei- ger allow zero time for the 4 minutes 10 seconds (250 seconds) that the screen was “up” on this non-specific callback? The answer is that Geiger gave no credit on the basis it would take no more than a minute to do a non-specific callback. (14:2744, 2894) Because a single minute would be wiped out by the process of rounding to the nearest tenth of an hour, Geiger would simply start Lee’s logged-on time at the 3:37 completion time of the first call—the one for 250 seconds. (14:2743–2744) As the nearest tenth for the call of just over 4 minutes is .1 hour (6 minutes), this 250-second call is an example of how Geiger, at least at the trial, strictly applied her own standards of effi- ciency. By not allowing the full 4 minutes that the screen was up, Geiger in effect made her first deduction, even though she declined to call it an adjustment, when she refused to back up the starting time for the project by the 250-second duration of the time that the screen was up. This failure reduced Lee’s logged-on time by .1 hour. When that .1 hour deduction is added to the .1 hour extra time that Lee added at the beginning, we see how, as between Lee and Geiger, they have a difference on the logged-on time of .2 hour. Recall that the logged-on time is important, because that (and not the time claimed) is the span of time to which Geiger makes her adjustments (additions or subtractions). Perhaps Geiger would have done it differently had her inves- tigation included a face-to-face meeting with Sherri Lee. Lee would have showed Geiger her personal copy of the paysheet (GCX 47 at 1), containing her personal notations of the starting and ending times as given to her by the computer, and that documentation possibly would have persuaded Geiger that Lee in fact was sitting at the computer ready to work for the first 5 minutes and that Lee quite likely did spend the full 4 minutes on the non-specific callback. Gallup, however, apparently pre- fers computers to people for its investigations. This ties to the ultimate question, which is, what was Geiger’s motive, or Gallup’s motive, for declining to recognize that this call could well have lasted the 4 minutes that the screen was up. Beyond the 1 minute that Geiger would allow, the remaining 3 minutes would be a bit short for any kind of break. But then, where is the disparity evidence showing that Gallup, or particularly Gei- ger, normally is nowhere near this strict? Then again, perhaps Geiger would not have been swayed by Lee’s personal notes. Aside from the call in issue, there are five other non-specific callbacks on the study that day. (JX 17 at 1) The average time for the five calls is (Envelope, please): 62.4 seconds. When rounded to the nearest tenth of an hour, per Gallup’s practice, the average time is exactly one minute! Before we jump to any conclusion that Stephanie Geiger is prescient, recognize that Geiger had the intcheck before her and doubtlessly saw these other five and computed their average time. That, plus her own knowledge and experience, resulted in her hard-nosed stand. Perhaps she would have been swayed by a personal interview with Lee, but again, perhaps not. As it is, by following Gallup’s practice of investigation by computer, it would seem rather a stretch to call Geiger’s allowance of only one minute to be unreasonable. Indeed, based on these num- bers, her allowance of a single minute (rounded to zero) seems close to reasonable. At first glance, one might consider it a bit troubling that Gei- ger makes no allowance whatsoever for one of those calls that runs up the average time in most every category. “Average” means that some calls will take longer than others, as she con- cedes in relation to completes. (14:2892–2893). When the cross examination came to this point respecting non-specific callbacks, Geiger explained (14:2894): A. Well, usually if they’re telling you that they don’t have time to do the survey, they wouldn’t sit there and talk to you afterwards. Even though Geiger also concedes that no one knows “with certainty” how long any given callback will take, she still makes it clear that her estimate for non-specific callbacks is one minute. (14:2894) The cross examination did not specifically explore why Geiger did not allow the 250-second call as one of those that, on occasion, will go beyond the average time— GALLUP, INC. 1287 especially since there were five other non-specific callbacks here that, as a group, made the average time. At this point, I simply note that Supervisor Geiger, apparently guided by her experience plus the average time of the other five calls by Lee in this category on that study for that day, adhered to her opin- ion that one minute would be enough. Strict, perhaps, but a stretch to call Geiger’s tenacity unreasonable and therefore suggesting another motive. Finally, remember that Gallup does not recognize, as part of actual hours, the first 5 minutes (before the 250-second call) when Sherri Lee was sitting before her computer waiting for a number to drop. Move now to Geiger’s deduction number 2 (my count)46— that being the first line item on Gallup’s convenience summary (RX 66; 15:3030–3032) for the analysis of nine of Lee’s calls. By Geiger’s count this is her first “adjustment.” The affected line item for deduction number 2 appears on page 1 of the intcheck (JX 17) at line 30, showing there a gap (from line 29 to line 30) of 31 minutes from 16:06 to 16:37, with the screen up for 1872 seconds (31.2 minutes), for a “complete.” This complete came on the same MCI International A894 that the 250-second call did that I was just discussing. At trial Geiger first asserted that she allowed Lee 20 minutes for completed surveys on this study, “And it’s not a 20-minute survey.” (14:2745) Geiger then corrects herself to say that she allowed Lee 13 minutes for this call and, stating that 31 minutes was excessive for the completed survey, made a deduction of .3 hour, or 18 minutes. (14:2745–2746) This call displays the very core of the major—very major—dispute between the par- ties on what really is happening on those calls that exceed, by a significant amount of time, the typical time screen “up” time for specific categories of calls. On this complete Gallup con- siders the extra time excessive and Sherri Lee and the Govern- ment consider it simply as time worked on a tough interview. Recall from Janice Rinehart’s case that when a call ends and is disconnected, the screen stays up a few seconds so that the interviewer can copy her case ID number for the complete. Thus, Janice Rinehart copied her case ID numbers onto her copy of her paysheets, as we saw with GCX 16. The point is, once the case ID number issues, the computer clock (after the screen remains up a few seconds for copying the ID number) stops counting the seconds and the call, as far as Gallup is con- cerned, has terminated. But if the interviewer delays asking for a case ID number, and instead takes a break, then the screen remains “up” and the clock running, even though MCI’s re- cords will show that the telephone connection ended several minutes earlier. The cross-examination (by attorney Shultz) of Sherri Lee was as follows on the topic of whether the time on this call was excessive (7:1422–1424, emphasis added):47 A. That 1800 minutes [seconds]—a lot of times it de- pends on where you were in the study. If you had to drag someone through the study, if you had to listen to chit-chat in order to do this study or abuse to do this study—it de- pended on whether you were making any money. That— 46 All the deduction numbers that I use are by my count of 44 for the number of adjustments made by Supervisor Geiger. 47 With further references at 7:1433–1434 and 8:1516. trying to make money or trying to pay your bills. There were a lot of factors why you would stay on the phone with someone long past the rate that you were getting paid. . . . . Q. Why, if you were trying to pay your bills, would you stay on it longer? [In other words, the question asks, why not cut your losses and move on to the next call that might yield quicker results.] A. Because if I had electricity that was due at the end of the month or next week, then I’m going to talk to God- zilla in order to get a survey done. If he was abusive to me, if I could cajole him and get the survey so if it meant that I’d be on the phone 30 minutes or two hours or what- ever it was to get that survey completed, that’s what I did. It was all based on financial need at this point. Q. [Counsel Shultz next inquires whether it is Lee’s contention that this complete took 1872 seconds when Lee obtained another complete on line 3, same study, in 746 seconds (12.4 minutes).] A. [Lee answers yes, observing that it happened and that not all survey respondents are alike, and that she wishes she could do all surveys in 2 or 3 minutes.] Q. Well, you’d have to agree with me, would you not, that that 1,800 is a long survey? A. Would I agree? Extremely long, and if I did that survey at that minute, it was extremely long and hard on me. Q. If you completed the survey but kept the screen up, would the time continue to run? A. It probably would, but I never did that. I knew oth- ers that did do that, but I did not. JUDGE LINTON: What would be the purpose in doing that? THE WITNESS: The purpose of it is not to get any case ID numbers so you could go to the bathroom, so you could take a break or whatever, and so you wouldn’t lose any time. That would be the purpose. That was something that I never did. I resent the implication. It’s called falsi- fying your time is what it’s called. MCI’s records (RX 61, first line item) tell us this for the call. From the start of dialing at 16:06:49 (6 minutes 49 seconds after 4 p.m.), to the answer pick up at 16:07:00, plus 18 minutes 43 seconds to the termination of the connection (at 16:25:43), there was a total “dialing” time or “phone” time (Gallup termi- nology) of 1134 seconds (18.9 minutes or .3 hour). As the screen was up 1872 seconds (31.2 minutes, or .5 hour), this means that, following the termination of the 18.9-minute call, Lee’s screen was “up” for an additional 12.3 minutes (1872- 1134=738÷60=12.3) until, at 16:37 (4:37 p.m.), she asked for a case ID number and the computer then terminated the call. Gallup contends (Br. at 98) that the MCI record demonstrates that, contrary to her protestation that she would not so falsify her time records, Sherri Lee did exactly that on this call. At trial Geiger’s difference with Lee was .5 hour in that Lee claimed 1.5 hours for her work on the project and Geiger, mak- ing two adjustments (the first, by my count, reducing the logged-on time at the start of the project by .2 hour, and the DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1288 second, by my count, being the just described .3 hour deduc- tion) totaling .5 hour, thereby allowing actual time of only 1.0 hour. (14:2746) But Geiger was a little more strict, by .1 hour, at trial than she was in June 1999 when she allowed 1.1 actual hours. (14:2747; RX 59) The MCI record also shows that Supervisor Geiger was closer to correct on her initial allowance (14:2745) of 20 minutes (.3 hour) for the call, and making a deduction of .2 hour (14:2746). Thus, a .3 hour allowance for the call, as Geiger made initially at trial, hits the .3 hour bull- seye displayed on the MCI record (RX 61) It therefore appears, and I find, that Geiger’s initial allowance at trial of 20 minutes, or .3 hour, is what really happened in June 1999, for the initial trial deduction of .2 hour from the logged-on time of 1.3 hours would yield the figure of 1.1 hours (RX 59) that Supervisor Geiger allowed as actual hours for Lee’s June 21 work on the project. So what conclusions are to be drawn respecting this second item? Vindicated in her June 1999 assessment by the MCI record, Supervisor Geiger, I find, made a reasonable deduction of .2 hour that June. The other factor, rather than a conclusion, is that I will tend to give stronger consideration to Supervisor Geiger’s assessments of the time that it takes for the categories of calls—to the extent that it can be determined that such as- sessments were made in June 1999. (But it will be troubling if she never allows for an infrequent occurrence of a long-winded person who answers the telephone.) By endeavoring to be stricter at trial in her estimates, Geiger may well have overex- tended herself. For the third and fourth examples, take the second and fourth items on Gallup’s convenience summary (RX 66). These are two more non-specific callbacks. By Gallup’s computer, the first one lasted (that is, it remained on the computer screen) 630 seconds, or 10.5 minutes—.2 hour. (JX 17 at 14 line 23) The study is MCI A909 on Wednesday, June 23, and is Geiger’s adjustment number 16. The second one (adjustment 19) lasted 536 seconds, or 8.9 minutes—.1 hour. (JX 17 at 15 line 57) Only the second one appears on Lee’s tables (GCX 46 at 3), and there Lee counts all 9 minutes as actual work time. From her analysis in preparing her chart (RX 59), Supervisor Geiger deducted .1 hour (6 minutes) as to both adjustment 16 (14:2774) and adjustment 19 (14:2789–2790). On cross-examination (which first covered adjustment 19— line 57), Lee confirms her testimony that the call took all 536 seconds. Then (8:1511): Q. Why would that take so long? A. And as I mentioned before and as I’ll say again, it depends on the respondent. Some respondents don’t take any time in order to reach them or to get to another party, to speak to that individual, to convince them to allow you to call back, or the situations varied. So in this particular situation, the amount of time that I spent on that call was to set up a callback at another time; maybe have talked to a child to get through to an adult. And so, yes. I agree that this nine minutes was not unusual for that particular number. At that point the questions move to whether it would be Lee who would be the person calling back. After explaining that, at that late date in the month, she was struggling to make “any amount of money” and therefore would talk with anyone who appeared to be even the possibility of a survey (a “complete”), Lee describes her practice (as we saw in the general description in Janice Rinehart’s case) of making a note of such telephone numbers and then calling them back manually. In this respect, Lee would “bank” the number to call back herself even though, as the number went back into the pool, she had no guarantee that someone else might not call the number first. (8:1511– 1514) Eventually, Lee again confirms that she claimed all 9 minutes as work done for the reasons that she just explained. (8:1521) Before directing Lee to the 630-second call at line 23 (Gei- ger’s adjustment 16), Gallup’s counsel asked (8:1522) Lee to confirm the length of the other five non-specific callbacks (not counting the two long ones in issue here) shown for that day on MCI A909. (JX 17 at 14–15) The average time for the five is 50 seconds, with the longest being 62 seconds (not the 50 sec- onds appearing in Lee’s answer at 8:1522). That exchange carried to the next question about the 630-second non-specific callback. Lee again states that the time required depends on who answers the telephone. (8:1522–1523) As perhaps could have been anticipated, at trial Gallup had an ace in the hole as to each call. Thus, MCI’s record (RX 61, second line item) reflects that the total “phone time” (dialing plus the connection time) for the 10-minute gap at line 23 lasted just 51 seconds. This tells us that the number remained on Lee’s computer screen for 9.7 minutes (.2 hour) after the tele- phone connection terminated. (630–51=579÷60=9.65=9.7) Addressing the category of “no answer,” Lee at one point suggests that, after she enters her code for the call, if the screen remains up it is because there is a “flaw in the system.” (8:1517) Without further development of this possible conten- tion (one not advanced by the Government at trial or on brief), and in light of Lee’s failure to address MCI’s records during her testimony at the rebuttal stage (16:3136–3166), I shall give no weight to this hint of a possible contention. In any event, Lee did not expressly tie it to non-specific callbacks. Respecting the 9-minute gap at line 57, MCI’s record reflects that the entire call, from dialing to termination, lasted a mere 27 seconds. (RX 61, fourth line item) Thus, the number appar- ently remained on Lee’s computer screen for an additional 8.5 minutes (.1 hour under Gallup’s system of rounding), after the telephone connection was terminated. (536–27=509 seconds ÷60=8.48=8.5) Not only does Lee not address, at rebuttal, these “hole-card” aces of Gallup, the General Counsel on brief simply asserts that, in addition to the MCI record’s not being available for the discharge decision, the MCI record would cause only 1 hour of time to be deducted from Lee’s total of 38.4 hours for the week, thereby leaving Lee liable for, at most, a warning. (Brief at 46) There is a twofold problem with the Government’s argument. First, the 38.4 figure is based on the defective count of hours compiled by Lee and the General Counsel, and second, even though the record was not available for the discharge, it clearly serves to support Supervisor Geiger’s analysis of the time in- volved. An important byproduct of the latter factor is that Gei- ger’s credibility receives a boost. GALLUP, INC. 1289 I shall move past most of the remaining entries on Gallup’s convenience summary. (The fifth, sixth, and ninth items are not among those for which Supervisor Geiger made adjust- ments.) The eighth entry, a category for “Spanish interviewer wanted,” JX 17 at 24 line 90, adjustment 41 by my count, ap- pears at first glance to support Lee. On this call, not listed on Lee’s tables (GCX 46 at 6), there is a 13-minute gap at lines 89-90, from 13:46 to 13:59, for a 770-second call (13.2 min- utes) coded as “Spanish interviewer wanted.” (JX 17 at 24 line 90) At trial Lee refused to concede that this call did not require all this time. Lee defended the time on the basis that many factors are involved on such a call, such as a language problem that was not seen as substantial until after several minutes of efforts. Moreover, this study was Isuzu R528, and for this study anyone in the household 21 years of age or older, as a potential purchaser of an automobile within the next 2 years, was eligible to do the survey. (8:1632–1634, Lee) MCI’s re- cord reflects that the call, from dialing to disconnect, lasted 12 minutes 44 seconds, or 720 seconds—12.7 minutes—6 seconds shy of what Gallup’s computer recorded. Despite MCI’s record corroborating the duration of the call, Supervisor Geiger would make a deduction of .2 hour (wiping out any credit for the call because of Gallup’s rounding proc- ess) on the basis that the interviewer can determine win the first minute whether English is spoken. If English is not spoken, the interviewer is to terminate the call and code it for a “Spanish interviewer wanted.” If the interviewer attempts a survey, she does so “at your own risk kind of thing.” (14:2828–2830, Gei- ger) In short, what Geiger is saying is that the interviewers are to avoid wasting time, and that, in Geiger’s opinion, Lee wasted about 12 minutes here. Conclusion. Poor Sherri Lee—even when the MCI record corroborates that she was working hard for Gallup, she still gets penalized. But the issue is not whether you or I would deduct the .2 hour from her logged-on time, but whether Gallup was prompted to do so, in moving part, by unlawful considerations based on Lee’s union activities. Resolving that question re- mains for later. Turn now to an overview of some of the remaining deduc- tions, particularly the no-dispute categories “Answering ma- chine” and “No answer.” As to most of these (and excluding those that begin a survey, where the issue is complicated by that position), there seems to be very little difference. There are a handful of calls coded “answering machine” where Su- pervisor Geiger deducted anywhere from .1 hour to .7 hour. (JX 17 at 10 line 175, a 5-minute gap on CSAT F040 on June 22 (adjustment 5) is an example of the former (14:2752, Gei- ger), and number 36, a 39-minute gap, JX 17 at 18 line 40, MCI A909 on June 25, is the latter; 14:2819) As to most of these, there is little difference. Thus, respecting the two examples, Lee assigns only a minute to the first example (my number 5), and deducts 3 minutes from a stated 4-minute gap (GCX 46 at 2), and subtracts all 39 minutes of the latter (GCX 46 at 6) be- cause she acknowledges that, during nearly all the gap, she, in effect, took a break (6:1163–1164; 8:1618–1619) The half dozen or so calls coded “no answer” (not associated with major breaks for changing studies, or as the first call of a study) show a similar history. The significance of a comparison of these categories is sim- ply this—the parties (as I suggested much earlier) are traveling in parallel universes. The Government is located in the GCX 46 (Lee’s tables) universe, and Gallup is in the RX 59 (Geiger’s analysis chart) universe. In the Government’s universe, the General Counsel and Sherri Lee have attempted to account for all the time by removing gaps when she was not working. But there are two problems with the Government’s universe. One, Sherri Lee’s concept (adopted by the Government) of “actual hours” includes several items (such as 5 minutes for start up whenever Lee begins the day and every time she switches to a different study during her shift; 6:1086–1087, 1122) that are outside Gallup’s pay system. As Lee herself observes (6:1191– 1192; 7:1276), Gallup has a modified piece-rate system. Aside from the pay for the pieces (the “completes”), a team’s hourly rate, based on “actual hours,” is a factor. As summarized ear- lier, only “phone time” is counted as part of “actual hours.” Thus, breaks are expressly excluded from “actual hours” in the training manual. (JX 23 at 24) Instead, a grace allowance of .1 hour (6 minutes) is applied by management when deter- mining whether an interviewer, whose hours are short of those claimed per an intcheck, should be disciplined. One of the perks of membership in the 35-hour team is that the members are accorded an enhanced grace allowance of 5 hours per week. Underscoring the reality that the 5-hour period is not a break allowance is the concept that the grace period is really a disci- plinary zone—landing in it, particularly toward the end, can subject an interviewer to discipline, normally just a warning. This “concept” is substantially blurred in the record. That is not surprising because Gallup seeks to define the .1 hour grace allowance in two inconsistent ways. First, it is a vacuum period for all non-dialing activities, including smoke breaks, restroom breaks, handling quota fills, and the whole list. Second, it is a disciplinary zone. In short, interviewers are free to use it for all non-dialing activities, but if they use what their supervisor con- siders as too much, then they subject themselves to possible discipline. In any event, by adding 10 minutes per hour for breaks to her tables (GCX 46; 6:1087–1088), Sherri Lee potentially increases the amount of time that Gallup allows for the grace period. Thus, if the actual hours claimed are 30, that would be 300 minutes or 5 hours for breaks. That would match the grace allowance for the 35 hour team. But if the interviewer claimed 35 actual hours, then 350 minutes would be over 5.8 hours, thereby exceeding by nearly an hour the extra grace allowance even for the 35-hour team. Also, as summarized earlier, Sherri Lee faults Gallup’s intcheck system for not including various work activities (such as handling quota fills). It is clear that Lee included the time she devoted to at least some of these work activities in her report of actual hours. As we also have seen, Lee’s concept of what should be counted as actual hours, such as sitting at her computer monitor waiting for a number to drop, is incorporated into her tables (GCX 46). In short, by feeding improper data into her tables, that exhibit (GCX 46) is substantially unreliable. The second problem with the Government’s universe is re- lated to the first. By simply submitting and relying on its own (and Lee’s) concept of hours (GCX 46), the Government fails DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1290 to invade Gallup’s universe and to attack the methodology fol- lowed by Supervisor Geiger in compiling her chart (RX 59) that reflects the results of her analysis of the intcheck (JX 17) cover- ing Sherri Lee’s work for the week of June 21 through June 27, 1999. By that strategic decision to stay out of Gallup’s universe, the Government has forfeited any chance it had of uncovering whatever deficiencies there might be in the 44 “adjustments” (subtractions) that Geiger made respecting Sherri Lee’s logged- on time. (6) Methodology questions (a) The 5-hour grace allowance Whatever the correct figure is for Sherri Lee’s hour for the week, there is no dispute that, as is shown on Supervisor Geiger’s chart (RX 59), the number claimed (excluding buddy time) of 35.4 hours, and the number, 30.5, found by Supervisor Geiger to be correct, are not too far apart. (Far closer than, for example, the gap in Janice Rinehart’s case.) Therefore, if Geiger’s methodol- ogy was infected by a substantial error, or by a small error re- peated across many of Geiger’s 44 adjustments, such a defect, if tied to an unlawful motive, could be very important. Other than stating “despite the admission of numerous items of documentary evidence directed at demonstrating that she [Lee] misreported her hours” (Br. at 50), the General Counsel does not pause to critique any of the 44 “adjustments” made by Supervisor Geiger that led to the actual hours, reflected on RX 59, that she allowed Sherri Lee. Moreover, the Government makes no complaint about Gei- ger’s methodology or her arithmetic. When the Government does not complain about any aspect of a manager’s methodology, or arithmetic, it is not the obligation of an ALJ to sift through the record trying to find that which the Government does not com- plain about. Nevertheless, I briefly will address a couple of items. First, it might be asked whether the 30.5 hours that Geiger allowed as actual hours for the week (RX 59) includes the 5-hour grace al- lowance. At trial the General Counsel appears to contend, well before Supervisor Geiger testified, that Gallup’s numbers did not include an allowance for the “breaks,” and that if such allowance were added, the numbers of both parties would be essentially the same. (7:1466–1467) Indeed, at that point (7:1468) we took a short recess to see whether the parties could reach an agreement on the numbers. Unfortunately, the effort was unsuccessful, and we continued through all the arithmetic. Geiger’s testimony perhaps persuaded the General Counsel that Geiger included breaks in her figures. On brief (Br. at 92), Gallup asserts that Geiger’s chart (RX 59) does include the break allowance of .1 hour per hour [which is less than the 5-hour grace period enjoyed by members of the 35- hour team], citing to Geiger’s testimony at 14:2826. Geiger did so testify. (14:2825–2826, 2830, 2843–2847) However, some of her description indicates that, instead of finding the “pure” time and then adding back the .1 hour per hour (5 hours for the week), Geiger added it as she went along by not subtracting for a “ten- minute gap here, a six-minute gap there.” (14:2825) Except on occasion, these grace allowances are not specifically marked in the record. Thus, it is difficult to verify Geiger’s testimony as to this. In Janice Rinehart’s case, Geiger’s foregoing description clearly applies, for there, as she testified, Geiger focused on the major gaps, generally those exceeding 10 minutes. But Geiger gave no such testimony respecting Sherri Lee. And certainly at trial Geiger frequently shaved the time rather close as to Lee. Indeed, at several points Geiger indicated, either expressly or by her figures, that she was being more restrictive than she was in June 1999. (For example, 14:2747, 2758, 2768– 2769, 2791, 2800, 2807–2808, 2815, 2821, 2830, 2845) Unfor- tunately, at trial Geiger was not asked to report, as she went through the intcheck, which of the smaller 6 to 10-minute gaps that she had bypassed in June as part of her process of, in effect, building in a break allowance as she proceeded. On cross exam- ining Geiger, the Government touched on the 5-hour grace al- lowance generally, particularly in reference to discipline (14:2865–2866), but the General Counsel did not focus on Gei- ger’s assertions that her charted 30.5 hours for Sherri Lee in- cluded the 5-hour grace allowance. The fact is, when we look at Geiger’s chart (RX 59), we can- not tell by looking whether either the claimed column, 35.4 hours, or the allowed actual column, 30.5 hours, does or does not include time for “breaks.” From Lee’s testimony at trial, we know that Lee asserts that she included break time of 10 minutes per hour some of the time, assertedly not claiming all in order to make sure she could not be charged with falsifying her time. (7:1396) Lee either would not, or could not, be specific, testify- ing, “Some days I did. Some days I did not.” (7:1385, 1400, 1402) Nor did she record on her personal copy of her paysheets (GCX 47) what days she added the breaks, and by what amounts. (7:1400–1401) Indeed, recall that Lee recorded on her personal copy (GCX 47) the actual minutes, as shown on the computer screen, for each of the studies that she did. She apparently did not include any breaktime because the computer screen would not show that. From the small numbers that she wrote on GCX 47 (some cannot easily be read because she wrote over them, in a heavy mark, to show the number of hours claimed for each pro- ject), it would appear that, at least as to most of the days, she did not add any time for breaks. As the heavy numbers (for hours claimed) are the ones that she transferred to the original pay- sheets that she turned in (GCX 18), it would seem that the “some days” that she added time for breaks may have come in weeks different from the one in issue. But even if we, with some hesitation, were to conclude that the 35.4 hours claimed (GCX 18; RX 59) did not include time for breaks (or if so, it was minimal), Supervisor Geiger does not tell us whether she considered that the claimed 35.4 hours did, or did not, include time for breaks. The relevance of this line of thought is this. Presumably Geiger would want to compare apples with apples. If Geiger thought that the claimed 35.4 hours did not include breaks (and since the claim is not supposed to include breaks, then one would assume that Geiger would proceed on the basis that the 35.4 hours did not include breaks), then presumably Geiger would record her own column, for allowed actual, as not including breaks. On the other hand, if Geiger made break al- lowances as she proceeded, 10 minutes here, 6 minutes there, then her allowance of 30.5 hours would include some time for breaks—more than the 5 hours permitted, by her testimony. But wait. Try a different approach. From the 30.5 hours that Geiger allowed in June, subtract 5 hours. That gives the total of GALLUP, INC. 1291 25.5 hours—rather lower than the restrictive 27.8 hours that Gei- ger reached at trial. Reverse direction. To Geiger’s total allowed at trial of 27.8 actual hours, add the 5-hour grace allowance. That yields a new total of 32.8 hours. One would think that if Geiger’s 30.5 hours did include the 5 hours of grace, then sub- tracting the 5 grace hours would produce the same number that Geiger reached at trial. Or, one would suppose that adding the 5 grace hours to the 27.8 would equal the 30.5. As we see, such thinking would be wrong. Suppose, however, that instead of using the grace figure of 5 hours, Geiger used the standard grace allowance of .1 hour per actual hour. From the trial figure of 28 hours (27.8 rounded to 28), multiply by 6 minutes. That equals 168 minutes, or 2.8 hours. Adding 2.8 hours, as the grace allow- ance, to 27.8 hours equals 30.6 hours—almost a perfect match with the 30.5 hours that Supervisor Geiger reached in June 1999. Go further. Add the 2.2 hours difference (5 hours less 2.8 hours plus 2.2 hours equals 5 hours) to Geiger’s June 1999 count of 30.5 hours. That yields a total of 32.7 actual hours. Closer than before to the claimed 35.4, but still 2.7 hours beyond the grace allowance of 5 hours. Consider this calculation. The General Counsel (Br. at 46), apparently describing Lee’s tables (GCX 46), states that, after stripping out all breaks, the actual hours left are 33.4. Then this (Br. at 46): “Approximately 30 of those [33.4] hours were dial- ing hours.” But “dialing hours” is a synonym that the managers use for “actual hours.” Thus, if the 5-hour grace time were added to the 30 dialing hours, that would yield a figure of 35 hours, still leaving an overclaim of .4 hour (24 minutes). Perhaps Geiger would have exercised some supervisory discretion and simply recommended that Supervisor Davenport, Lee’s immediate su- pervisor, just issue a warning. The problem with this calculation is that it is not linked to the analysis and calculations made by Supervisor Geiger. While all these calculations are interesting, the real problem is that whatever Geiger did cannot be verified at this point because this part of her procedure was not litigated at trial. Thus, even if Geiger gave breaks with one hand in June 1999, she took at least some of them away with the other hand by her restrictive ap- proach (more generous than at trial, but nevertheless more restric- tive than the time claimed by Sherri Lee) to the time that she would allow on the various call categories. Again, however, the question is not fairness by someone’s perception. The question is whether Geiger’s approach was based on an unlawful motivation. In this respect, recall that the MCI record (RX 61), although con- taining only a limited number of calls, rather dramatically sup- ports Supervisor Geiger on some of those calls. (b) Backing up the start time A second item for discussion is this. To counter a complaint by Lee that the intcheck cheats by not showing the starting time of a call, Gallup asserts (Br. at 84–85) that Supervisor Geiger’s entire analysis [presumably the June 1999 version] “included backing up her computation to include the actual start time (14:2743–2845, Geiger).” Gallup’s assertion generally is not so.48 This particular dispute apparently refers to the process of computing the logged-on time, the time that the study is “up.” Recall that the supervisor, Geiger in this case, makes her adjust- ments (subtractions) from the logged-on time, not from the time claimed. For that reason, the logged-on time is very important. Because the study “up” time is very important, the method for computing the “up” time becomes determinative of an important matter. What would spark dispute at the trial (generally more during the cross examination of Lee than during Geiger’s testi- mony) would be Gallup’s tendency, at least at times, to compute the start time from the completion time of the opening call to the ending time of the last call. (No dispute about the ending time.) Although this matter is more emotional than material, I shall address it as briefly as possible. Otherwise the matter could pro- ceed on to higher levels without receiving any attention here. As Supervisor Geiger’s June 1999 chart (RX 59) reflects, she studied 19 projects from the intcheck (JX 17) involved here. A review of the opening calls for these 19 projects shows that most such calls (15 in number) were not subject to being backed up because they averaged only 32.5 seconds in duration. These are, for example, the 30-second calls that produce the coded entry “Answering machine” or “No answer.” A “Busy” is even shorter, such as one (CSAT F040 on June 24) for 10 seconds (14:2796–2797), and another, the last call (Beth Israel on June 27), being for a mere 8 seconds (14:2836) Even the longest of this batch took only 78 seconds, a “Second refusal” on June 21 on the Isuzu R528 project. (Not the “Aces” project as rendered on the transcript at 14:2747:13). The remaining four opening calls are more controversial. Three of the four perhaps can be summed up in this fashion. They are fairly short calls of around 4 minutes—long enough to receive a .1 hour (6 minutes) in the rounding process, except for one thing. The one thing is that Supervisor Geiger is of the opin- ion that such items as callbacks, specific and nonspecific, should not take more than a minute. By disallowing everything over a minute, each call is dropped back into the no-adjustment zone and the start time for the study begins with the call’s completion time. (The completion time is used because that is what appears on the intcheck. If the intcheck were designed so as to show the start time, rather than the completion time (the later could be figured by converting the seconds that are shown, as is done now to find the beginning of the call), it probably would be Gallup feeling the squeeze from the rounding process (many short calls would be rounded to the start time) rather than the employees being the ones squeezed). These three calls are: One, the very first call, the 250-second (4.2 minutes) “Non-specific callback” for MCI A894 on June 21 (discussed earlier in the section for sample deductions). (JX 17 at 1 line 1) Geiger made it clear that she also would have used the completion time, because of the rounding process, in June 1999. (14:2744) Two, the 198 second (3.3 minutes) “Specific call- back” coming on June 25 in Beth Israel W693. (JX 17 at 20 line 1) Completion time, 13:04, used. (14:2802–2803. Three, same for the 210-second (3.5 minutes) “Specific callback” in MCI 48 In the cited 103 pp. of Geiger’s direct testimony on the 19 pro- jects, Geiger at no point asserts that she backed up the start time as to all opening calls. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1292 A909 on June 26. (JX 17 at 21 line 1—a single call) “I did not give her credit for the one phone call.” (14:2831, Geiger) Finally, for the fourth of the longer opening calls, a 17-minute (1024 seconds; 17.1 minutes) “Complete.” (JX 17 at 17 line 1) Because completed surveys on this study take about that much time, Geiger testified, she backed up the start of the call the full time, adjusting the start time by .3 hour, or 18 minutes. (14:2816–2818) With this clarification of the matter, it is time to turn to the next topic. (d) The Government’s prima facie case (1) Primary theory The Government’s primary theory for a prima facie case ap- pears to be that the asserted “trigger,” the claim (based on the testimony of Supervisor Geiger) that Sherri Lee complained about two other employees (Sarah Lakey and Shari Patty), is “patently false.” (Br. at 50) In short, this principal theory of the Government depends on obtaining a finding that Stephanie Gei- ger testified falsely in her claim that Sherri Lee had suggested that management should check the hours of Lakey and Patty. Lee flatly denies. As Geiger has no corroborating witnesses, this is a swearing match between Geiger and Lee, with the Govern- ment, of course, having the burden of persuasion. Apparently in support of the Government’s “patently false” argument, the General Counsel (Br. at 49-50) argues that (1) Gallup was unhappy with Lee’s union activities; (2) no evi- denced that anyone else’s hours have been so scrutinized, and (3) the “red flags” as to others were open and obvious. The first item has no support in any of the 8(a)(1) findings I have made. Thus, no threat or other unlawful statement was directed to or about Sherri Lee. The second ground appears to be mostly inaccurate. For example, in October 1998 Supervisor Uria-Ruiz ran an intcheck on Vanessa Villareal to check her hours, and then fired Villareal when she admitted at least the facts. Thus, there was no occasion for as much detail there as we have here. In February 1999 Supervisor Barlow terminated Jack Burroughs, a high pro- ducer, for falsifying his hours. Barlow’s methodology was very similar to that used here after he saw a “red flag” on Burroughs’ paysheet (high number of hours compare to a low number of dials). As supervisors have used intchecks in the past to check hours and to discharge employees, I find no merit to this ground. Respecting the third ground, no similar “red flag” incident, that is correct insofar as the exhibits are concerned. Thus, al- though the supervisors describe a past practice of checking the hours of any accuser along with the accused, no supporting documentation was offered. Still, the claimed practice is cer- tainly logical (if for no other reason than to preserve morale). Crediting the supervisors as to this, I find that such a practice preexisted Sherri Lee’s discharge. Return now to the face-off between Supervisor Geiger and Sherri Lee. In several areas I have found Geiger to be credible (although I generally do not credit her in the case of Lynne Zieler). Moreover, respecting Geiger’s description of the time estimates that she had made, she dramatically is corroborated by the MCI record (RX 61) concerning her adjustment number 2 concerning the 31-minute call shown on the intcheck. Supervisor Geiger generally made a good impression as a witness in the cases of both Janice Rinehart and Sherri Lee. By contrast, Sherri Lee generally made a less than favorable impression. Although she appears to be intelligent and articulate, as a witness Lee was verbose and tended to argue with Gallup’s lawyer. Even though I recognize that the passage of time was lengthy, I still give some weight to the fact that Lee had no sav- ing explanation for the 31-minute call (Geiger’s adjustment num- ber 2) that was exposed by the MCI record (RX 61) as only a 19- minute call. As for Lee’s September 1999 EEOC charge (RX 33, race, sex, age, retaliation), as I did with Lynne Zieler, I find as a negative Lee’s initial testimony, on cross-examination, that she thought her union activities were the “sole” reason for her dis- charge. (7:1303–1305) Later she corrected that to include the grounds of her EEOC charge. (7:1305–1307; 8:1667–1668) I consider the switch to be a negative factor in the matter of credi- bility. The need to correct that point suggests that the witness, even if unconsciously, tends to recall in a manner selected to support her felt needs at the moment. Zieler’s overall credibility was sound, but Lee’s, as I have mentioned, was a bit unfavorable. In light of the foregoing findings crediting Supervisor Geiger as to the triggering event, I find no merit to the Government’s primary theory that Sherri Lee’s asserted complaint about two other interviewers is a fabrication. I need not determine whether, as Supervisor Geiger asserts (14:2808–2811, 2850) and Sherri Lee denies (16:3164–3166), that, following the discharge of Janice Rinehart, Lee uttered a threat against, in particular, Super- visor Geiger that Geiger overheard. (2) Alternative theory For an alternative theory, the Government argues disparity re- specting Lisa Miller and Russell Lund. (Br. at 50) Recall that the Lisa Miller matter is discussed in Janice Rinehart’s case un- der the heading, “Supervisor Geiger checks others.” I found no discrepancy there, when Supervisor Geiger did not follow a po- tential discrepancy on one track, because Geiger was not investi- gating that track. Because the General Counsel failed to show that Geiger deviated from her normal practice (that she normally follows all potential discrepancies, no matter that they are outside the area she set out to investigate, yet did not do so here), I found no disparity. If the Lisa Miller matter has some arguable rele- vance to Janice Rinehart’s case, it has practically none as to Sherri Lee’s. Respecting Russell Lund, recall that he is the person whose paysheet Payroll Coordinator Mike Cota failed to cover ade- quately. Beverly Robinson and Catherine Wagley, speaking with Cota, observed Lund’s name on the paysheet. Cota had just told the two that the person (Lund) had submitted a paysheet claiming 8 hours when he had worked only 4 hours. Lund is still em- ployed. When discussing the Government’s prima facie case respecting Janice Rinehart, I found that the Russell Lund matter, which ostensibly showing disparity, was only one of several factors the combination of which established a prima facie case. With all other mentioned factors found not applicable, or oth- erwise lacking in merit, the Russell Lund matter—an isolated incident when compared with several other discharges, including those of Vanessa Villareal and Jack Burroughs—would not, standing alone, establish a prima facie case before the NLRB. (If this were a Civil Rights case, perhaps so, for all the plaintiff need do there is to show enough to cast on the defendant the burden of GALLUP, INC. 1293 producing evidence as to a valid reason, not the burden of per- suasion as to an affirmative defense such as we have in NLRB practice. Procedure before the NLRB in an unfair labor practice trial is essentially the same as that which prevails in any contract or tort suit in a state or federal district court. That is, the General Counsel leads off and fires everything the Government has. If the Government’s barrage is strong enough to establish its burden of persuasion on the unfair labor practices alleged, and thereby avoid a motion to dismiss, then the Respondent (who can elect to rest and seek to win on appeal) proceeds and must establish its affirmative defense by a preponderance of the evidence. Strict rebuttal and surrebuttal follow. The Government retains the overall burden of persuasion to establish the unfair labor prac- tices alleged. Thus, and contrary to the suggestion in some re- ported decisions of one or more federal circuit courts, unfair labor practice trials before the NLRB do not employ the three- step procedure of some Civil Rights cases. Stated differently, dances before the NLRB are the Texas Two Step, not the Civil Rights Three Step.) Based on the foregoing, I find that the Government failed to establish, prima facie (that is, persuading by a preponderance of the credible evidence), that a moving reason for Gallup’s dis- charge of Sherri Lee was her activities on behalf of the Union. Even if it were determined that the Government carried its bur- den, contrary to my finding, I still would find no violation of the Act on the ground that Gallup carried its own burden of demon- strating that, regardless of any union finding, it would have dis- charged Sherri Lee for falsification of her paysheets in any event. Accordingly, I shall dismiss complaint paragraph 25 as to Sherri Lee. 5. Patrick C. Snyder a. Introduction Recall that Patrick Snyder was fired on August 6, 1999. Ad- mitting the fact but denying liability, Gallup affirmatively pleads that Snyder “was terminated for making offensive comments to a co-worker of a sexual nature and would have been terminated for this event and prior events based upon their cumulative effect by the Respondent even if he was not involved in union activities.” There is no dispute that Snyder was very active in the union organizing activities. Hired in November 1994, Snyder worked as an interviewer until his discharge. (4:677–678, Snyder) Sny- der’s quality evaluations (GCX 9) averaged 49.82 out of a possi- ble 50. (1:64, 136, Uria-Ruiz) While at Gallup, Snyder received many awards (GCX 35), including “Interviewer of the Year” for Houston North for 1998–199 (GCX 35 at 1). So far as the record shows, until the summer of 1999 Snyder had never had any dis- ciplinary problems at Gallup. Indeed, at Houston he was known as, “The Chosen One.” (4:730) But beginning the evening of CEO Clifton’s May 26 speech to the assembled Houston em- ployees, Dame Fortune began frowning when she looked upon Snyder, for the first and second (Snyder asserts that the second never occurred) of the three incidents leading to Snyder’s dis- charge assertedly occurred the evening that May 26. Incidents One and Two, both involving Supervisor Stephanie Geiger, occurred that evening following CEO Clifton’s speech. Agreeing that the first incident (although disagreeing with much of Gallup’s version) and verbal warning (issued by Supervisor Doug Barlow) happened that May 26 (4:717, 723), Snyder insists (4:733–734) that incident Number Two is a complete fabrication. Incident Number Three, an “event” on August 4, triggered Snyder’s August 6 discharge. On Snyder’s termination notice (RX 44), the incident is described as “offensive comments of the [a] sexual nature.” (12:2266, Barlow) Present at Snyder’s termi- nation were Supervisors Douglas Barlow and Chad Murray and Interviewer Patrick Snyder. Barlow conducted the termination meeting. (4:730–731, 736, Snyder; 12:2263, Barlow) Barlow filled out the “Comments” portion of the termination notice (12:2261, 2268–2269), the text of which reads (RX 44): Based on 2 previous incidences [incidents] that happened on May 26th 1999, 1) Disruption of workplace, 2) Insubordi- nation of supervisor, In addition to the incident on Aug. 4th of offensive comments of the [a] sexual nature, Patrick Sny- der is terminated. At the bottom right of the termination notice, Snyder (12:2261, Barlow) printed, signed, and dated (top and bottom) the follow- ing note (RX 44): There was no warning verbal or otherwise given by Stepha- nie to me on Aug. 4th. It is a complete fabrication. The charges of sexual harassment are deliberately exaggerated. This termination is motivated by my continuing union orga- nizing activities, and is illegal. Snyder can be excused, in the context of his termination, for dating the second incident as having occurred on August 4. (Au- gust 4 is when the complaint was made about Snyder’s com- ments “of a sexual nature.”) The second incident assertedly oc- curred the evening of May 26. Incidents Two and Three (one denied and the other disputed) I describe in some detail below. Incident One I do not describe in detail, as I need not resolve what happened, because there seems to be no dispute that, about 8 p.m. by Mission Control and just outside the break room (RX 38), Supervisor Barlow reacted to a loud altercation between interviewers Snyder and Therron Hume by calling both into his office. Present in Barlow’s office were, besides Barlow, Super- visor Jason Becerra, Hume, Snyder, and Glen Lambert who at- tended on behalf of Snyder, and Supervisor Geiger whom Barlow called in to hear her version of one point. Barlow let Snyder and then Hume each give his version of the altercation. Agreement is near unanimous that, following the reports by Snyder and by Hume, Supervisor Barlow stated that he was giving each a verbal warning. (Only Lambert recalls that no verbal warning was given. 3:543) There is a dispute whether Barlow added that any further disturbance would result in termination, and I address that later. The complaint does not attack the verbal warning issued by Supervisor Barlow concerning incident Number One. Supervisor Becerra is no longer employed with Gallup, having departed in early October 1999. (1:38–39, Uria-Ruiz) There is no evidence that Gallup has an employee handbook containing rules of conduct and the penalties for violations. So far as the record reflects, to the extent that employees are made aware of whatever disciplinary rules Gallup imposes or follows, those rules are reproduced in the Training Manual’s Appendix DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1294 A—“Workplace Policies and Agreements”49 (JX 23 at 61–74) But these are just major policies, such as for no harassment, fam- ily and medical leave, an alcohol and drug abuse policy, and a policy covering use of the computers. Appendix A contains nothing about a prohibition of or penalty for such mundane things as spitting on the floor or even poor attendance. Supervi- sor Uria-Ruiz mentions a rule against “Disturbing the work- place,” but it is not clear from her testimony whether she is de- scribing one of a set of written rules, or simply a tradition or practice. (1:133–134) Thus, when Gallup terminated Snyder for an asserted third incident of disturbing the workplace, the number of incidents may or may not fit some disciplinary schedule that exists somewhere outside the record. b. Overview Respecting Snyder, the General Counsel contends that Gallup “concocted his second warning out of whole cloth and grossly exaggerated the subject of his third warning, an innocuous com- ment made concerning pajamas at a pajama party sponsored at work by Respondent.” (Br. at 40, 51) Expanding as to the latter point, the General Counsel argues (Br. at 37): Respondent’s attempt to justify the discharge of Snyder for the innocuous comment made to Tristan Lucario on August 4, 1999 is patently absurd in view of the widespread sexual innuendo generally omnipresent in the Gallup workplace. This included the virtual invitation for employees to engage in sexual banter which was inherent in holding a workplace pajama party. Thus, Respondent cannot be heard to com- plain when comments are made concerning pajamas at a company-sponsored pajama party. Indeed, such comments were to be expected and were part of the intended “fun” of the occasion (TR 731). [4:731, Snyder] Failure to join the fun and participate actively at Gallup implied that one was not a “team player” (TR 731). Gallup argues that Snyder’s assertion that the second incident never happened should be subjected to “strict scrutiny” because it is the uncorroborated testimony of a witness with an economic interest in winning the case. (Reply at 33) I strictly scrutinize the testimony of all witnesses, and I apply the law that imposes on the Government, as the party having the burden of persuasion on establishing unlawful motivation, to carry that burden by a preponderance of the credible evidence. As for an economic interest, Gallup is not a disinterested party in that regard, and its supervisors have the economic interest of their jobs to ponder as they contemplate their testimony. Even former supervisors, such as Stephanie Geiger, have the double interest of, one, remaining in the good graces of their former employer (Gallup, here) for the purpose of obtaining good references for future employment, and, two, the desire (both from natural desire and for professional reputation purposes related to employment prospects) to see their judgments and actions as a former supervisor found to be proper. 49 At one time the title for Appendix A may have been “Legal Agreements,” as appears in the Table of Contents (JX 23 at 4), and in a separate exhibit of that name (GCX 34). The latter also has a trade secret agreement and a payroll agreement, and is in the format of policy forms that employees, especially new employees (4:688), sign. But nothing expressly covers insubordination. In short, in these unfair labor practice trials before the NLRB, it is rare to have a witness who is a truly disinterested bystander. Gallup’s argument about that the no corroborating witness fac- tor requires “strict scrutiny” appears to be closely related to the contention that one witness opposed by several should always lose. The Board has disposed of that argument as follows: Moreover, a greater number of witnesses on one side of an issue is not controlling. It is the weight of the credible evi- dence, not the numerical superiority of witnesses, which is controlling. Sahara Coal Co. v. Fitts, 39 F.3d 781, 782–783 (7th Cir. 1994); Riley-Beaird, 259 NLRB 1339, 1367 fn. 115 (1982). The rationale of the two cited cases would also apply to an argument that witness “A” should be disbelieved when he or she is opposed by different (but single) wit- nesses on separate incidents. Parts Depot, Inc., 332 NLRB 670, 703 (2000). (In Riley- Beaird, five managers were arrayed against a solitary welder in a 5 on 1 contest. The welder was credited.) I interpret the forego- ing principle to include the concept that a witness with an eco- nomic interest in the outcome of the case who gives uncorrobo- rated testimony will have the lack of corroboration considered as a factor in the process of resolving credibility, but that such wit- ness is under no credibility taint simply because he or she has no witnesses or other evidence to corroborate his or her testimony. Any other rule removes the demeanor factor from the process of resolving credibility. Thus, in theory, and perhaps it would be in a rare case, a single witness, with an economic interest in the outcome of the case, but with no corroborating witness or other evidence, could be so persuasive against an array of opposing witnesses, that the demeanor factor would persuade the trier of fact. Thus, consider the following quotation from Parts Depot, Inc., id.: In resolving credibility in favor of Vivian Fortin, who testi- fied with apparent sincerity, against her opposing witnesses, including the articulate Peter Bassett, I am reminded of the description of demeanor which Justice Henry Lamm deliv- ered nearly 90 years ago. Writing for the Missouri Supreme Court, Justice Lamm rendered a description of unequaled perception and eloquence: 11 We well know there are things of pith that cannot be pre- served in or shown by the written page of a bill of excep- tions. Truth does not always stalk boldly forth naked, but modest withal, in a printed abstract in a court of last resort. She oft hides in nooks and crannies visible only to the mind’s eye of the judge who tries the case. To him appears the furtive glance, the blush of conscious shame, the hesitation, the sincere or the flippant or sneer- ing tone, the beat, the calmness, the yawn, the sigh, the candor or lack of it, the scant or full realization of the so- lemnity of an oath, the carriage and mien. The brazen face of the liar, the glibness of the schooled witness in reciting a lesson, or the itching overeagerness of the swift witness, as well as honest face of the truthful one, are alone seen by him. In short, one witness may give testimony that reads in print, here, as if falling from the lips of an angel of light, and yet not a soul who heard it, GALLUP, INC. 1295 nisi, believed a word of it; and another witness may tes- tify so that it reads brokenly and obscurely in print, and yet there was that about the witness that carried convic- tion of truth to every soul who heard him testify. _____________________ 11 Creamer v. Bivert, 214 Mo. 473, 113 S.W. 1118, 1120–1121 (1908). Moreover, why should a discharged employee, alone of all the parties in the land, be singled out to carry some additional credi- bility burden? No such taint is attached to parties in a personal or commercial damage or contract suit in State or Federal district court even though they may not have corroborating witnesses. Parties there are not required to come to court wearing sandwich boards bearing the legend “Greedy Plaintiff” or “Evil Defen- dant.” Consequently, and particularly since the statute imposes no such burden, I impose no sandwich-board burden here either on the discharged employees or on Gallup simply because they have an economic interest in the outcome of the case. Instead, under the statute, I strictly scrutinize the testimony of every wit- ness. Of course, the lack of a corroborating witness is a factor to be considered in weighing credibility. But it is not the determin- ing factor. Nevertheless, I so scrutinize, not because some rule requires that I do so, but simply because that process, rather than anything less, helps lead me to more comfortable credibility reso- lutions. When The Congress assembled hath constitu-tionally established the legal burden, let no judge put asunder. As for the third incident, Gallup observes (Reply at 34) that all the Government’s posturing does not change the fact that the triggering incident arose, not from any action by a supervisor, but from coworker Tristan Lucario’s complaint about remarks made to her by Snyder. Testifying nearly a year after the incident, Lucario had reached the advanced age of 19 as of the trial. (15:3103) Thus, as of the incident with Snyder, Lucario was all of 18. (12:2264, Barlow) Notwithstanding her youth, Lucario showed more wisdom on August 4, 1999, than Gallup’s Houston management, for on that date she came to work in street clothes rather than in pajamas,50 thereby prompting comments not only from Snyder (himself in street clothes) but, in more aggravated form, from one David Brown. Actually, the “event” (12:2263, Barlow) was a Gallup- sponsored pajama contest, or “Pajama Jama,” for the stated pur- pose, per the announcement (GCX 36), to help put surveys “to rest in style.” And (GCX 36): We will have prizes for the most surveys put to rest and for the ones who did so in the best style. At the bottom of the announcement, employees were cau- tioned, “Please, Tasteful pajamas only.” The “tasteful” caution underscores what a mature management, exercising common sense in light of all the sexually provocative photographs permit- ted to be posted in various cubicles (GCX 26, for example), would have recognized immediately—such a “pajama jama” was adolescent, unwise, unprofessional, and calculated to generate 50 A testimonial not only to her own maturity and good sense, but a reflection of fine rearing by parents of sound judgment, high moral principles, and good character. incidents that some employees would consider sexual harass- ment. Moreover, such a company-sponsored event was likely to impose on at least some employees the very offensive working environment that is contrary to the standards required of every- one in Gallup’s non-harassment policy statement (JX 1 at 1; JX 23 at 65), and in its sexual harassment policy statement (JX 23 at 68; GCX 34 at internal 65), for avoiding conduct that creates “an offensive work environment.” Finally, a mature management would see that corporate sponsorship of such an event would serve only to compromise the company, and cripple its credibil- ity, if it later were faced with the prospect of having to invoke its anti-harassment policies against employees who might them- selves act immaturely in an environment of bedroom attire in the workplace. Enforcing its rules then, from its compromised posi- tion, would only open the company to the criticism of corporate hypocrisy. Perhaps CEO Clifton will send down a plaque for each Houston manager to post on the wall of his or her office, with a legend reading (as somewhat modified by me, with apolo- gies to St. Paul): When I was a child I would speak as a child, think as a child, reason as a child; when I became an adult, I put aside adolescent behavior. [1 Cor. 13:11.] Nevertheless, the question here is not whether Gallup com- promised itself (it did), or whether its discharge of Patrick Snyder was tinged with hypocrisy (it was), but whether, in terminating Snyder, Gallup was motivated (in moving part) by animus against him because of his strong and open support of the Union. Turn now to the incidents. c. The asserted three incidents (1) Number One—May 26, 1999 (a) “Real” supervisor Although I need not describe all the details of the confronta- tion between Therron Hume and Patrick Snyder, I am satisfied that, in general, Hume was the aggressor as Snyder (and also Glen Lambert, 3:504–505, 540) reports. I also credit Snyder that, when Snyder moved over to the Mission Control desk in an effort to obtain the protective intervention of Supervisor Geiger, Geiger did not intervene and, instead, sat there with “a bemused smile on her face.” (4:718) There is relevance in how this incident began. I credit Super- visor Geiger that, as Snyder approached the breakroom (the Mis- sion Control desk is just outside the entrance to the breakroom; RX 38; 2:345, Uria-Ruiz), she was asking Hume whether he was leaving, and Hume said he did not know. Geiger intended to ask Hume to get them some dinner. (13:2684, Geiger) Hume and Geiger are social friends. (12:2451, Hume) At that point, over- hearing Geiger ask someone if he were going to leave, Snyder admittedly interrupted their conversation assertedly to ask Geiger whether there had been a change in policy, allowing employees to go home. Snyder does not believe that he spoke in a “snide” tone. (4:717) Geiger reports that Snyder asked whether Gallup had made any new policies about employees leaving. She replied no. Geiger tells us that Snyder’s question had a sarcastic tone. She reports that Hume then approached Snyder in the break area (apparently they met at about the entrance) and told Snyder that he had been rude to Geiger and needed to apologize to her. As DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1296 the confrontation progressed, Snyder said that they needed to get a supervisor. Hume replied, “Stephanie is a supervisor.” To this Snyder re- marked, “We need to get a real supervisor,” emphasizing the word “real” in a sarcastic and hostile tone as he said it. (13:2685, Geiger) Hume reports that Snyder asked a question about leav- ing early or (Hume offers alternative versions) stated, “I hope there’s no extra overrides being given.” (Recall the mention in Janice Rinehart’s case that overrides are a formal procedure by which a supervisor has some discretion in excusing a shortage in reaching the required team hours.) Hume then reports that Gei- ger told Hume that lately she had been receiving a lot of rude tones such as that from some of the employees. Consequently, as Snyder exited the breakroom, Hume confronted him in “a very mild tone.” (12:2451–2452, Hume) During the ensuing confron- tation, Snyder remarked that they needed a “real” supervisor. (12:2453, 2459, Hume) In one of his pretrial affidavits (RX 18 at 1, dated August 16), Snyder concedes that, while he does not recall ever telling Geiger that he wanted to talk to a “real” manager, it is “possible” that he had said it. (4:753) This point becomes part of the credibility mix. Thus, on cross-examination, after reading the foregoing statement from his affidavit, Snyder asserts that, on reflection and now with better recall (testifying some 3 months later on No- vember 19), he never made the statement. (4:754) Gallup con- fuses the issue when citing the quote from Snyder’s affidavit, by not differentiating the occasion. (Br. at 112 fn. 19) In this con- nection, it is relevant to note for context that the paragraph im- mediately preceding the paragraph expressing the concession of possibility reads (RX 18 at 1 lines 3–5): On May 26, 1999, I had a confrontation with Therron Hume. I was given a verbal warning for that incident from Doug Barlow. I went home immediately after getting the verbal warning. I had no further contact with Stephanie Geiger that day. I find that Snyder’s pretrial concession of possibility refers to incident Number One, not incident Number Two. Not believing Snyder’s “better” memory at trial, and crediting Hume on this point, I find that, in the confrontation when Snyder exited the break room, Snyder did make the attributed remark about a real supervisor (emphasis in original). (b) The disciplinary meeting Respecting the disciplinary meeting that followed, there is a dispute concerning whether it was reported to Supervisor Barlow that Snyder had made his “real” supervisor remark. According to Supervisor Geiger, she did so report (13:2686), although she does not include it in her August 6 report (JX 21b) of the incident. Snyder (4:723) and Glen Lambert (3:542) each asserts that the item was not mentioned. Hume does not mention it as coming up in the meeting with Supervisor Barlow, and in fact, he reports (12:2460), he was so focused on Snyder that he does not recall whether Geiger was present. For his part, Barlow also denies that the topic was mentioned in the meeting. (12:2250) I find that it was not mentioned. Before leaving the disciplinary meeting in Supervisor Bar- low’s office, I need to resolve one other point. There is a dispute as to whether, in decreeing verbal warnings to both Hume and to Snyder, Barlow added that any repetition of their conduct would result in their termination. While conceding that, at one point, Barlow said that disrupting the office could be grounds for termi- nation, Snyder denies that Barlow put it in terms of a warning of discharge for any repetition. (4:735) Lambert does not address this item because he recalls no warning at all (3:543), while Bar- low (12:2249) and Hume (12:2456) assert the positive. Supervi- sor Geiger does not include the capital punishment aspect when she mentions the warnings at trial (13:2687), although she does so in her August 6 written report (JX 21b) of the incident. The General Counsel contends (Brief at 37) that Geiger’s written report of August 6 (the month, written as a numeral, is a strike- over, with one number apparently being a “7” and the other an “8”) was “concocted as an afterthought” to shore up the allega- tions against Snyder of sexual harassment. If Supervisor Barlow prepared any report of the meeting that he held with Snyder and the others, and describing the disciplinary warnings that he is- sued, there is no mention of that report in the record. (Two or three times at trial Gallup is at some pains to correct the Government’s assertion (4:689–690) that the final basis for Snyder’s discharge was an allegation that Snyder had engaged in sexual harassment. Gallup’s position there (4:691–692, 749) and on brief (Brief at 109; Reply at 34–35) is that the allegation was not processed under Gallup’s sexual harassment policy. Instead, the allegation was treated as a charge of disturbing the workplace by making offensive remarks of a sexual nature. It appears that the main difference in the nomenclature is that the sexual har- assment policy contains certain steps in a formal procedure. By contrast, “disturbing the workplace” is far more informal. More- over, what might constitute sexual harassment under the policy, if the determination were to follow the law of court cases, no doubt would have a less flexible definition than comments “of a sexual nature.” In short, management would have more flexibility, and less formality, by proceeding on the basis of “disturbing the workplace” than in treating the incident under Gallup’s sexual harassment policy, even though some of the concepts are very similar or even identical.) Respecting the disputed point, I credit Patrick Snyder that Bar- low did not expressly issue a capital punishment warning. Even so, this may be a finding of little import. First, Snyder concedes that, at one point, Barlow did say that disrupting the office could be grounds for termination. Second, as I have mentioned, the record contains no rules of conduct and stated disciplinary penal- ties. There is no Gallup rule of record requiring that an employee be given a capital punishment warning in the event of another infraction against managerial serenity. Thus, whether Barlow issued a capital punishment warning or not, it matters not insofar as Gallup’s past practice or rules are concerned. The absence of such a warning possibly will be a factor to consider when dis- cussing motivation. Move now to the next incident. (2) Number Two—May 26, 1999 (a) “Real” supervisor—(Again?) The first issue here is a timing question. Thus, Snyder’s initial testimony implies that he left the facility immediately after visit- ing his cube and signing out—that is, within about 5 minutes of the conclusion of the disciplinary meeting. (4:723; 5:836) Under GALLUP, INC. 1297 this implied theory, Snyder would not have been present after 9:10 p.m. when the second incident supposedly occurred. In- deed, Snyder contends that this second incident never occurred. (4:733) Supervisor Barlow testified that the disciplinary meeting lasted about 20 minutes, concluding about 8:40 p.m. (12:2247, 2250) He is supported in this by Glen Lambert’s intcheck for that eve- ning. (RX 41 lines 307-308; 12:2250–2253) About 30 to 45 minutes after the disciplinary meeting ended, Supervisor Geiger asserts (13:2687–2688), [or about 9:10 p.m. to 9:25 p.m.], she and Supervisor Jason Becerra were eating their pizza dinner at Mission Control when [between about 9:10 p.m. to 9:25 p.m.] Patrick Snyder approached and asked whether he would be given an hour’s exemption from his weekly requirement. Geiger re- plied no, explaining that there was plenty of work to do that week and that the interviewers were required to work their hours. (13:2687–2688) At trial Geiger’s explanation about “exemp- tions” leads to the point in issue (13:2688–2689): Q. What do you mean by exemption or hours off? A. When the workload is slow, sometimes Gallup will give you an exemption which means you don’t need to get your hours in, if you want to take extra time off. Your pay wouldn’t be adjusted because we’re giving you time off. And so he was asking if that week they were going to get hours off. Q. And you responded negatively? A. I said no, there was plenty of work to be done that week and that they were required to get their hours in. Q. And did he say anything to that? A. Yes. Q. What did he say? A. He said, “Are you sure, or do we need to ask a real supervisor?” Q. And how did you respond to that? A. I became upset and I said, “Okay, that’s it; this is the second time you’ve done this tonight, you’re being insubor- dinate, this is your last warning.” Q. What, if anything, did he say? A. Actually, he just smiled and turned away and walked away. Jason Becerra did not testify. Recall that Becerra left Gallup in early October 1999. (1:38–39, Uria-Ruiz). Overlooking this fact, the General Counsel (Br. at 36) requests that an adverse inference be drawn against Gallup from its failure to call Super- visor Becerra to testify. Because Becerra was no longer em- ployed by Gallup during the time of the trial, no adverse infer- ence can be drawn from Gallup’s failure to call him as a witness. Reno Hilton, 326 NLRB 1421, 1421 fn. 1 (1998), enfd. on point 196 F.3d 1275, 1284 (D.C. Cir. 1999). Interviewer Joyce Niehouse asserts that, approaching Mission Control to ask a question, she came upon the conversation, and heard Snyder tell Geiger that he needed to speak to a manager. When Geiger replied, “I am a manager,” Snyder stated, “I need to speak to a real manager.” (Emphasis in original.) Deciding that this was not the place for her to be, Niehouse left. (12:2447) Niehouse acknowledges that she was vocally opposed to the Union. 12:2448) Recall that in Lynne Zieler’s case I found Nie- house not to be a credible witness. I also note that on this occa- sion Supervisor Geiger recalls no one else being present besides the three already named. (13:2688) Moreover, in her June 2 “Incident Report” (JX 21a) that, it appears, Geiger prepared after returning from a 5-day Memorial Holiday that she took (13:2689–2691), Geiger names Jason Becerra as a witness, but not Joyce Niehouse either by name or simply as an unidentified person. Again observing that Niehouse had an unfavorable de- meanor, and in light of the foregoing, and struck by the fact that Niehouse supposedly was so conveniently present and heard only this specific part of the conversation, I do not credit Joyce Nie- house concerning this event. As Snyder was on Supervisor Charles Davenport’s team (5:829–830; JX 22; RX 44), it is not clear why Snyder would not have asked Davenport rather than Geiger. Indeed, as Snyder had left the disciplinary meeting (in which he had received a verbal warning with at least an implied warning from Barlow that any repetition could result in his discharge) no more than a few min- utes earlier, where Supervisor Geiger had stated her view that Snyder’s question to her in incident Number One was voiced in a sarcastic manner (4:722, Snyder; 12:2249, 2389, Barlow), it is very strange that Snyder would ask Geiger about this matter. On the other hand, Davenport was not even present for the discipli- nary meeting conducted by Supervisor Barlow, so perhaps the jurisdictional lines among supervisors and their teams at Gallup are a bit blurred. Stranger still, of course, is the thought that, so shortly after receiving the warning for insulting Supervisor Gei- ger, Snyder would be so foolish as to repeat his insult. And stranger still more is the assertion that Snyder delivered the insult in the presence of a witness, Supervisor Becerra, a witness osten- sibly friendly to Geiger. As for the timing of the occasion, although Snyder’s initial tes- timony suggests that immediately after the disciplinary meeting he packed up, signed out at Mission Control, and went home (4:723; 5:836), and therefore would not have been on the prem- ises for any second incident with Supervisor Geiger, his subse- quent testimony on cross-examination discloses that he had con- versations with several union committee members following his disciplinary meeting (5:837–841, 847, for example). Moreover, Snyder states that when he left the facility and reached the park- ing lot, he met interviewer Rene Ontiveros and his wife who were in their car. They asked if he wanted an escort to his car. (5:846, Snyder) Now it so happens that Ontiveros’ intcheck reflects that his last call for the day ended at 9:04 p.m. (RX 42 at 3 line 124; 12:2253–2255, Barlow) Supervisor Barlow reports that, when he left the facility about 9:10 to 9:15 that evening for the parking garage, he saw Snyder outside the building. (12:2255–2256) It takes about 5 minutes for someone leaving Gallup’s Houston office to reach the parking garage. (11:2158–2159, Barlow) The evidence strongly suggests that Rene Ontiveros left the building about 9:10 p.m., and that Snyder (who encountered Ontiveros in the parking garage) left perhaps a minute or so later for Barlow saw him outside the building about 9:10 to 9:15 p.m. From this evidence it is clear that, from a time standpoint, it is a very tight squeeze to place Snyder at Mission Control at some point be- tween 9:10 and 9:25 that evening. Indeed, as the pizza arrived about 9:10 p.m., even placing the time of Snyder’s asserted visit DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1298 at soon after Geiger and Becerra began their feast, say 9:15 p.m., it becomes a stretch to say that Snyder could have been at Mis- sion Control at 9:15 p.m. when that is the very time Supervisor Barlow saw him outside the building (presumably on the way to or in the parking garage). Accordingly, I find that the timing factor, although not a fully clear factor, tends to favor the conten- tion (4:735) of Patrick Snyder that this incident “never hap- pened”—or, as Snyder also describes the allegation (4:733), “It is a lie.” [Regarding this timing matter, I disregard Gallup’s inadvertent reference to an exhibit (RX 29) that is not part of the record. (Brief at 120 fn. 23) After I sustained the General Counsel’s objection to the offer of this pretrial affidavit, Gallup proceeded to ask questions based on the affidavit, and did not reoffer it. (5:834)] (b) The two documents Following her long Memorial Day holiday, Supervisor Geiger returned to work on Tuesday, June 1. At a meeting of the super- visors that day, she reported this incident and that she had warned Snyder. The told her to document the incident. (13:2691) The following day, Wednesday, June 2, Geiger prepared a document (JX 22) reflecting the verbal warning that she had given to Sny- der the previous Wednesday, May 26. (13:2691) Presumably she also prepared the June 2 “Incident Report” (JX 21a) at the same time. (On this report, she records the time of the incident as being 8:15 p.m., some 30 to 45 minutes earlier than she gave at trial. Of course, that would have been 5 minutes before even the disciplinary meeting began, and therefore it obviously is off base.) Respecting the documented form for the verbal warning (JX 22), the text of the comment section originally had three sen- tences, but the third is crossed out on the exhibit. Geiger con- cedes that she crossed it out, at the August 6 discharge meeting, because it is an incorrect statement. (13:2691–2692; 14:2881) The original text reads (JX 22): Insubordination—he was rude and asked to speak to a “real” manager. He was given a verbal warning. He refused to sign the warning. Now the legal world is familiar with the judicial procedure of correcting an order or judgment “nunc pro tunc” (“now for then”)51 in order that the court’s records will “speak the truth.” At the August 6 discharge meeting, when Snyder was first shown this document, aside from his denial of the incident, he pointed out to Geiger that, because it assertedly was a verbal (oral) warn- ing that was given, there was nothing for him to sign. (4:735– 736) Agreeing, Geiger reached over and scratched out the last sentence. (4:736, Snyder; 13:2691–2692; 14:2881, Geiger) The troubling issue is not so much the scratching out as it is a ques- tion of why Geiger saw fit to write it to begin with. Had the sen- tence been part of the preprinted items, it possibly could be ar- gued that she simply had overlooked its presence. But this is something that she actually had to write out by hand—something that, on June 2, she knew was false. Why would Supervisor 51 B.A. Garner, A Dictionary of Modern Legal Usage 607 (2d ed., Oxford Univ. Press, 1995). Geiger have done that? On cross-examination, the General Counsel, going for the gold, let potential wealth slip through her fingers when she moved on to other topics without following up on her question. The exchange reads (14:2881): Q. And did you ever ask Mr. Snyder to sign that docu- ment? A. No. Q. Why did you originally write on Joint Exhibit 22 that Mr. Snyder refused to sign the warning? A. I had noticed that he had not signed it, and I was mistaken and wrote he refused to sign it, but it was a ver- bal warning; he was not given an opportunity to sign any- thing. Does Geiger’s response simply tell us again why she scratched out the sentence, not why she wrote such a false statement in the first place? Or is she (a person both intelligent and articulate) saying, in a rather inarticulate and indirect man- ner, that at some later date when she inspected the warning memorialization (JX 22), she observed that Snyder had not signed it so she therefore, on that unspecified later date, added the third sentence that, on August 6, she scratched out? But how can that be, for she tells us, as quoted, that she never asked Snyder to sign the document. Moreover, she admits that she has never asked anyone to sign a verbal warning. (14:2881) Indeed, at Gallup employees normally are not even told that such a memorial document is being placed in their personnel file (14:2881), and Geiger did not tell Snyder on this occasion (13:2691) Whatever actually happened here, the appearance of a dissembling answer is troubling. Gallup argues that Supervisor Geiger, having just arrived in Houston on May 17 (13:2467), had no motive to fabricate this second incident. (Br. at 119) Indeed, Geiger observes that, before May 26, she and Snyder had never had any conflict. Therefore, Snyder’s rudeness on May 26 surprised her. (13:2699–2700) In contrast, Gallup contends (Br. at 119–120), Snyder’s motive for expressing his insubordination in this sec- ond incident is that he was upset that Geiger (a young manager and he (roughly 45 to 47 I would say) an experienced and supe- rior interviewer with a cube full of awards—“The Chosen One”) had just rebuffed his request for an exemption of hours. That is, Gallup argues, Snyder wanted the union supporters to be able to leave work (assertedly because they were being har- assed following CEO Clifton’s speech) without being penal- ized. A major problem with this theory is that Geiger merely testified that Snyder asked if “he” was going to be given an exemption. (13:2688) The asserted request did not extend to anyone else. Only in her explanation of what the request meant does Geiger state that Snyder was asking if “they” were going to get an exemption. In her answer to Snyder, Geiger states that “they” would be required to work their required hours. The use of “they” is nothing more, I find, than a generic refer- ence to all employees, utilized in order to show that she was not mistreating him by stating, in effect, that all employees would have to work their required hours. I so find. Gallup suggests (Br. at 113) that Snyder’s inability to re- member who the supervisor was who was at Mission Control when he tendered his paysheet (RX 41) for initialing at his GALLUP, INC. 1299 signing out, or to recognize the initials on his paysheet (4:723; 5:825, 866–868), indicates that Snyder simply is trying to con- ceal the fact that it was Supervisor Geiger, who was on duty that night at Mission Control, to whom he presented his pay- sheet for initialing. Nothing prevented Gallup from establish- ing through Geiger whether the supervisor’s initials on the pay- sheet (GCX 41 at 1) are hers. Although there is some similarity with the initials that she apparently placed on the scratch-out on the verbal warning (JX 22), there are significant differences. For all we know, some other supervisor spelled Geiger at Mis- sion Control while she took a break, and it could have been then that Snyder arrived and obtained that supervisor’s initials on his paysheet. But even if Geiger was the supervisor who was at Mission Control and initialed Snyder’s paysheet, that would support Snyder more than Gallup. Thus, if that happened, it seems a bit inconsistent with Geiger’s story that Snyder came up and stated that he needed to see a manager, then a “real” manager. If all that went on, when did Snyder tender his paysheet and when did Geiger calmly initial it? In any event, I find that Geiger did have a potential motive for preparing false documentation of a warning (an aside from any nefarious mission to eliminate lead- ing union supporters). Thus, recall Therron Hume’s account that, in the first incident, Geiger complained to Hume that Gei- ger had been getting comments of rude tones from several em- ployees lately. Moments later, as I have found, Snyder insulted Geiger by saying that a “real” supervisor was needed. The record possibly suggests that Supervisor Geiger lifted that insult from the first incident and incorporated it into and fabricated this second incident. However, pursuing that sug- gestion would implicate Supervisor Becerra who apparently signed the document as a witness (JX 22). The Government advances neither evidence nor theory as to why Supervisor Becerra would accommodate Geiger by signing, as a witness, something that he knew to be false. Granted, Supervisors Gei- ger and Becerra obviously were on friendly terms as of May 26 even though Geiger had been there fewer than 2 weeks, and they were still on friendly terms even after both had left Gallup’s employment, as the record (14:2906–2908) reflects. Still, Becerra is not named in a single count of 8(a)(1), and there is not a single quote of animus attributed to him. Nor is there any evidence that Becerra and Snyder were on unfriendly terms. Indeed, as Snyder asserts, before the confrontation with Therron Hume, Snyder apparently had not experienced a con- flict with anyone at Gallup because he devoted himself to his work and hardly knew anyone outside his own GTE track. (5:849–850; 16:3161). In short, nothing in the record suggests that Becerra would sign a false document as to Patrick Snyder. Moreover, Becerra also signed the June 2 incident report (JX 21a) that Geiger prepared. Aside from what I already have described, the text of that report concludes by stating that Gei- ger had told Snyder that by questioning her position and being hostile he was committing insubordination. “I told him that if he did this again he would lose his job. This was his last warn- ing.” Snyder asserts that this report, including the “last warn- ing” portion, is all “a lie.” (4:733, 735) (c) Conversation of June 2, 1999 Snyder asserts that, about a week after incident Number One, or about June 2, on impulse he stopped at Mission Control and asked Geiger, the duty supervisor, whether she felt that he had insulted her in the incident with Therron Hume. According to Snyder, Geiger replied “No.” (4:724) Geiger denies. (13:2700) By coincidence, this would be the same date that Geiger drafted her June 2 incident report (JX 21a) and the documentation of the verbal warning (JX 22). A “No” response sounds a bit unlikely. I am not persuaded that this conversation occurred. (d) Meeting of July 8, 1999 About Thursday, July 8, Patrick Snyder met with Supervi- sors Heidi Roberts (the new lead manager), Doug Barlow, and Stephanie Geiger in Barlow’s office. Snyder said that there was an office rumor that he had been targeted for termination, that at the disciplinary meeting Barlow had told Snyder and Hume to write up their versions of the (first) incident, that Sny- der had done so and now wanted to submit his version to be placed in his personnel file. And because he also had heard that his file contained two harassment charges, he requested to see his file and see what the harassment charges were about. After assuring Snyder that rumors are just rumors and that Snyder had done the proper thing by coming to them, Supervi- sor Barlow, joined by the other two, replied that the office had “gone paperless,” and that all personnel records were now in Lincoln (Nebraska). Barlow said that they did not have to show him his personnel file, that the file contained no harass- ment charges to his knowledge, but that he would find out and let Snyder know. Snyder then directly asked Geiger whether she had written up her version of events and placed them in his file and she replied, “No.” He then asked whether there was anything in his file, and Barlow and Geiger each said, “No, there’s nothing in your file at all, period.” (4:724–727; 5:825, 832, Snyder) In their testimonial appearance at trial, neither Doug Barlow nor Heidi Roberts address this meeting (nor does Gallup do so on brief). Supervisor Geiger does so. Geiger’s account is gen- erally consistent with Snyder’s up to a point. The main differ- ences are that, in Geiger’s version, Snyder seeks to ascertain whether there are any written (emphasis added) warnings in his file, and the supervisors assure him that there are no written (emphasis added) warnings in his file. Snyder assertedly was told, however, that his file would contain documentation of “both” events of May 26, and that the documentation was just of the verbal warning, plus now his statement would be in- cluded. (13:2692–2695, Geiger) I am skeptical of Geiger’s version. It sounds so legalistic and so full of propaganda, whereas Snyder’s, as in his delivery, has a natural ring. With such major differences in these versions, it would be nice to have the report of a disinterested bystander. Hooray! We do. It is a tape recording, produced by the General Counsel (4:805, 812, 814) that Snyder made of the meeting. (5:825, Snyder) Unfortunately, neither the General Counsel nor Gallup ever offered this tape recording as evidence, not even during the rebuttal stage. Thus, I am left with a major conflict on the main point of this event, and neither party offers the one piece of independent evidence that would resolve the conflict. So DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1300 which party is more in need of the evidence? The answer, of course, is the party who has the burden of persuasion on the issue—the Government. Accordingly, I draw the inference, adverse to the Government, that had the General Counsel intro- duced the tape and a transcript, that such exhibits would show that, as to the major points of conflict, Supervisor Geiger’s version would be supported rather than the version of Patrick Snyder. At the same time, I draw an inference, adverse to Gallup, that had Supervisors Barlow and Roberts given their reports of the meeting, that such reports would be consistent with Snyder’s version rather than with Geiger’s account. How- ever, I consider the adverse inference against the Government stronger than the adverse inference against Gallup (because the former involves the General Counsel’s failure to offer a tape recording (capturing the event) and transcript). The testimonial difference is not enough to carry this point for the Government. In short, the General Counsel has failed to demonstrate that the July 8 meeting went as described by Patrick Snyder. Snyder also testified that he understood a verbal warning was no more than that, and, being verbal (oral), nothing was re- corded or memorialized and placed in his personnel file. (4:728) That may well be his understanding, but there is noth- ing unusual about memorializations of oral warnings. In any event, at Gallup, Supervisor Geiger asserts, oral warnings nor- mally are memorialized, and employees normally are not told that such written accounts are being placed in their personnel files. (13:2691; 14:2881) If that is so, then why did Supervisor Barlow not prepare a memo memorializing the verbal warning that he gave to Snyder on May 26? In fact, he never did, and it was not until August 6, the date of Snyder’s discharge, that Supervisor Geiger prepared an “Incident Report” (JX 21b) covering the Therron Hume incident and verbal warning. Even so, I decline to find any “invidious motive,” as requested by the General Counsel (Br. at 37), in Supervisor Geiger’s early fail- ure to tell Patrick Snyder that she had prepared a documenta- tion of her verbal warning given for the second incident, along with an incident report for the second incident, and that such items would be sent to Nebraska to be placed in his personnel file. Nothing shows that Geiger departed from her own past practice. (e) Conclusion I find that the Government has not carried its burden of es- tablishing (as part of a pretextual motive theory) that the second incident did not occur. However, I leave open for later discus- sion whether it could be found that Gallup could not have relied in good faith on the second incident. In this connection, any finding of bad faith by Gallup (that is, an unlawful motive) must go beyond any personal animus that Geiger may have developed against Patrick Snyder. Recall that the entire first incident, including the “real” supervisor insult, was all of a personal nature. The evidence suggests that any animus Super- visor Geiger felt against Snyder was based on his personal at- tack on her, and was not directed against him because of his union activities. That also would be true of the asserted second incident. Therefore, the record would have to show that, de- spite this original personal nature of the matter, Gallup know- ingly converted the matter into an antiunion scheme to get rid of Patrick Snyder because of his union activities. At this point the evidence does not get beyond the stumbling block of Su- pervisor Jason Becerra’s presence and his signature on Supervi- sor Geiger’s two documents. Finally, and respecting Supervisor Jason Becerra, It is a mys- tery how the Government could issue and proceed to trial on a complaint as to Patrick Snyder without having or advancing a theory neutralizing the presence of Supervisor Becerra at this asserted second incident This is all the more true if, during the investigation stage, Becerra gave an affidavit reaffirming the two documents (JXs 21a and 22) that he signed as a witness. Simply ignoring Becerra is not a effective litigation strategy. (3) Number Three—August 4, 1999 (a) Gallup’s policies Respecting “offensive comments of a sexual nature” (RX 44, the termination notice), it is relevant to consider Gallup’s poli- cies on inappropriate language even though Patrick Snyder was not discharged under Gallup’s policy against sexual harass- ment. Despite that technical distinction, as we see in just a few minutes, at Snyder’s termination meeting, Supervisor Barlow handed to Snyder a copy of Gallup’s policy against sexual har- assment. (4:729–731, Snyder) Now the record contains two versions of a sexual harassment policy. The older one-page copy, it appears, is or was the form (GCX 34 at internal 65) signed by Snyder when he was hired (it possibly has been su- perseded since 1994 when Snyder was hired). (4:688, Snyder) It is shorter than the one appearing in the January 1999 ITM (JX 23 at internal 68–70), but that mainly is because the current three-page model has extensive coverage of a complaint proce- dure. Then there is the four-page May 20, 1999 Equal Em- ployment Opportunity and Non-Harassment Policy Statement (JX 1) which is broad enough to cover every complaint known to contemporary American society, especially with the more recent addition of harassment about “the union issue” that was addressed much earlier in this decision. Although not entirely clear, it appears that the document that Supervisor Barlow handed to Patrick Snyder at the termination meeting was a copy of the older one-page version (GCX 34 at internal 65), for that is the indication from Snyder (4:688–690, 729, 731–732), and Supervisor Barlow never disputes that sug- gestion. Even under this older form, sexual harassment is de- fined in broad terms, and includes “other verbal or physical conduct of a sexual nature when” “(3)” it creates an “offensive work environment.” However, under the policy, it could be argued that a single incident of less than a blatant nature would not qualify, and that the less than blatant “prohibited acts” would need to be (GCX 34 at internal 65): (2) continual or repeated verbal abuses of a sexual na- ture including graphic commentaries on the person’s body, sexually suggestive objects or pictures placed in the work area that may embarrass or offend the person, sexually de- grading words to describe the person or propositions of a sexual nature. In any event, notwithstanding that at trial the General Coun- sel was expressly proceeding on the theory (as mentioned above in my summary of the disciplinary meeting) that the final GALLUP, INC. 1301 ground for which Gallup discharged Patrick Snyder was sexual harassment (4:689–690), on brief (Br. at 37, 51) the Govern- ment makes no such claim. Even so, it is clear that the third ground for Gallup’s termination of Snyder relies on the concept of “sexual harassment,” for Supervisor Barlow admits (12:2266) that “Snyder’s comment refers to the term ‘sexual harassment.’” It is just that, as I summarized earlier, Gallup wanted the flexibility and informality of proceeding under the rubric of “disturbing the workplace” rather than Gallup’s for- mal and less flexible sexual harassment policy. (b) “Teddies” Recall that Wednesday, August 4, was pajama day, or the day of “Pajama Jama.” Arriving at work in street clothes about 3 o’clock that afternoon, Patrick Snyder, as he acknowledges (4:696–697), commented to four female employees as he pro- ceeded to his cube. To Supervisor Heidi Roberts, wearing a silk top and shorts ensemble, he asked if those were her paja- mas. She answered yes. “That’s cute,” he said. Any reply by Roberts is not recorded. Walking past Mission Control, and observing Supervisor Stephanie Geiger in apparent street clothes, Snyder remarked, “Gee, I’m disappointed; I took you for a teddy girl.” (4:696) Snyder asserts that Geiger merely laughed. (4:696) Geiger reports that she responded, “Knock it off, Patrick.” Geiger filed no complaint over Snyder’s remark even though Snyder assert- edly, “throughout the summer,” had made other unwelcome remarks, including one or more of this nature. (13:2699–2700) During his rebuttal testimony, Snyder does not address these assertions of Geiger. Attaching weight to Snyder’s failure to rebut, I credit Supervisor Geiger’s description. Teddy pajamas, Geiger informs, are a skimpy, one-piece outfit that is “pretty revealing.” She reports that the idea for the “pajama jama” came from the Support Team, which does the taping and moni- toring at the Houston office, on the theory that it would “invoke some spirit in the workplace.” (13:2698–2699) Supervisor Uria-Ruiz testified that it is the retention specialist’s job to make sure that these events “go well.” (1:125) That person’s overall job, Uria-Ruiz advises, is to motivate employees and to keep turnover low. (2:227–228) The record does not show what grade the retention specialist received for this event. Proceeding on, Snyder next met Joanne Passig, a person of mature years dressed like a grandma figure” and who was wearing “bunny slippers.” “Those are cute bunny slippers,” Snyder commented. (4:697) Passig’s reply, if any, is not re- corded. Snyder began work at his desk. As irony would have it, about 15 minutes later Tristan Lucario, a member of the support staff who walked around “mapping” the work that in- terviewers were doing (15:3102–3103, Lucario), came by ap- parently wearing street clothes (4:697, Snyder). To Lucario, whose name Snyder did not learn until his discharge 2 days later (4:732, Snyder), Snyder acknowledges saying, “Gee, I’m disappointed; I took you for a teddy girl.” “No,” she assertedly replied, “I wear a tee shirt and shorts.” She then left. (4:697) Lucario tells us that the time of the incident was about 3:15 p.m. (15:3107) She informed some of her fellow support staffers of Snyder’s remarks. They told her to “write it.” She intended to do so, but first resumed her work. About 5 p.m. (15:3104, 3107) interviewer David Brown also made comments (much worse than Snyder’s, as we see shortly). Both sets of comments made Lucario uncomfortable and she considered them improper. (15:3105–3106) She then wrote up both inci- dents and submitted her written complaints to Supervisor Heidi Roberts that same evening. (15:3104, 3106, 3108) Actually, as we learn in a moment from Supervisor Geiger, complained orally to Supervisor Roberts, with Supervisor Geiger present, the evening of August 4. They told her to reduce the com- plaints to writing, and it was the next day that she submitted her written (hand printed, actually) one-page complaints against Snyder and Brown. (13:2695–2697) Lucario’s complaint as to Snyder is printed as a verbatim transcript of testimony by “Patrick” (“PS,” below) and “Me” (“TL,” below) as follows (GCX 10): PS: So where’s your pajamas? TL: I didn’t wear any. PS: Man, I was looking forward to seeing you in a teddy. Do you wear teddies? TL: No. PS: Well, then what do you wear? TL: I wear boxers and a T-shirt. In a paragraph following that report, Lucario printed as fol- lows (GCX 10): And then I walked off. The situation started when I was mapping and I was about to ask him for his start up code. The time was about 3:15 p.m. The situation made me feel really uncomfortable. [On the exhibit copy, a photocopy of a hand printed docu- ment, the bolding that I have displayed above does not appear, although the underlining of the single word does. Instead, an asterisk appears on each end of the last (and now bolded) sen- tence. As I type in the second asterisk, using Microsoft Word 97, the software converts everything between the asterisks (the asterisks disappear) from normal text to boldface text. Assum- ing that, by so placing the asterisks, Lucario intended them to mean just what the software has done, I leave the last sentence bolded.] Lucario testified that she prepared and submitted the docu- ment because Snyder’s comments “made me feel uncomfort- able. I didn’t think it was right for someone in the workplace to say that type of comment to me.” (15:3104–3105) Indeed, the fact that Lucario was in street clothes, and not in pajamas, should have served as an alarm bell for Snyder that the 18-year old Lucario would consider as unwelcome any sexually sugges- tive pajama comments about her bedroom habits. Lucario’s complaint as to David Brown (GCX 6) reports, in the same transcript format, a conversation that begins very much as the one with Snyder, only Brown quickly becomes far more direct and aggressive (“You’ve got the whole package and a nice body ... Let’s just say, you’re all that and a bag of chips, snickers, and skittles.” Then, grabbing Lucario’s arm, Brown remarked, “Your boyfriend’s a lucky guy.” In the paragraph she added to this complaint, Lucario states that she walked away thinking that it was the “second time that an older man had spoken to me so vulgar. It made me feel DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1302 really uncomfortable and I almost started to cry. It was about 5:45 p.m. when this happened.” No asterisks are included in this paragraph. Observing Lucario to testify in a persuasive fashion, I find that the August 4 incidents with Patrick Snyder and David Brown occurred as she describes. As noted earlier, Supervisor Geiger reports that Lucario first orally reported these incidents to Supervisor Heidi Roberts, in the latter’s office, while Geiger was present. Actually, Geiger states that Lucario came into the office, crying as Geiger re- calls, and said that something had happened between her and Patrick Snyder that made her extremely uncomfortable. As Geiger describes the meeting, Lucario never mentions David Brown. (13:2695–2697) In light of the substantial differences between the conduct of Snyder and Brown, with Brown’s con- duct, besides referring to Lucario’s body, also including his grabbing Lucario’s arm, I find that whatever crying Lucario did (and she possibly was tearful), was generated by David Brown’s assaulting behavior rather than by Snyder’s ill-advised words. I further find that Lucario’s oral report the evening of August 4 to Roberts and Geiger specifically included a descrip- tion of David Brown’s conduct, and that at trial Geiger deliber- ately, and in an effort to mislead me, sought to omit Brown and to portray Patrick Snyder as the cause for Lucario’s tearful condition. Finally, I find that Supervisor Geiger’s deliberate attempt to mislead me was driven by an intense animus against Patrick Snyder. (c) Patrick Snyder is discharged Snyder’s termination took place in Supervisor Chad Murray’s office moments after Snyder arrived for work about 3 p.m. on Friday, August 6. Present were Snyder, Murray, and Supervisor Doug Barlow, with Barlow seated at Murray’s desk. Barlow did most of the talking for Gallup. (4:728–731, Snyder; 12:2261–2263) Although, in his description, Barlow omits tendering the sexual harassment policy (12 2263), I credit Sny- der (4:729, 731) that, at the beginning of the conversation, Bar- low handed Snyder a copy of the sexual harassment policy. This apparently was a copy of the older version. (GCX 34 at internal 65) The conversation then went to the August 4 pa- jama contest.52 During his own testimony, Supervisor Murray did not address the subject of this meeting. At first Barlow hesitated about releasing Tristan Lucario’s name, and instead asked whether Snyder had asked a “support girl about a teddy.” Admitting that he had, Snyder said that such was in the spirit of the pajama party. (4:731) At trial Snyder explains that for days employees had been joking about pajamas. The event was to be fun. “And at Gallup, you were expected to participate in the fun events; otherwise, you were not considered a team player.” (4:731) Notwithstanding that testimony, as mentioned earlier (4:696), Snyder came to work in street clothes, not in pajamas. Barlow then read Lucario’s account (apparently the tran- script portion of GCX 10), and Snyder protested that he had not said that. Snyder stated that he asked Lucario why she had not 52 Crediting both Snyder and Barlow in part, the sequence and con- tent I render here is my finding of how the conversation generally pro- ceeded. dressed up and that “I would have figured you for a teddy girl.” To this Barlow asked if Snyder admitted making a comment. Snyder replied, “Yes, but it wasn’t the way that Tristan said.” (12:2264–2265, Barlow) Then handing Snyder a document listing “three charges” (4:732, Snyder) (apparently RX 44, the termination notice), Barlow said that, based on the two previous incidents of dis- rupting the workplace, in addition to the most recent incident, “You are terminated.” (12:2265, 2267, Barlow) Snyder then asked what the two other incidents were. (12:2267) Barlow said the first one was the disruption with Therron Hume. (4:732; 12:2265) Snyder acknowledged that he knew about that one. (12:2267) Apparently inspecting the termination notice, Snyder then asked (4:732), “What is this insubordina- tion charge?” Handing Snyder a copy of Supervisor Geiger’s June 2 “verbal” warning document (JX 22) (4:732, 735), Bar- low said it was in telling Supervisor Geiger on May 26 that Snyder needed to find a real manager (12:2265). Snyder pro- tested to Barlow, “This never happened.” (4:732, 735) Barlow then went and brought in Supervisor Geiger. (4:735, Snyder; 12:2267, Barlow) When Barlow returned with Geiger, Snyder looked at her and said, “You know this never happened.” Geiger replied, “Yes, it did.” Referring to the third sentence that alleged he had refused to sign the (verbal) warning, Snyder told Geiger that in a verbal warning there is nothing to sign. Geiger then reached over and (as earlier noted) scratched out the third sen- tence. (4:736, Snyder; 12; 2267, Barlow; 13:2692, Geiger) Geiger then left the meeting. (4:736) Snyder repeated that the incident never happened, and that it was fabricated. Barlow repeated that, based on the incidents of disruption in the work- place, the verbal warning by Geiger, and the offensive com- ments of a sexual nature, “you are terminated.” (12:2267– 2268) Barlow added that he would not have terminated Snyder solely on the last incident, but because of the combination of all the incidents,53 “we are terminating you.” (12:2265) Unrebutted testimony by Snyder reflects that, at one point, Supervisor Bar- low said (4:737): You know, you’re one of our best workers; you’re not being terminated for quality or performance. You’ve had this change of attitude a couple of months ago. A couple of months earlier (late May—early June), of course, was a time when the union movement was still very active—and also a time, May 26, when Snyder received two verbal warnings. Barlow denies that Snyder’s union activities were a factor in the decision to discharge him. (12:2315) On the other hand, Barlow does not tell us who made the decision to discharge, nor does he describe for us the process (by meet- ing, memos, or whatever) in which the decision was made. Before the meeting ended, Snyder asked about references. Barlow said that the only statement given would be dates of employment and position held (4:736, Snyder; 12:2268, Bar- low) Snyder asserts that Barlow also said that callers would be 53 Supervisor Uria-Ruiz confirms that it was the combination, and that without the last incident, the “last straw,” Snyder would still be employed, although on probation. (1:60, 132–133, 138) GALLUP, INC. 1303 told that Snyder was not rehirable. (4:736, Snyder) At trial Barlow claims that if anyone asked whether Snyder was rehir- able, the answer would be that Gallup was not at liberty to say. (12:2268) I find that Barlow, learning the safer procedure after Snyder’s termination, did tell Snyder that a prospective em- ployer calling for a reference would be told that Snyder was not rehirable—a potential defamation lawsuit that well-advised employers avoid. The meeting then adjourned, and Snyder gathered his belongings and left. (4:736–737; 12:2268) d. Disparity (1) Introduction To show disparity of treatment, the General Counsel cites in- cidents of clear sexual harassment by male employees—David Brown, David Clewis, and Randy Jenkins—who received more lenient treatment than that administered to Patrick Snyder. (2) David Brown Earlier we saw that David Brown was verbally gross, very aggressive, and even physically assaulting as to Tristan Lu- cario. Brown did not stop with Lucario that August 4, for Rheannon Eddy filed a complaint (GCX 7) about very similar conduct the same evening. (1:125–129, Uria-Ruiz) For his “inappropriate comments of a sexual nature” involving the two incidents, Brown was given a written warning (GCX 8 at 1), his firs offense, by Supervisor Chad Murray and cautioned that the next time could result in his termination. (1:125–129, Uria- Ruiz) When the next time came on August 17, as documented by one victim plus two witnesses (GCX 8 at 2–4), such conduct involving pulling a garment off the legs of a female interviewer and gawking at her exposed thighs, he was fired by Supervisor Murray for sexual harassment. (1:131; 2:301–304, Uria-Ruiz; GCX 8 at 5; RX 14) The General Counsel (Br. at 38-39) fails to fully articulate the discrepancy here. Apparently the Government’s contention is that the incidents involved several women and the comments and actions were more serious than the “innocuous comment” by Patrick Snyder. I see no disparity here. Brown was given a written (not a verbal) warning for the Lucario and Eddy inci- dents. (Snyder received a verbal warning for each of two inci- dents on May 26.) Brown was fired when he repeated his mis- conduct, and Snyder was fired for an asserted third disruption of the workplace. There is practically no evidence in the record calibrating the various levels of comments or conduct of a sex- ual nature. I note, however, that on March 16, 1999 (some 2 months before the union activity began), Supervisor Uria-Ruiz gave David Haynes only a verbal warning (RX 76) for sexually implicit and explicit remarks to and in the presence of another interviewer, such remarks including that he would like to “get in Tristan’s pants.” Uria-Ruiz gave only a verbal warning largely because, it appears, the complaining witness, Angela Lund, did not want to reduce her complaint to writing. (15:3110–3112) (3) David Clewis On August 12, 1998, Supervisor Joan Dorwart gave a verbal warning (documented on August 17; GCX 12 at 1) to David Clews for “harassment”—such conduct being, in effect, unwel- come advances over a 2-day period to a female employee, con- cluding by coming up behind her, as she was on the telephone, and running his hand over her shoulders and around to her up- per chest just above her breasts. After the victim ran Clews off, she complained to Dorwart and reduced her complaint to a two- page complaint (GCX 12 at 2–3). (1:148–149, Uria-Ruiz) This effort to show disparity is ineffective. While it shows that Gallup, or at least Supervisor Dorwart, gave a verbal warning for conduct more serious than the conduct of Patrick Snyder, it does not show that conduct such as Snyder’s received no warn- ing. (4) Randy Jenkins (a) Facts This case, involving interviewers Vanessa Zieler and Randy Jenkins, has the potential to show disparity in treatment. Un- fortunately, the asserted events occurred largely in 1994 and 1995, and, because of the 4 to 5-year passage of time since the events, Zieler’s memory understandably recalled details as to time and sequence only generally and vaguely. No personnel or other records from that time period were offered to assist in at least getting a time frame for the sequence of major events. On May 22, 1998, Supervisor Tom Langenegger fired inter- viewer Randy Jenkins. (2:231, Uria-Ruiz) The termination notice (GCX 22) lists the reasons as (1) a long history of not putting in his required hours for each week and pay period; (2) falsifying his timesheets so as to over-represent his hours; (3) disrupting the workplace, and (4) soliciting other trackers to misrepresent their hours. (The record does not tell us what conduct of Jenkins constituted “disrupting the workplace,” nor does it give us the date.) The General Counsel’s contention here is that Gallup re- sponded to oral complaints and requests for help for a period exceeding a year from interviewer Vanessa Zieler (Lynne Zieler’s daughter, 3:447, 457) by nothing more than exhorting Jenkins to stop. (Br. at 38) Among the managers to whom Zieler assertedly reported Jenkins’ unwelcome verbal and physical advances was Supervisor Doug Barlow. (3:450–451, V. Zieler) According to Zieler, she complained 10 or 12 times over several months without any material success. She even showed them one of the love letters that Jenkins wrote her, but which she no longer has. (3:450–451) Zieler concedes that for the first 6 months or so of this conduct, she did not complain to management because she was “young, naïve, and stupid.” (3:465) She began work at Gallup in August 1994 as she turned 17 (3:447), and as of the trial she was 22 (3:456). Zieler asserts that at one point she telephoned officials in Nebraska, and 2 weeks later one of them came to the office and held a meeting, attendance not mandated, with employees. One of the topics discussed was sexual harassment. The official said that such conduct was not tolerated in the workplace. (3:454–455, 464) Eventually Zieler went to Supervisor Bar- low’s office crying and telling him that she no longer could take the harassment and that if Gallup would not do anything she would have to leave. Barlow tried to talk to her, but she did not want to talk. She then left, apparently quitting. This was in February 1996. (3:455, 464–465) Zieler is not sure whether she ever filed a complaint in writ- ing (3:457-458). In the meeting with the supervisors, when DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1304 Zieler was present, Jenkins was told that his conduct was not appropriate and that he would be terminated if he did not stop. (3:461) Nevertheless, the harassment continued, and when Zieler reported this to Supervisors Barlow and Dennis Welch they said they would speak with Jenkins. (3:463) For his part, Supervisor Barlow asserts that Vanessa Zieler never complained to him about Randy Jenkins, and that she never had any discussions with her about Jenkins, either with or without Supervisor Dennis Welch. (12:2271) Moreover, Gallup asserts that Vanessa Zieler was employed twice. She did not give notice when she left the first time, nor the second time either. She did not report directly to Barlow, and she did not discuss her departure with him. Apparently the reemploy- ment process is almost automatic for someone who has per- formed well. Even so, when she asked to come back for her third employment, that request was denied for failure to give notice when she left the previous time. At no time was there any mention in these rehirings about any comments of a sexual nature by a coworker. (12:2272–2276) Gallup neither identi- fied nor offered copies of any personnel records of Vanessa Zieler showing her dates of employment or anything else about her employment with Gallup. On brief Gallup does not address the Vanessa Zieler matter. As noted at the beginning of this topic, many of the asserted incidents of conduct by Jenkins asserted occurred during the years of 1994–1995 (3:448), and most of Zieler’s description if set in a generalized time frame. The most specific incident occurred about 6-8 months after she began work (so about early 1995), as she and Jenkins got on the elevator to go down to take a smoke break together. Note that some of the offensive inci- dents already had occurred before this occasion. (3:459) On the elevator, Jenkins pushed a button for the 10th floor, of the 12-story building, rather than the down button to the first floor. However, Zieler was talking to Jenkins and not paying attention to what he had done, and then they arrived at the 10th floor which, at the time, was vacant and dark. Notwithstanding those conditions, Zieler got off the elevator with Jenkins who, appar- ently moments later, began trying to pull off Zieler’s clothes and to forcefully have sex with her. (3:448, 453–454) This apparently was the incident for which Jenkins assertedly was told to stop or he could be terminated. Nevertheless, after this 10th floor assault, Zieler continued to join with everyone, in- cluding Jenkins, downstairs on smoke breaks. (3:462) (Credit- ing Supervisor Barlow that the building has only 10 floors (11:2156), I gather that Zieler was referring to another floor, possibly the 8th. To avoid confusion, however, I use Zieler’s description that it was on the 10th floor with two other floors above that one.) (b) Discussion Vanessa Zieler was not a persuasive witness. I recognize that, some 4 to 5 years after the events, Zieler would have diffi- culty recalling details of time and sequence as well as of the incidents themselves. Even so, I noticed that, when the cross examination asked about any continued association with Jen- kins for smoke breaks after the 10th floor assault and the warn- ing to Jenkins, Zieler became defensive and vague as to the sequence of anything. Adding to the Government’s problem here is that Zieler apparently either no longer has, or cannot find, the love letters that Jenkins either sent to her or gave to her. Moreover, Zieler apparently never filed a written com- plaint about the conduct of Jenkins. While a written complaint is not necessary to initiate action, it certainly would help here where the question is whether any of this ever happened. Respecting Gallup’s defense, it can hardly be said that Gallup did much to support Supervisor Barlow’s testimony. Thus, copies of personnel records of Vanessa Zieler would at least, presumably, corroborate Barlow’s assertions that Zieler worked twice at Gallup and, on her third try, was rejected. On the other hand, the General Counsel did not call Vanessa Zieler at the rebuttal stage to correct or to clarify the assertions made by Barlow about her employment history with Gallup. Finally, in light of the documented incidents where Gallup has issued warnings for comments or conduct of a sexual na- ture, plus several other for nonsexual matters, issued under the catch-all category of “disturbing the workplace,” it seems very odd that Gallup’s supervisors (Zieler asserts that she had com- plained to several supervisors, not just to Supervisor Barlow) would show no interest in documenting a warning, even a ver- bal warning, to Randy Jenkins. On the other hand, recall that Supervisor Barlow did not document his verbal warning of May 26 to Patrick Snyder—that was done by Supervisor Geiger, and then not until August 6, the day of Snyder’s discharge (JX 21b). All in all, the evidence from both sides is unimpressive. Consequently, because the General Counsel carries the burden of persuasion as to the Vanessa Zieler matter, and that burden has not been met, I find that the Government has failed to prove disparity through the evidence as to Vanessa Zieler and Randy Jenkins. I emphasize that I do not find that Randy Jenkins did not harass Vanessa Zieler substantially as she describes. I sim- ply find that, on this record, no findings of such harassment by Jenkins and complaints by Zieler to Gallup are warranted. e. The Government’s prima facie case As I have found, Supervisor Stephanie Geiger’s personal dis- like for Patrick Snyder is the nature of her animus against him. About the only item the Government can point to at this point possibly showing union animus by Gallup against Snyder is the remark that Supervisor Barlow made to Snyder during the dis- charge meeting. Recall Barlow’s comment that Snyder was one of Gallup’s best workers, and was not being terminated for quality or job performance. But, “You’ve had this change of attitude a couple of months ago.” (4:731, Snyder) What “atti- tude” was Barlow referring to? The General Counsel argues that it was Snyder’s union activities. (Br. at 36) The General Counsel does not pause to explain why it is those activities rather than the disciplinary actions taken against Snyder on May 26. The latter are just as likely as the former. I therefore find that this comment by Supervisor Barlow has not been shown to refer to Patrick Snyder’s union activities. The final possibility is that the third incident, the comments of a sexual nature to Tristan Lucario, were seized on as a pre- text to get rid of Patrick Snyder because of his union activities. Although Snyder’s brief comments were ill advised, they clearly are mild compared to some others. But, Gallup would not have terminated him for that one incident. Unfortunately GALLUP, INC. 1305 for Snyder, that incident made his third disturbance of the workplace. Short of a showing of disparity, Snyder cannot prevail. As already summarized, no disparity in treatment has been shown. Accordingly, finding that the Government has failed to establish, prima facie, that Snyder’s union activities were a moving reason for the decision to discharge him, it fol- lows that the Government’s complaint as to Patrick Snyder must be dismissed. I therefore shall dismiss complaint para- graph 25 as to Patrick Snyder. CONCLUSIONS OF LAW 1. By making certain coercive statements to employees at its Houston, Texas office, Gallup has violated Section 8(a)(1) of the Act. 2. Gallup has not violated the Act as otherwise alleged. 3. The unfair labor practices found affect commerce within the meaning of Section 2(6) and (7) of the Act. REMEDY Having found that the Respondent has engaged in certain un- fair labor practices, I find that it must be ordered to cease and desist and to take certain affirmative action designed to effectu- ate the policies of the Act. On these findings of fact and conclusions of law and on the entire record, I issue the following recommended54 ORDER The Respondent, Gallup, Inc., Houston, Texas, its officers, agents, successors, and assigns, shall 1. Cease and desist from (a) Restricting employees from posting, distributing, or pos- sessing union-related literature, flyers, or notices anywhere in the Houston, Texas office that employees are permitted to post, distribute, or possess other nonwork related materials. (b) Removing or confiscating posted or distributed union- related materials from places where employees may post, dis- tribute, or possess other non-work related items. (c) Requiring employees to notify a supervisor before dis- tributing union literature at work where no such notification is required before distributing other non-work materials. (d) Instructing or requesting new employees, or trainees, to report to supervision any attempt by other employees to speak to them about union matters as the new employees, or trainees, are working, where the instruction or request does not also extend to other nonwork related matters. 54 If no exceptions are filed as provided by Sec. 102.46 of the Board’s Rules and Regulations, the findings, conclusions, and recom- mended Order shall, as provided in Sec. 102.48 of the Rules, be adopted by the Board and all objections to them shall be deemed waived for all purposes. (e) In any like or related manner interfering with, restraining, or coercing employees in the exercise of the rights guaranteed them by Section 7 of the Act. 2. Take the following affirmative action necessary to effec- tuate the policies of the Act. (a) Within 14 days from the date of this Order, rescind all rules restricting employees from posting, distributing, or pos- sessing union literature, flyers, or notices in the Houston, Texas office more so than it restricts such posting, distributing, or possessing other nonwork related materials, and contem- poraneously with the removal of such restrictions, notify its Houston, Texas employees in writing that such rules have been rescinded and that they are free to post, distribute, or possess union-related materials in the same manner that they are free to post, distribute, or possess any other non-work related items. (b) Within 14 days from the date of this Order, notify all em- ployees, especially new employees, or trainees, that they are as free, during their working time, to talk with other employees about union matters and without notifying supervision of such talk, as they are permitted to talk, without such notification factor, about any other nonwork related matter. (c) Within 14 days after service by the Region, post at its of- fice at Houston, Texas, copies of the attached notice marked “Appendix.”55 Copies of the notice, on forms provided by the Regional Director for Region 16, after being signed by Re- spondent’s authorized representative, shall be posted by the Respondent and maintained by it for 60 consecutive days in conspicuous places including all places where notices to em- ployees are customarily posted. Reasonable steps shall be taken by the Respondent to ensure that the notices are not al- tered, defaced, or covered by any other material. In the event that, during the pendency of these proceedings, the Respondent has gone out of business or ceased its operation at the facility involved in this proceeding, the Respondent shall duplicate and mail, at its own expense, a copy of the notice to all current em- ployees and former employees employed by the Respondent at any time since May 18, 1999, the date of the first unfair labor practice found in this proceeding. (d) Within 21 days after service by the Region, file with the Regional Director a sworn certification of a responsible official, on a form provided by the Region, attesting to the steps that the Respondent has taken to comply. IT IS FURTHER ORDERED that the complaint is dismissed inso- far as it alleges violations of the Act not specifically found. 55 If this Order is enforced by a judgment of a United States court of appeals, the words in the notice reading “Posted by Order of the Na- tional Labor Relations Board” shall read “Posted Pursuant to a Judg- ment of the United States Court of Appeals Enforcing an Order of the National Labor Relations Board.”