Indiana Metal Prod., 442 F.2d at 51. An administrative law judge is not bound

to believe or disbelieve the entirety of a witnesss testimony, but may choose to believe only certain portions of the testimony. See Altemose Constr. Co. v. Natl Labor Relations Bd., 514 F.2d 8, 15 n.5 (3d Cir. 1975)(citing National Labor Relations Bd. v. Universal Camera Corp., 179 F.2d 749, 754 (2d Cir. 1950), vacated and remanded on other grounds, 340 U.S. 474 (1951)).

   Moreover, based on the unique advantage of having heard the testimony firsthand, I have observed the behavior and outward bearing of the witnesses from which impressions were garnered as to their demeanor. In short, to the extent credibility determinations must be weighed for the resolution of issues, I have based my credibility findings on a review of the entire testimonial record and exhibits with due regard for the logic of probability and the demeanor of witnesses.

   The transcript of the hearing in this case is 957 pages in length, comprised of the testimony of six different witnesses. Although the record contains some inevitable inconsistencies in testimony, I find the separate testimony of all six witnesses to be credible.

   Specifically addressing the testimony of Mark Kelly, I found his answers to be sincere, despite the difficulties of pro se testimony. It was apparent that Mr. Kelly believed the version of the facts as he was relaying them. On specific incidents, however, Kellys testimony appears to be exaggerated, as no corroborating evidence supports his version of the facts and reason alone disagrees with Kelly. These few incidents will be noted in my analysis. Furthermore, Kellys honesty and trustworthiness were undercut upon cross-examination to some extent. The Respondents inquiry into previous statements made by Kelly under oath at an Unemployment Compensation hearing, demonstrated inconsistencies in his testimony, at a minimum. Alternatively, it demonstrated that Kelly allowed his answer to a question regarding the level of knowledge about his NRC technical investigation that he possessed at the time of his resignation to be interpreted in a misleading light. (Tr. 445-447). When questioned with the inconsistencies, Kellys explanation changed several times. (Tr. 445-48).

   I specifically find the testimony of Paul Prevey to be credible.

III. ISSUES

1. Whether Mark Kelly engaged in protected activity in his inclusion of various recommendations in the G.E. Nuclear Quality Assurance Incident Report and his subsequent refusal to sign a version of said report without such recommendations.

2. Whether Respondent discriminated against Mark Kelly by constructively discharging him through events from September 1999 to February 2000.

3. Whether Respondent discriminated against Mark Kelly by threatening to fire him for "wasting time" after Mark Kelly ignored Paul Preveys editorial changes to the G.E. Nuclear Quality Assurance Incident Report.

IV. STATEMENT OF THE CASE

   This issue in this case is whether Lambda Research, Inc. discriminated against Mark Kelly because of his engagement in protected activity.

   Complainant contends that his recommendation of certain safety and procedural changes

in a Quality Assurance Incident Report [QAR] and his concomitant refusal to sign a QAR with his recommendations omitted was a "protected activity" under the Energy Reorganization Act. Complainant alleges that the respondent discriminated against him and constructively discharged him because of his recommendations and refusal to acquiesce in an edited QAR by: 1) giving him a poor performance evaluation; 2) receiving a small bonus in Fall 1999; 3) changing company policies to only allow information to be provided to clients in final report form; 4) reducing his employment responsibilities; 5) placing negative evaluations in his personnel file without reviewing them with him 6) engaging in angry outbursts and tense confrontations with him; 7) shortening the delivery date on a laboratory project; 8) threatening to fire him for "wasting time" on the GEN QAR; and 9) reducing the staff in Lab II.

   It is Lambda Researchs position that no protected activity took place, no retaliation occurred, and, furthermore, no nexus between the two can be demonstrated. Lambda provides explanations for each of the events that Kelly credits as discriminatory.

V. FINDINGS OF FACT

    Lambda Research, Inc. is a research laboratory, providing research services such as Quantitative and Qualitative Phase Analysis, Residual Stress Management, and Pole Figure Preparations to governmental, industrial, and educational clients. The laboratory specializes in X-ray diffraction. Lambda Research, Inc. is located in Fairfield, Ohio.

   Mark Kelly received a Ph. D. in analytical chemistry from the University of Cincinnati in 1988. In the summer of 1998, he was interviewed by Paul Prevey, the owner, director of research, and president of Lambda Research. Prevey hired Kelly to be a technician in "Lab II." In December 1998, he was promoted to a supervisor position in Lab II.

   Kelly developed expertise in x-ray diffraction and texture analysis of materials as an employee at G.E. Aircraft Engines. (CX 22). After his employment at G.E. Aircraft Engines, Kelly worked as a senior analytical chemist for Shepherd Color for four and one-half years. After beginning his employment at Lambda Research, Kelly trained in Lambda procedures, including several x-ray diffraction methods and x-ray texture analysis methods performed for specific clients, such as General Electric Nuclear [GEN]. Kellys work as a technician was reviewed favorably. His immediate supervisor, Perry Mason, testified that Kelly was good following quality assurance procedures and understood x-ray diffraction. (CX 17, 63).

   During the time of Kellys employment with Respondent, Lambda Research performed x-ray diffraction texture analysis of zirconium [Zr.] alloy materials for GEN. (CX 3-5, 30, 36). In order for Lambda to qualify to perform such work, Lambda and its quality systems were audited by GEN. In addition, Lambda participated in a round-robin exchange of research data with GEN to demonstrate their capability to perform accurately the necessary work for GEN.

   In January 1999, Mr. Richard Calcaterra of GEN contacted Kelly concerning an upcoming audit of Lambda Research. (Tr. 51). The purpose of the audit was to determine if Lambda was proficient enough to perform research services for GEN. One goal of the audit was to determine if Lambdas procedures complied with various nuclear safety regulations, specifically 10 C.F.R. § 21 and 10 C.F.R. § 50(b). Mr. Calcaterra relayed these regulations to Kelly and their importance in ensuring nuclear safety. (Tr. 53, 55). Calcaterra, however, did not tell Kelly that the regulations would apply to any specific work to be sent to Lambda. (Tr. 54-55). Calcaterra asked Kelly to arrange for x-ray diffraction texture analysis on zirconium specimens. (CX 46, 5). The purchase order for the first type of work that Calcaterra had described to Kelly did include the 10 C.F.R. § 21 notification requirement. (Tr. 56-57; CX 5, p. 3). Kelly testified that his understanding of 10 C.F.R. § 21 required him to contact GEN when problems with the requested research occurred. (Tr. 58). Kelly also asserted that this responsibility to notify clients of problems stemmed from Lambdas corrective action procedure in its quality system. (Id.).

   In June 1999, Kellys Lab II performed further texture analysis on the zirconium tubes. This work was performed after a purchase order generated on May 12, 1999, was received. (Tr. 927, RX 1, p. 3). This purchase order did not contain the 10 C.F.R. § 21 notification requirement. (RX 1). After the testing was completed, Michael Glavicic, a senior research assistant at Lambda, analyzed the collected data. In his report, Glavicic noted some irregularities. Specifically, some of the data appeared "atypical," and Glavicic included a notification that alerted GEN to this fact in his report. The final report sent to GEN stated that some of the data was "not typical." (CX 2; Tr. 120). Because of the unusual results and a request by Chuck McKinney of GEN, further testing was performed on another sample - this time with no unusual results. (Tr. 121-22). The report submitted to GEN regarding the re-test concluded that "sample mounting error" had produced the unexpected results, not an atypical sample. (CX 2).

   After the problems with the GEN project were identified, Kelly wrote Prevey a memorandum suggesting several changes in the laboratorys procedures on July 2, 1999. (Tr. 225; CX 19, p. 15). Kelly asserted that Prevey often reacted to problems brought by staff by insulting them, accusing them of being the problem instead of the procedures. (Tr. 225). Because Kelly perceived a communication problem with Prevey, he wrote the memorandum addressing the problems he identified in the lab. (Tr. 226). In addition to changing some procedures and ensuring correct personnel training, Kellys memorandum suggested contacting GEN about the problem of sample flatness and its effect on past work performed by Lambda. (CX 19, p. 15). Kellys memorandum clearly indicates that Kellys concern with nuclear safety is not a byproduct of that specific project, but rather previous work that GEN has sent to Lambda which was subject to the same inadequate procedures. (Id.).

   Kelly and Prevey discussed his memorandum, and Prevey ended the discussion by throwing the document in the trash can. (Tr. 227). This period of time was a tense time in the lab, according to Kelly. He worried about Prevey coming into the lab "launch[ing] into a tirade." (Tr. 244). For example, during a review of standards in the lab, Kelly alleges that his job was threatened because he was "stealing time" from Prevey by reviewing procedures. (Tr. 244-45). The atmosphere was so tense that Kelly immediately called the Nuclear Regulatory Commission [NRC] after his job was threatened. (Tr. 245-46). He did not identify himself to the NRC, however. (Tr. 246).

   Whenever a problem occurs with testing at Lambda, a Quality Assurance Incident Report [QAR] is required to be prepared. (Tr. 520, 860). The QAR is an internal document, and is not sent to external clients. The document, however, is available for review, if requested by external parties, during system audits of the facilities. As mentioned above, GEN performed such an audit before it authorized Lambda to perform research for GEN. (CX 46, Tr. 856). The maintenance of a high quality QAR system is integral to the successful operation of such a research facility. (Tr. 696-697). As with other laboratories, Lambdas product is information. To move that product, Lambda must convince clients that its information, and the systems used to procure it, are maintained at the highest level and followed zealously. (Tr. 697). The QAR promotes consistent, effective information-gathering procedures by ensuring proper procedures are consistently enforced. (Tr. 860). When a problem arises with the procedure or the results obtained from the procedure, quality assurance incident reports leave "bread-crumb" trails throughout the procedure, enabling Lambda and, if necessary, external clients to retrace the steps of the research to determine where and when mistakes occurred. (Id.). As Lambda frequently worked with materials for the nuclear industry, the QAR system also served to ensure clients that defects in data could be contained and minimized, thus minimizing the possibility of a catastrophic mistake. (Tr. 59). When working with hazardous materials, this guarantee is obviously important to the clients. (Tr. 861-62).

   On July 16, 1999, Kelly, along with lab technician Chris Barger, prepared a QAR addressing the problems with the testing of the GEN zirconium tubes identified by Glavicic and the final report to GEN. (CX 1; Tr. 247). As standard procedure dictated, Kelly forwarded the completed QAR to Paul Prevey, president of Lambda Research and Kellys supervisor, for approval. (Tr. 249). The QAR included the following recommendations: 1) previous zirconium texture analysis results should be reviewed to

determine if significant distortion of texture analysis occurred [because of the sample mounting error]; 2) the client should be notified that other data may have been distorted so that they could assess the significance of the problem; and 3) the pole figure procedure and sample preparation procedure should be revised to prevent reoccurrence of the sample preparation problem and failure of the peak program. (CX 1, p. 2). Kelly attributed most of the problem to technician error, stating, "Most of the flatness problem occurred because the technician did not follow the directions for use of the contact cement required by [the sample mounting procedure]." The QAR submitted by Kelly did not mention or implicate nuclear safety; however, the memorandum written to Prevey by Kelly on July 2, 1999, addressing the exact same problem, clearly identified that the presence of the 10 C.F.R. 21 requirement in previous purchase orders from GEN was a driving force in Kellys concern and recommendations. (CX 19, p. 5).

   On August 16, 1999, Perry Mason returned the QAR to Kelly. (Tr. 249). Prevey had not signed the QAR, but he had removed the recommendations to revise the sample mounting and pole figure procedures and notify GEN of the problems. (CX 1; Tr. 250-51). Prevey decided that the current procedures, if followed correctly, already insured sample flatness. (Tr. 663-67). Prevey maintained that no further notification was necessary because GEN had already been notified in the second report that a "sample mounting error" had occurred and because all data had been previously checked against archival data. (Tr. 685). Prevey believed that Kellys QAR recommendation to notify GEN of the sample mounting error was actually referring to the 90 degree rotation problem the lab had experienced months earlier. (Tr. 653). Accordingly, Prevey noted on the QAR that the client had already been notified. (Tr. 654). Prevey also dismissed Kellys recommendation to re-check former projects for GEN because he believed that this was the first time that data produced for GEN had not matched the archival data or the data set received from the round-robin Lambda had participated in with GEN. (Tr. 659-60, 686).

   Kelly disagreed with Preveys changes, labeling them "inappropriate." (Tr. 250). Kelly asserts that "Preveys directions to have the recommendations to correct the procedure removed from the QAR are evidence that his intent was to avoid admitting and documenting the problems with...procedures used to obtain texture results for GEN up to that point and...during audits of Lambda Research." (Complainants Post-Hearing Brief, p. 26).

   After the original draft of the QAR, Lab II identified a random mechanical problem that may have influenced the results sent to GEN involving the "peak finding program." (Tr. 254-57). Kelly informed Prevey of the problem, but Prevey rejected a change in procedure, stating that procedures were not changed to fix mechanical problems. (Tr. 259).

   After the first submission of the QAR, Kelly and Prevey discussed the proposed changes. (Tr. 659). Prevey maintained that the procedures were sound, the problems existed in execution, and that it was pure speculation on Kellys behalf that problems had occurred in the past redounding from sample flatness errors. (Id.). Prevey based his conclusion that there were no problems with the procedures as they stood based on the hundreds of previous samples examined by Lambda and, more importantly, the consistent accuracy of Lambdas work compared to the round-robin exchange of data it had participated in with GEN. (Tr. 660).

   In September 1999, Western Zirconium requested that Lambda perform x-ray diffraction texture analysis on some samples. (Tr. 272). Kelly quoted the client a turn-around time of three weeks, but, upon the clients request, Prevey, unbeknownst to Kelly who was out of the lab at that time, shortened the delivery time in half to approximately a week and a half. (Tr. 868, 274). When Kelly returned, Prevey confronted Kelly about quoting three weeks for the turn around time and instructed that turn around times should be shorter. (CX 25, p. 18). The project was not completed in time. Kelly had not performed the analysis because he believed that he was not certified to do so and to have performed the work uncertified would have been a QA violation. (Tr. 287). When the only employee certified to perform that specific type of analysis Mike Glavicic returned from a funeral the day after the deadline Prevey quoted, Kelly attempted to ask Glavicic to quickly perform the analysis. (Tr. 288).

   Prevey overheard the request and confronted Kelly regarding why the project had not been completed. (Tr. 289). Prevey and Kelly disagreed as to whether it was appropriate for Kelly to have performed the analysis. (Tr. 288, 700). Kelly was aware of a new procedure being implemented by Glavicic, and he had seen a demonstration of the procedure. (Tr. 287-88). Kelly, however, had not seen the final procedure. (Tr. 288). Prevey reprimanded Kelly because Kelly could have performed the work under the existing procedure, but chose not to. (Tr. 700-01). Kelly alleges that the reprimand was laced with cursing, and that he was later told that he was "stupid" and "incompetent" and using the quality system to hold up work (Tr. 289-90, 294). Kelly admits that Prevey said "that we were all stupid and incompetent." (Tr. 290).

   Prevey subsequently noted Kellys failure to complete the project in Kellys personnel file. (Tr. 290, 700). Prevey also noted that work on the project not related to the new procedure had not been completed. (Tr. 701-02). Regardless of any perception on Kellys part of an inability to perform the procedure, he could have completed the QA checks unrelated to Glavicics new procedure. (Tr. 701). Kelly did not see or know of the memorandum placed in his personnel file until the discovery phase of the instant case. (Tr. 271).

   On September 14, 1999, Kelly noticed some oil on the milling machine and began to wipe it off. (Tr. 299). At that time, Prevey walked through the laboratory and inquired about what Kelly was doing. (Tr. 298). Prevey stopped because the milling machine was making a loud noise. (Tr. 708). When Kelly attempted to explain that he was attempting to locate the source of the oil, he was accused of dropping maintenance assignments. (Id.). Kelly admitted that it was his responsibility to have seen to the machines maintenance. (Tr. 300). Kellys mistake was caused by a misunderstanding in the allocation of responsibilities after a co-worker transferred into the laboratory. (Tr. 299-300). Kelly maintains that Prevey began yelling and cursing at him, but no other testimony supports this claim. (Tr. 300). Chris Barger was there at the incident, (Tr. 298, 708; CX 17, p. 31), but he offered no testimony on any cursing or yelling by Prevey. (Tr. 913). Indeed, Barger testified that he never heard Prevey swear at Kelly. (Id.).

   Prevey indicated that his reaction was not caused by the lack of maintenance on the machine, but by Kellys reaction to Preveys inquiry. Prevey recorded in a memorandum to Kellys personnel file that Kelly expressed the belief that Lambdas policy was for employees to wait for failure of the apparatus before interceding with maintenance. (Tr. 708-09; CX 17, p. 31). Prevey was "aghast" at Kellys response, especially since the remark was made in front of two employees that Kelly supervised, Beth Schumaker and Chris Barger. (Tr. 709). Kelly advances that Preveys reaction was not called for because all of his former personnel reviews demonstrate responsibility with company property and Prevey reacted much less harshly to dropped assignments by other employees. (Tr. 300-06). Kelly, however, later admitted that he "was aware of other situations at Lambda Research where people had done similar things and he [Prevey] may have lost his temper." (Tr. 312). Kelly did not see the memorandum Prevey authored until he initiated the instant case and received it in discovery. (Tr. 307, 309).

   Kelly and Prevey again confronted each other when a developer solution ran out at the laboratory. (Tr. 314). Kelly was responsible for the inventory, and, between inventories, someone in the lab had used an unusual amount of developer, causing the solution to run out. (Tr. 315). When Prevey was made aware of the situation, he called Kelly to his office and held him responsible for the upheld project. (Tr. 316). Kelly alleges that Prevey told him that he should be held responsible for delaying a space shuttle flight. (Id.). Kelly took this as a sarcastic remark, but he knew that Prevey was angry. (Id.).

   Prevey called Kelly into his office and informed him that he was somehow "screwing up" a report generated through the offices word processing system. (Tr. 317). Prevey said that Kellys corrections on his computer to word processing templates was causing the front office an undue amount of work. Kelly alleges that Prevey accused him of stealing from him and wasting untold amounts of time in senseless work. (Id.). Kelly later discovered that he was not the individual causing the error in the reports, and informed Prevey. (Tr. 318). When told, Prevey simply walked away. (Id.).

   In early September 1999, Kelly took a vacation day. When he was gone, Prevey checked on the laboratory and discovered two technicians sitting at their desk with their diffractometers empty. (Tr. 319, 338). When Kelly returned, he was called into Preveys office where Prevey proceeded to inform him of the inactivity in the lab, that the inactivity affected the bottom line profitability of the laboratory, and that he should be fired for that. (Tr. 319). Kelly agreed that profitability was very important, but he did not agree with the hostile tone with which Prevey spoke or Preveys assumption that the workers were unproductive intentionally. (Tr. 339). Kelly claimed that Prevey cursed at him and engaged in what he termed "verbal abuse." (Id.). After Kelly investigated what the technicians were working on during the day of his absence, he went back to talk again to Prevey. (Id.). Prevey accused the technicians of "playing games" with Kelly, and that he was not properly managing his employees. (Tr. 340). To verify the technicians stories, Kelly compared their time cards against past time cards, concluding that nothing was irregular. (Id.). Prevey asserted that the time cards were being filled out incorrectly, money was flying out the window, and Kelly was mismanaging the lab. (Id.). In a later QA management meeting, it was discovered that the technicians were not incorrectly filling out the time cards. (Tr. 340-41). Kelly felt that nothing came of the time card issue, but "the verbal reprimands and accusations were extremely stressful and very difficult." (Tr. 341). Kelly attributed the treatment to his handling of the GEN incident. (Id.). Kelly stated that Preveys behavior would get better, then worse. (Id.).

   On several occasions, Prevey stated that Kelly should be shot for raising the procedural issues and holding up work to address QA issues. (Tr. 337). Kelly took the statements as only expressions and did not take them as physical threats, but rather took them as indication of the anger and hostility with which Prevey would approach problems. (Id.). No other witness at the hearing testified to hearing Prevey ever physically threaten Kelly. Furthermore, no witness testified to ever being physically threatened by Prevey.

   Kelly alleges that his personnel reviews were ranked lower in retaliation for his refusal to sign the GEN QAR. (Tr. 345) The record contains four personnel reviews for Kelly. (CX 17, p. 9-26). The first two reviews were completed by Kellys supervisor before Prevey, Perry Mason. Those reviews occurred in September and November 1998. (CX 17, p. 9-14). Kellys final two reviews were completed by Prevey. (CX 17, p. 15-26). Those reviews occurred in June and September 1999. (Id.). In between the latter two reviews, the GEN QAR issue arose, and Kelly refused to sign the edited version. (Tr. 358). In September 1999, Kelly received a performance appraisal. (Tr. 345; CX 17, p. 22-24). Kellys rankings were lower in almost every area compared to his previous performance appraisal by Paul Prevey. The evaluation is separated into two categories: general work characteristics and attitude and behavior characteristics. In the general work characteristics section, Kellys second evaluation from Prevey is lower in five of the seven categories, with the biggest drop in "Supports and complies with QA policies and procedures." (CX 17, p. 15, 22). In the attitude and behavior characteristics category, Kellys evaluation dropped in three of the seven categories. (CX 17, p. 16, 23). Kelly exaggerates when he claims that all of his scores, except one, were lower in September than June. (Tr. 347). Of the fourteen rated categories, only eight were ranked lower, not thirteen as Kelly intimates.

   Prevey attributes the lower rankings to several factors. First, Prevey stated that he tends to give higher grades the first time he reviews an employee as an incentive tool. (Tr. 769). While this factor seems questionable, his remaining reasons are convincing. Secondly, Prevey claims that some of Kellys performance did not match the level of competence he assumed Kelly possessed. (Tr. 768). The second evaluation, unlike the first evaluation, reflected the fact that Prevey had an opportunity to observe Kellys work for a substantial period of time. (Id.). In addition, Prevey attributed Kellys higher marks on the June evaluation due to Kellys expressed interest at that time of "ferret[ing] out any loose ends in the QA system." (Tr. 770). Kellys enthusiasm for the task earned him higher marks. (Id.). When the second review came around, the promise of Kellys performance, according to Prevey, had not fulfilled itself. (Tr. 770-71). Finally, Prevey admitted that Kellys personnel reviews, like all such reviews, are subjective. (Tr. 769, 772).

   Kelly specifically points to one particular evaluation mark. During the period encompassed by the June evaluation, Kelly missed three days of work for sick time. His attendance record was judged "outstanding" by Prevey. (Tr. 772-74). During the period of time encompassed by the September review, Kellys attendance improved, using only one day of sick leave. (Tr. 774). Yet, in the September review conducted by Prevey, Kellys better attendance was actually judged worse, receiving a "very good" instead of the "outstanding" rating he received when he missed three days worth of work due to illness. (Tr. 772-73). When presented with the seeming inequity, Prevey responded, "I dont review these [attendance record sheets], really, when I look at that. I mean, thats not really a quantifying, I mean, Im not trying to quantify this that way." (Tr. 774).

   Kelly alleges that his receipt of the smallest third-quarter bonus in Lab II evinces retalia-tory discrimination. (Complainants Post-Hearing Brief, p. 49). The record does not support this allegation.

   In his testimony, Prevey described the two separate bonus programs implemented at Lambda. (Tr. 874). Managers at Lambda, including Kelly, participate in both programs, and all other employees participate in only one bonus program. (Id.). Prevey participates in neither program. (Id.). In the bonus program participated in by all employees, ten percent of a quarters profits are divided up based upon the following criteria: one-third based on years of service and two-thirds based on salary. (Id.). After the application of a formula, the money is divided up. The first half is a mandatory bonus, and the second half is a discretionary amount that the supervisor may award. The second bonus program, participated in by only managers, is more arbitrary and awarded on an annual basis. Prevey will take around 10% of the profits and divide it on a discretionary basis between the managers. (Tr. 874-75).

   In the first and third quarters of 1999, the company lost money. (Tr. 877). No bonuses were given in the first quarter, except for one to Chris Barger out of a back-up pool of money. In the third quarter, Kelly did receive the smallest bonus in Lab II, but only because he dispensed money to the employees he supervised under the discretionary bonus fund. Before Kelly himself dispensed funds, he actually received the largest bonus in Lab II. In December, Kelly received a bonus of ,599.89 from the bonus program all employees participated in. (RX 31, p.5). In addition, Kelly received an annual bonus given only to managers totaling $2,000. (RX 31, p.3). The record reflects that Prevey actually gave all managers bonuses larger than the bonus program normally produced in the fourth quarter of 1999. (Id.; Tr. 878-79). Kelly admits that he received a "good" bonus in December. (CX 25, p.22).

   Kellys pay was never decreased at any time during his employment at Lambda. (Tr. 872).

   In late January and early February 2000, Kelly alleges that events took place indicating that Kellys responsibilities and authority were being removed from him. Prevey announced that Glavicic would begin attending weekly Lab II meetings. Also, Glavicic began working on Lab II analysis, and, to come up to speed, began reviewing all of the Lab II work. (Tr. 403). Kelly viewed Glavicics entrance into the laboratory as a precursor to Glavicic taking over Kellys job. (Tr. 404). Kelly believed he was training his eventual replacement based upon a similar situation he had previously witnessed at Lambda involving two other employees. (Tr. 406-07). Glavicic testified that Prevey never informed him about any plan for him to take over Kellys job. (Tr. 124). Glavicic was told that he was helping out in Lab II to expedite the processing of late lab reports. (Id.). Bargers testimony supports this proposition, as he testified to increased production and timeliness when Kelly resigned. (Tr. 912-13). Glavicic did manage Lab II after Kelly left. (Tr. 124).

   Another change made on February 24, 2000 the day before Kelly resigned was that Prevey announced that partial results would no longer be issued to clients. (Tr. 414). Clients would only receive the final report. This rule was not aimed at Kelly, but was implemented lab-wide.

   Producing work became increasingly difficult, Kelly alleges, as the number of staff in Lab II fluctuated and, over time, decreased until only Kelly and Barger were working in the lab. When Kelly began work at Lambda, he made the fourth employee in the lab. (Tr. 321). The lab was staffed with employees until approximately December 1998, until John Haas was hired. Haas became the fifth employee in the lab. (Tr. 323). Shortly thereafter, Perry Mason was assigned to a different project. (Id.). In May 1999, Haas was transferred to another lab, leaving only three staff in the lab. (Tr. 325). Lab IIs staff was further undercut when Beth Schumaker, research engineer, became physically unable to perform some of the necessary work. (Id.). All of these events took place before the disputed GEN QAR.

   In November 1999, Schumaker left Lambda to follow her husband to another job in Indiana. (Tr. 328). From November 1999 until Kelly left the lab, only he and Barger worked in Lab II on a permanent basis. With only two staff, completing work became very difficult, according to Kelly, as QA checks became impossible to complete when one or the other was not working. (Id.). During that time, potential staff was interviewed, but none were hired. Prevey rejected the one applicant Kelly recommended. (Id.). The workload had not changed from the time Kelly started at Lambda, but the staff had fallen from four to two. (Tr. 329). The staffing situation created a lot of tension and overwork, according to Kelly. (Tr. 330).

   On February 18, 2000, Kelly arrived at work to see Prevey and Glavicic working in Lab II. They were changing the pole figure procedure, as Kelly had requested. (Tr. 398). Then, Prevey confronted Kelly about not changing the procedure as he had been instructed to in the previous week. (Tr. 736-37). Kelly responded that his actions were directed by the QA system and, furthermore, disputed that he received any clear directive to change the procedure. (Id., Tr. 399-400). Prevey states that Kelly was irrationally maintaining that GE had to be contacted before any procedure could be changed. (Tr. 738). Kelly maintains that he simply wanted to provide a copy of the procedure to GE, not request their approval. (Tr. 399-400).

   Without Kellys knowledge, this disagreement was documented in his personnel file by a memorandum written by Prevey. (Id.; CX 17, p. 36). Kelly disputes the content of the memorandum, claiming that, if he had acted as Prevey claimed, he would have violated the QA system. (Tr. 399). Kelly interprets this as evidence of Preveys continuing irrational behavior toward him. (Id.). Kelly never saw this memorandum before he resigned, despite the fact that the memorandum references a "counseling statement" that would have been required to have been reviewed with Kelly. (Id.). Prevey testified that the words "counseling statement" were added by Kathleen Bower, not him. (Tr. 730).

    The same day Prevey placed a memorandum in Kellys file regarding the thin sheet texture mounting procedure, he placed yet another memo in Kellys file regarding Kellys failure to use the semi-quantitative fluorescence analysis. (Tr. 718). When Prevey noticed the unusual conclusion of the presence of sodium chloride in one of Kellys report he was reviewing, he looked for the elemental analysis and discovered that it had not been done as required by laboratory procedures. (Id.). Kelly and Prevey had engaged in several discussions concerning Lambdas elemental analysis procedure, with Kelly unsatisfied with the procedure. (Id.). After the elemental analysis was run, no sodium chloride was present, indicating that the report was incorrect. (Id.) The report was edited and then shipped to the client, sans Kellys elemental conclusions. (Tr. 720). Prevey viewed the failure to use the elemental analysis as "insubordination." (Tr. 718). The record demonstrates that the analysis was completed before Kelly issued his report, but it was not so noted on the report submitted to Prevey. (Tr. 722-24).

   From September 1999 through February 2000, Kelly alleges that Preveys angry outbursts became increasingly severe, supposedly in an attempt to pressure Kelly into signing the QAR or to justify his eventual firing. (Complainants Post-Hearing Brief, p. 32). Kelly alleges that Prevey would blow-up over non-events. Kelly further alleges that abusive, foul language was directed toward him, and that Prevey routinely threatened to fire him. Specifically, Kelly recalls that, in a week in September, his job security was threatened every day by Prevey. (CX 25, p. 17).

   No witness corroborates Kellys stories of obscenity-laden tirades. Glavicic testified that Prevey would raise his voice, and had done so to him. (Tr. 125). Glavicic, however, testified that he never heard Prevey swear at Kelly. (Id.). Perry Mason testified that he never heard Prevey swear or be abusive toward Kelly. (Tr. 900). Barger also testified that he never heard Prevey threaten or swear directly at Kelly, although he had heard Prevey swear. (Tr. 913). Marie Marawi also testified that she never heard Prevey yell or swear at Kelly, nor had she ever witnessed Prevey act abusively toward him. (Tr. 935). No witness, except Kelly, testified to ever perceiving Prevey as physically threatening.

   To deal with the various incidents catalogued above, Kelly began the practice of writing resignation letters and bringing them to work with him, ready to be used if needed. (Tr. 342). Kelly testified to pre-writing his resignation letters after three particularly bad situations. (Id.). Kelly overrode all of them, however. (Id.) Eventually, Kelly realized he was preparing resignation letters several times a week. (Id.). On cross-examination, Kelly testified that it was reasonable to assert that he wrote approximately eight resignation letters during September 1999 that were never submitted to Prevey. (Tr. 437).

   Before his September 1999 performance evaluation, Kelly had requested the NRC to perform a technical evaluation of the GEN problem. After Kellys performance evaluation, he requested the NRC to perform a discrimination evaluation of the recent events he had endured. (CX 23, p. 4-7; Tr. 360-61). Kelly later withdrew the discrimination evaluation, primarily because of the risk of losing his job. (CX 23, p. 7; Tr. 362). In addition, the pole figure procedure was changed in early October 1999, and Kelly felt like things may be improving in the lab, further diminishing the need for a discrimination evaluation. (Tr. 367).

   On December 16, 1999, Kelly received an answer to his inquiries from the Nuclear Regulatory Commission. (CX 23, p. 11). In their final report, the Commission stated, "[W]e have concluded that the errors resulting from the texture analysis at Lambda Research in the development of engineered components is not a safety concern. NRC did not pursue the issue further since we determined it is not a safety concern." (Id.). Despite the NRCs finding of no safety concerns, Kelly persisted in his worries because he felt like the NRC evaluation was not completely thorough. (Tr. 372). Kelly worried that the NRCs evaluation was limited to the June testing Lambda performed. Kelly admitted that GEN had received a corrected report to let them know that the original results of the GEN test were incorrect. (Tr. 374). Kelly was concerned with past testing and the possibility that some of that testing was subject to the 10 C.F.R. § 21 requirement. (Tr. 375).

   On February 16, 2000, the new quality assurance manager, Marie Marawi, brought another version of the GEN QAR for Kelly to sign. That same day, Chris Barger came to Kelly with another zirconium specimen and informed him that the contact cement failed again. (CX 19). Kelly informed Marawi that he could not sign the QAR until he reviewed it. (CX 25). Before he signed the QAR, Kelly attempted to contact Prevey to discuss the contact cement problem. This attempt led to a confrontation in the lab between Kelly and Prevey regarding the efforts of Barger and Kelly on the GEN studies. On February 18, Prevey took the QAR that Marawi had delivered back from Kelly. Because of the confrontation in the lab, Kelly had not had a chance to review or sign the new draft of the QAR.

   The QAR next came up for discussion on February 25, 2000 in a meeting between Kelly, Prevey, Glavicic, and Barger. Prevey handed Barger and Kelly two quality incident reports to sign, one of which was the QAR relating to the GEN test from June 1999. (Tr. 416, 680). The other QAR concerned an unrelated project. Prevey indicated that they had five minutes to read and sign the QAR. (Tr. 418). Then, as Kelly was reading the QAR in Preveys office, Prevey told him that Kelly would sign the QAR before he left his office. (Id.). After a period of time, Prevey gave Kelly one hour to look over the QAR and sign it, unedited. (Tr. 418-19). When Kelly reviewed the QAR, he identified, from his viewpoint, three problems: 1) there was no mention of the failure of the peak finding program; 2) there was no indication that any review of older data for the problems that had been identified would be done; and 3) there was no indication that GEN was to be notified of the full extent of the problem with procedures used in past work, which would notify GEN that it should evaluate the problem. (Complainants Post-Hearing Brief, 46-47). Kelly refused to sign the QAR. (Tr. 424). Kelly went back to Preveys office, and, when Prevey asked for the QAR, he handed Prevey his resignation. (Tr. 423, 438, CX 17). At no time did Prevey solicit Kellys resignation.

VI. CONCLUSIONS OF LAW

   A. Applicable Law

   Since this case has been fully tried on the merits, the relevant inquiry is whether Kelly prevailed by a preponderance of the evidence on the ultimate question of liability. Carroll v. Bechtel Power Corp., Case No. 91-ERA- 46, Secy Dec., Feb. 15, 1995, slip op. at 9-11, affd Carroll v. U.S. Dept. of Labor, 78 F.3d 352 (8th Cir. 1996); Adornetto v. Perry Nuclear Power Plant, Case No. 97-ERA-16, ARB Case No. 98-037, Fin. Dec. and Ord., Mar. 31, 1999, slip op. at 3. See also Jackson v. Ketchikan Pulp Co., 93-WPC-7 and 8, Secy Dec., Mar. 4, 1996, slip op. at 4-5 n.1. Thus, it must be determined whether Kelly has proven, by a preponderance of the evidence, that he engaged in protected activity under the ERA, that Lambda Research took adverse action against Kelly, and that Kellys ERA-protected activity was a contributing factor in the adverse action that was taken. Dysert v. Secretary of Labor, 105 F.3d 607 (11th Cir. 1997); Simon v. Simmons Foods, 49 F.3d 386 (8th Cir. 1995); Ross v. Florida Power and Light, Case No. 96-ERA-36, ARB Case No. 98-044, Fin. Dec. & Ord., Mar. 31, 1999, slip op. at 6. See 42 U.S.C. §5851(b)(3)(C). Once a case has been tried fully on the merits, it no longer serves any analytical purpose to address and resolve the question

of whether the complainant presented a prima facie case. Instead, the relevant inquiry is whether the complainant prevailed by a preponderance of the evidence on the ultimate question of liability. Carroll v. Bechtel Power Corp., 1991-ERA- 46 slip op. at 9-11 (Secy, Feb. 15, 1995), affd Carroll v. U.S. Dept. of Labor, 78 F.3d 352 (8th Cir. 1996).

   In order to prevail in a case based upon circumstantial evidence of retaliatory intent, the Complainant must establish by a preponderance of the evidence that the employer was subject to the Act, that he was engaged in activity protected under the Act, that he was subjected to adverse employment action, that Respondent was aware of the protected activity when it took the adverse employment action, and that the protected activity was the reason for the adverse action. See Trimmer v. United States Dept of Labor, 174 F.3d 1098, 1101-02 (10th Cir. 1999); Seater v. Southern California Edison, 95-ERA-13 at 14 (ARB Sept. 27, 1996). Complainant may carry his burden of proof on any element of a discrimination claim by direct or circumstantial evidence. Bartlik v. Tennessee Valley Auth., 88-ERA-15 at 2 (Secy Apr. 7, 1993), affd sub nom. Bartlik v. United States Dept of Labor, 73 F.3d 100 (6th Cir. 1996). It is not sufficient for Complainant to establish that the Respondents proffered reasons for the adverse action are unbelievable; he must establish intentional discrimination in order to prevail. See Leveille v. New York Air Natl Guard, 94-TSC-3 at 4 (Secy Dec. 1, 1995).

   Initially, I note that my jurisdiction is limited by law in this case to deciding only whether the complainant was discriminated against because he engaged in protected activity under the ERA. I am limited to deciding only this issue and cannot consider whether the employer acted properly in making decisions unrelated to the complainants protected activity. Likewise, I do not have the authority to decide whether the complainants supervisor acted improperly unless those actions were related to the protected activity under the ERA. My inquiry must focus solely on whether the complainants protected activity was the reason for the adverse actions taken by Lambda Research.

   A. Lambda Research as "Employer" Under the Act

   A necessary element of a valid ERA claim under the employee protection provision is that the party charged with discrimination is an employer subject to the Act. Mackowiak v. University Nuclear Systems, Inc., 735 F.2d 1159, 1162 (9th Cir. 1984); DeFord v. Secretary of Labor, 700 F.2d 281, 286 (6th Cir. 1983). Employers under the ERA are licensees, or applicants for a license, of the Nuclear Regulatory Commission, and their contractors and subcontractors. 42 U.S.C. § 5851(a); Billings v. OWCP, 91-ERA-35 (Secy Sept. 24, 1991), slip op. at 2; Wensil v. B.F. Shaw Co., 86-ERA-15, 87-ERA-12, 45, 46, 88-ERA-34 (Secy Mar. 29, 1990), slip op. at 11, affd sub nom. Adams v. Dole, 927 F.2d 771, 776 (4th Cir. 1991), cert. denied, 116 L. Ed. 2d 90 (1991);. Nichols v. Bechtel Construction, Inc., 87-ERA-44 (Secy Oct. 26, 1992), slip op. at 8.

   The evidence clearly indicates that Lambda Research is subject to the Act. The GEN purchase orders clearly indicate that GEN is subject to the Act, and the evidence of record clearly demonstrates that Lambda acted as a subcontractor to GEN. A subcontractor of an NRC licensee is an employer subject to ERA, 42 U.S.C. § 5851(a). Mackowiak v. University Nuclear Systems, Inc., 735 F.2d 1159 (9th Cir. 1984). Lambdas own procedures included portions of the Energy Reorganization Act. (CX 36, p. 56-57).

   B. Protected Activity

   In a retaliation claim under the ERA, the protected activity must relate to nuclear or environmental concerns or must further the purposes of that act. See 42 U.S.C. § 5851(1)-(3); 29 C.F.R. § 24.2; Tyndall v. United States Environmental Protection Agency, 93-CAA-6 (ARB June 14, 1996). To constitute protected activity under the ERA, an employees acts must implicate safety definitively and specifically. American Nuclear Resources v. U.S. Department of Labor, 134 F.3d 1292 (6th Cir. 1998). The ERA does not protect every incidental or superficial suggestion that somehow, in some way, may possibly implicate a safety concern. Stone & Webster Engineering Corp. v. Herman, 115 F.3d 1568, 1574 (11th Cir. 1997). Not every act an employee commits under the auspices of safety is protected under the whistleblower provisions of the ERA. Stone & Webster Engineering Corp. v. Herman, 115 F.3d 1568, 1574 (11th Cir. 1997). Raising particular, repeated concerns about safety issues that rise to the level of a complaint constitutes protected activity under the ERA. Bechtel Construction Co. v. Secy. of Labor, 50 F.3d 926, 931 (11th Cir. 1995). However, making general inquiries regarding safety issues does not qualify as protected activity. Id..

   The parties dispute whether Complainant engaged in any protected activity by refusing to sign the September and February QAR. Specifically, Kelly alleges three protected activities: 1) Kellys submission of the QAR report in July 1999 recommending various changes in the laboratory originating from problems with the GEN texture analysis; 2) Kellys refusal to make the changes suggested by Prevey and his subsequent re-submission of the QAR without said changes; and 3) Kellys further refusal to sign the QAR in February 2000. Each of these activities will be reviewed in turn. Each action will be reviewed individually.

1. Kellys submission of the QAR report in July 1999 recommending various changes in the laboratory originating from problems with the GEN texture analysis

   Kellys submission of the QAR clearly constituted protected activity. A legal dispute whether purely internal complaints or safety reports to management constitutes protected activity under the ERA no longer exists because the 1992 amendments to the ERA explicitly include an employees notification to his or her employer of an alleged violation of the ERA. See Section 2909(a) of the Comprehensive National Energy Policy Act of 1992, Pub. L. No. 102-486, 106 Stat. 2776, 3123.

   In Hermanson v. Morrison Knudsen Corp., 94-CER-2 (ARB June 28, 1996), the Board found that the ALJ correctly recognized that internal complaints were protected under the whistleblower provisions of the pertinent environmental statutes. The Board observed that an informal and internal safety complaint may constitute protected activity. Slip op. at 5, citing, Nichols v. Bechtel Construction, Inc., 87-ERA-44, slip op. at 10 (Secy Oct. 26, 1992) (employees verbal questioning of foreman about safety procedures constituted protected activity), appeal dismissed, No. 92-5176 (11th Cir. Dec. 18, 1992); Dysert v. Westinghouse Electric Corp., 86-ERA-39, slip op. at 1, 3 (Secy Oct. 30, 1991) (employees complaints to team leader protected).

   The scientific soundness of Kellys assessment of the problems does not alter the fact that his initial QAR report and its suggested changes constituted protected activity. The policy underlying the ERA and the employee protections it affords are designed to promote and encourage the full, unfettered flow of safety-related information and safety concerns not only to employers but the NRC as well. Nothing in the language of the Act conveys any intent to restrict its coverage only to those concerns which address actual violations or imminently hazardous conditions. Accordingly, in deference to the policy objectives of the ERA and similar enactments, the precedents which guide this adjudication have not required the ultimate substantiation of the employees concerns. Passaic Valley Sewage Commrs. v. Dept. Of Labor, 992 F.2d 474 (3rd Cir., 1993); Yellow Freight Sys., Inc. v. Martin, 954 F.2d 353 (6th Cir., 1992); Oliver v. Hydro-Vac Services, Inc., 91 SWD 1 (Sec. 11/1/95); Aurich v. Consolidated Edison Co., 86 ERA 2 (Sec. Order, 4/23/87).

   It is sufficient that a Complainant have a "reasonable belief" or a "good faith perception," that a potential violation has occurred or might occur or a potentially hazardous situation may exist. Passaic Valley, supra; Minard v. Nerco Delamar Co., 92 SWD 1 (Sec., 1/25/94); Yellow Freight, supra; Oliver, supra; Aurich, supra. Thus, the courts have specifically protected the disclosure of a "possible violation" even when a subsequent NRC investigation revealed the employee was mistaken. Kansas Gas & Electric Co. v. Brock, 780 F.2d 1505 (10th Cir., 1985), cert. denied, 478 U.S. 1011 (1986); Mackowiak v. University Nuclear Sys., Inc., 735 F.2d 1159 (9th Cir., 1984). In the instant case, Kellys previous work for GEN included the 10 C.F.R. § 21 requirement, and Kellys concern was specifically related to previous work Lambda had performed for GEN. Despite the fact that only a small percentage of the purchase orders received the 10 C.F.R. § 21 requirement (2 out of 21), it was not unreasonable for Kelly to believe that previous work on the purchase orders that did contain the safety regulation may have been affected by the faulty procedures or their execution. Kellys concern was imminently reasonable at this specific point in time.

   Respondent argues that Kelly did not engage in protected activity because the GEN QAR submitted in July 1999 did not implicate or communicate to Prevey that Kelly was concerned about nuclear safety. (Respondents Post-Hearing Brief, p. 6-7, citing Makam v. Public Service Electric & Gas Co., 1998-ERA-22 (2001)). I find this argument untenable. The QAR submitted by Kelly explicitly references the 10 C.F.R. § 21 requirement. (CX 1, p. 3). Furthermore, Kellys July 2, 1999, memorandum to Prevey, addressing the exact same problem, clearly identified that the presence of the 10 C.F.R. § 21 requirement in previous purchase orders from GEN was a driving force in Kellys concern and recommendations. (CX 19, p. 5).

   I find that Kellys submission of the GEN QAR in July 1999 was protected activity.

2. Kellys refusal to make the changes suggested by Prevey and his subsequent re-submission of the QAR without said changes

   Kellys refusal to make the changes requested by Prevey was not reasonable, and, thus, it was not protected activity. See Cox v. Lockheed Martin Energy Systems, Inc., 1997-ERA-17 (1999). When Kelly spoke with Prevey regarding his recommendations and received the edits by Prevey to the original QAR, Kelly was confronted with sufficient explanations for Preveys changes.

   First, Prevey rejected Kellys suggestion to notify GEN of the possibility of pole figure data distortion due to the lack of sample flatness because Prevey believed that the client had already been notified and the data already reviewed. Preveys conclusion was reached because he assumed that Kelly was referring to the 90 degree rotation error that had been detected before the sample mounting error had been located. (Tr. 653-58). Furthermore, Prevey knew that the data had already been reviewed because it is reviewed in real-time when the reports themselves are generated by comparing the report conclusions with archival data from the round robin exchange of data. (Tr. 686-91). Lambda had already contacted GEN regarding that issue, and thus Kellys recommendation was moot. Prevey assumed that Kelly was referring to the 90 degree rotation problem because he saw no basis for contacting GEN regarding the sample mounting problems on the June purchase order. (Tr. 653-55). Preveys determination that there was no basis to contact GEN had been communicated to Kelly. (Tr. 659-660).

   Prevey rejected Kellys final recommendation that two procedures being revised to ensure sample flatness because a procedure to ensure sample flatness was already in place. (Tr. 666). If procedures were followed correctly, there was no need to alter the existing procedures. Prevey considered Kellys concern that errors in previous work may have gone undetected as pure speculation. (Tr. 659; 686-91). Preveys explanation to Kelly sufficiently and completely obviates the need to revise the procedures, and Kellys failure to follow Preveys edits and insistence to keep his recommendations in the QAR were unreasonable.

   Durham v. Georgia Power Company and Butler Service Group, 86-ERA-9 (ALJ October 24, 1986), is directly on point. In Durham, the administrative law judge found that the complainant was not engaged in protected activity when he refused to sign a particular section on a quality control form. The complainant argued that by not signing the form, he was refusing to violate applicable quality control procedures, and therefore, his suspension was a result of protected activity. Prior to the suspension, the management had explained to the complainant that his signature was required to proceed with work and the signature did not violate or compromise the integrity of the quality control procedures. The administrative law judge noted that, had management requested the complainant to falsify control documents or violate quality control procedures in any way, his refusal would constitute protected activity. The administrative law judge found, however, that there was no request for a falsification of the document nor any compromise of quality control procedures; rather, the complainant merely did not agree with the managements request or explanation. The administrative law judge found that the complainant, after the explanation and before his suspension, knew or should have known that the requested signature would not violate applicable quality control procedures. His refusal to sign the form after such explanation was not protected activity, it was merely refusing to obey a valid order. Accordingly, the complainant failed to establish a prima facie case that his suspension was in retaliation for protected activity.

   In the instant case, Kelly was confronted with sufficient explanation as to the requested changes as to render his refusal to edit the QAR mere unwillingness "to obey a valid order."

3. Kellys further refusal to sign the QAR in February 2000

   Kellys second refusal to sign the QAR is also not a protected activity.

   Beyond Preveys explanation of the requested changes, Kelly was informed in December 1999 that the Nuclear Regulatory Commission [NRC] found no nuclear safety hazard in the sample mounting error that had occurred at Lambda. (Tr. 369, 444; RX 24). Kelly maintains that the NRCs report was incomplete and inadequate. (Tr. 371-72; Complainants Post-Hearing Brief at 40). Kellys chief complaint with the NRCs evaluation of the situation at Lambda was the NRC did not know for what purpose GEN would use the zirconium tubes. (Id.). The NRCs report however indicates that, regardless of the end use, no safety issues were implicated because "we believe that texture analysis, in general, cannot be used to inadvertently qualify unacceptable material." (RX 24, p.1). The report stated, "Therefore, we have concluded that errors resulting from the texture analysis at Lambda Research in the development of engineered components is not a safety concern." (Id.). Phillips v. Stanley Smith Security, Inc., ARB No. 98-020, ALJ No. 1996-ERA-30 (ARB Jan. 31, 2001), affirmed that NRC approval of a procedure or activity does not necessarily render a

complainants concerns unreasonable, but that conclusion was based upon the fact that the Complainant in that case suspected intentional deception upon the part of the Respondent with the NRC. No such fact pattern exists in the instant case. The NRCs report, when combined with Preveys numerous discussions with Kelly, should have obviated any concern for nuclear safety that Kelly possessed. I find that to advance that he continued to have such concerns is unreasonable.

   C. Adverse Employment Action

   To constitute an adverse action, Complainant must demonstrate by a preponderance of the evidence that the action had some adverse impact on his employment. See Trimmer, 174 F.3d at 1103 (citing Montandon v. Farmland Indus., Inc., 116 F.3d 355, 359 (8th Cir. 1997)); but see DeFord v. Secretary of Labor, 700 F.2d 281, 287 (6th Cir. 1983))(economic loss is not required for action to be adverse). The governing regulations define discrimination or an adverse employment action very broadly. See 29 C.F.R. 24.2(b)("Any employer is deemed to have violated the particular federal law and the regulations in this part if such employer intimidates, threatens, restrains, coerces, blacklists, discharges, or in any other manner discriminates against any employee because the employee has [engaged in protected activity]). Activities

found to be adverse employment actions include, but are not limited to, elimination of position, threats of termination, blacklisting, causing embarrassment and humiliation, constructive discharge, and issuance of disciplinary letters.

   In Graf v. Wackenhut Services, L.L.C., 1998-ERA-37 (ALJ Dec. 16, 1999), pet. for review withdrawn, Graf v. Wackenhut Services, L.L.C., ARB Nos. 00- 024 and 25 (ARB Feb. 16, 2000), the administrative law judge found that "[t]he Tenth Circuit liberally defines the phrase adverse employment action and takes a case-by-case approach to determining whether a given employment action is adverse." Jeffries v. Kansas, 147 F.3d 1220, 1232 (10th Cir. 1998) (employment action is not required to be materially detrimental). The judge wrote:

In Jeffries, for example, verbal interrogation and reprimand were sufficient to constitute adverse employment actions even though said actions did not actually have an adverse impact on the terms and conditions of the employees employment. Id. Other examples of adverse actions include "decisions that have demonstrable adverse impact on future employment opportunities or performances, demotions, adverse or unjustified evaluations or reports, transfer or reassignment of duties, [and] failure to promote." Fortner v. Kansas, 934 F. Supp. 1252, 1266-67 (internal citations omitted), affd sub nom. Fortner v. Rueger, 122 F.3d 40 (10th Cir. 1997). Nevertheless, it is not sufficient for a complainant to simply testify that he did not like the action or wished that the action had not occurred. Trimmer, 174 F.3d at 1103 (citing Greaser v. Missouri Dept of Corrections, 145 F.3d 979, 984 (8th Cir. 1998). See also Fortner, 934 F. Supp. at 1266-67 ("[N]ot everything that makes an employee unhappy is an actionable adverse action."). Speculative harm will not constitute adverse employment action. Id.

Id.

   Complainant alleges that the adverse employment actions he suffered as a result of his engagement in protected activity was a constructive discharge. (Complainants Post-Hearing Reply Brief, p. 8-13). Within the complainants constructive discharge evidence exists separate incidents

which standing alone may constitute adverse actions. After analysis of Complainants constructive discharge claim, the actions shall be reviewed independently.

   1. Constructive Discharge

   Complainant resigned his employment on February 25, 2000, but Complainant alleges that Respondent engaged in a pattern of activity so egregious as to render his resignation involuntary.

   To establish a constructive discharge:

the employee must show that working conditions were rendered so difficult, unpleasant, unattractive or unsafe that a reasonable person would have felt compelled to resign. ... It is insufficient that the employee simply feels that the quality of his work has been unfairly criticized.

Mosley v. Carolina Power & Light Co., 94-ERA-23 (ARB Aug. 23, 1996)(citing Henn v. National Geographic Society, 819 F.2d 824, 829-30 (7th Cir. 1987)). See also Wilson v. Firestone Tire & Rubber Co., 932 F.2d 510, 515 (6th Cir. 1991)(stating that constructive discharge occurs when employment is "so difficult or unpleasant that a reasonable person in the employees shoes would have felt compelled to resign."). "Constructive discharge" assumes that a complainant was not formally discharged, the issue being whether he or she was forced to resign or whether he or she quit voluntarily. Because a finding of constructive discharge requires proving that working conditions were rendered so difficult, unpleasant, unattractive, or unsafe that a reasonable person would have felt compelled to resign, i.e., that the resignation was involuntary, Johnson v. Old Dominion Security, 86-CAA-3 to 5 (Secy May 29, 1991), slip op. at 19-22 and n.11, the adverse consequences flowing from an adverse employment action generally are insufficient to substantiate a finding of constructive discharge. Rather, the presence of "aggravating factors" is required. Clark v. Marsh, 665 F.2d 1168, 1174 (D.C. Cir. 1981). Unless constructively discharged, a complainant is not eligible for post- resignation damages and back pay or for reinstatement.

   Circumstances held sufficient to render resignation involuntary include a pattern of discriminatory treatment and "locking" an employee into a position from which no relief seemingly can be obtained. Clark v. Marsh, 655 F.2d at 1174; Satterwhite v. Smith, 744 F.2d at 1382-1383. Transfer from a supervisory position to a "dead-end position requiring [the employee] to do virtually nothing was a form of enforced idleness both humiliating and detrimental." Hopkins v. Price Waterhouse, 825 F2d 458, 473 (D.C. Cir. 1987), revd on other grounds, 109 S. Ct. 1775 (1989) (constructive discharge occurred where employee subjected to what any reasonable senior manager in her position would have viewed as "career-ending action"). [additional citations omitted].

   The Secretary addressed the doctrine of constructive discharge in Perez v. Guthmiller Trucking Co., Inc., 87-STA-13 (Secy Dec. 7, 1988), slip op. at 24-27; Taylor v. Hampton Recreation and Hampton Manpower Services, 82-CETA-198 (Secy Apr. 24, 1987), slip op. at 7-9; Hollis v. Double DD Truck Lines, Inc., 84-STA-13 (Secy Mar. 18, 1985), slip op. at 8-9, employing in those cases an objective standard adopted under antidiscrimination legislation. See Simpson v. Federal Mine Safety & Health Rev. Comn, 842 F.2d 453, 461-463 and nn.8, 9 (D.C. Cir. 1988); see also Wilson v. Firestone Tire & Rubber Co., 932 F.2d 510, 515 (6th Cir. 1991)(applying objective standard test in determining constructive discharge). In Taylor v. Hampton Recreation, the Secretary concluded that the employees resignation was coerced where he had endured a pattern of abuse by his immediate supervisor, the supervisor repeatedly had refused to provide him with guaranteed job training, the confrontations and threats of imminent discharge adversely affected the employees health, and top management had manifested insensitivity and a marked lack of response to the employees grievances and requests for assistance.

   The record does not support the complainants allegation of constructive discharge. The record does reveal certain facts, however. First, there is no question that Prevey was a demanding boss. Several witnesses testified to this. There is also no question that Prevey occasionally yelled at his employees. Prevey admitted to this under oath. Those conclusions, however, are the extent to which the record yields negative evidence about the work environment at Lambda Research. None of the specific examples that Kelly advances as demonstrative of a working environment "so difficult, unpleasant, unattractive, or unsafe that a reasonable person would have felt compelled to resign" withstand the scrutiny of the weight of the evidence.

   Kelly advanced that Prevey unreasonably shortened the delivery time on the Western Zirconium project, but the evidence clearly demonstrates that the reduction in time for the project was at the clients request. Furthermore, there is no indication that this one example of a shortened deadline made completion of Kellys jobs on a daily basis impossible or unduly burdensome.

   Kelly alleges that Preveys reaction to the Specs Milling machine incident was unreasonable, but no other employees who witnessed that incident corroborate Kellys allegations of swearing. Furthermore, Kelly admits that the ultimate responsibility and failure rested with him, and Preveys notation of Kellys response in his personnel file is patently reasonable.

   Kelly alleges his responsibilities were decreased with the intervening presence of Glavicic in the lab, but there is no indication that Glavicic was in the lab to overtake Kellys job, beyond Kellys assumptions based upon what he perceived were Preveys tactics with previous employees. I grant Kellys perceptions on this issue little weight. Glavicic and Prevey testified that Glavicic was not there to overtake Kellys job, asserting that the only reason for his presence was to move backed-up work out of the lab. Testimony does reveal that the lab was busy, and that QA checks were more difficult with only two people. However, the record clearly reveals that the size of the lab fluctuated over time, and that events conspired in the lab to reduce its personnel that were out of the control of Prevey, such as Shoemakers resignation to follow her husband to another job. To the contrary, in the time period between Kellys refusal to sign the QAR in September 1999 through his resignation, Prevey seemed to support Kellys responsibilities in the lab by giving him bonuses, following some of his suggestions, and taking him to Colorado to a seminar on x-ray diffraction. There is simply no indication that Prevey was acting to fire Kelly.

   The crux of Kellys constructive discharge argument appears to be that Prevey was rude and irrationally cruel to Kelly. The record supports Kellys allegation that Prevey was a tough person to work for, who occasionally lost control of his temper, but no evidence demonstrates a working environment "so difficult, unpleasant, unattractive, or unsafe that a reasonable person would have felt compelled to resign." It is clear that Kelly and Prevey engaged in heated arguments. These types of arguments, however, are not uncommon to the work force, especially in the private sector where the demands of the market push companies to produce quality products in little time. There is no evidence that any of the "angry outbursts" of Prevey crossed the line, albeit they may be indicative of an inordinate lack of control. The record indicates that Kelly was not alone as a target of these "outbursts." Glavicic and Barger both testified to witnessing such outbursts; Glavicic admitting to being the recipient of some. At times, the heated exchanges were brought on by Kellys actions. Kellys failure to follow the fluorescence analysis procedure and his reaction to Preveys inquiry into maintenance responsibilities would reasonably draw harsh responses from employers seeking compliance with rules and procedures. Other times, such as the developer solution incident, the word processing template incident, and the timecard incident, it appears as though Prevey lost control of his temper before he had a complete understanding of the situation. A loss of temper, however, does not produce a working environment "so difficult, unpleasant, unattractive, or unsafe that a reasonable person would have felt compelled to resign." If it did, almost every American worker has likely been constructively discharged. That non sequitur speaks for itself.

   The record does not support Kellys allegation that he was threatened with physical harm. Preveys testimony concerning his recitation of a safety incident occurring years earlier in the lab necessitating his possession of a gun at the lab one night makes any belief that Prevey had a gun at the lab at all times unreasonable. Kelly himself denies such a belief, but advances that Preveys casual references to "shooting" are indicative of Preveys treatment of him. I find Kellys testimony on this subject lacking credibility and I afford it no weight.

   Kelly asserts that a change in company policy allowing only final reports to be sent to clients is further evidence of a work environment that is "so difficult, unpleasant, unattractive, or unsafe that a reasonable person would have felt compelled to resign." Kelly believed this action was to stifle his efforts to contact GEN regarding the zirconium testing. This assertion is unsupported by reason and case law. See Mosbaugh v. Georgia Power Co., 91-ERA-1 at 9 (Secy Nov. 20, 1995)(not an adverse action when complainant was treated like other similarly situated employees under company policy); Moody v. Tennessee Valley Auth., 91-ERA-40 at 2 (Secy Apr. 26, 1995). As the policy was company-wide, it cannot be an "adverse action" in retaliation for protected activity.

   Kelly also points to his lower performance evaluation and his small third-quarter 1999 bonus as evidence that Prevey was making it impossible to succeed. Any evidence of wrongdoing that Kelly argues flows from his third-quarter bonus has been completely obviated by evidence in the record that 1) Kelly received a large December 1999 bonus after his third-quarter bonus and, more importantly, 2) Kellys low third-quarter bonus was a product of his own discretion, not Preveys. Any argument that Kellys third-quarter bonus evinces retaliatory animus is a misrepresentation.

   Kellys September 1999 performance evaluation was markedly lower than his July 1999 evaluation. Two factors contribute to this result. First and foremost, Kellys September evaluation was the first evaluation that Prevey had been afforded the opportunity to supervise Kelly the entire evaluation period. Secondly, as Prevey testified, Prevey is historically easier on his first evaluation as a way of promoting better performance. Kelly did rank lower on eight of the fourteen categories in his September evaluation compared to his July evaluation, but the marks are not substantially lower. Furthermore, no tangible detriment redounded from the lower review. Kelly was not put on probation, denied bonuses, or placed on a performance improvement plan. See Ilgenfritz v. U.S. Coast Guard Academy, ARB No. 99-066, ALJ No. 1999-WPC-3 (ARB Aug. 28, 2001) (holding that a negative performance evaluation, absent tangible job consequences, is not an adverse action).

   While Kellys cross-examination of Prevey did demonstrate that there seemed to be little basis for his attendance to be graded harsher than after his July 1999 review, there is no evidence that this was not a simple, unintentional mistake. Preveys explanation on this particular mark is neither coherent nor satisfactory, but it also fails to evince any discriminatory intent. (Tr. 774). The scaled grades for the attendance section (Outstanding, Very Good, Satisfactory, Below Average, and Unsatisfactory) do not include objective criteria, such as limiting an "Outstanding" rating to those employees with one or less absence. The criteria are subjective. Furthermore, Kelly still received a "Very Good" rating, literally and figuratively. There exist a myriad of explanations for the loss in grade, the vast majority of which evince no discriminatory intent, such as 1) Prevey was a tougher grader after the first performance review, 2) Prevey did not

refer to the attendance sheet in his assessment, or 3) Preveys criteria for judging attendance of his supervisors toughened over time. Prevey himself admitted he was a "demanding" boss, but "fair." (Tr. 850). Marie Marawi also testified that Prevey was demanding but fair. (Tr. 934). Prevey admitted that the marks are necessarily subjective. There is absolutely no evidence that the lower September evaluation moved Kelly any closer to a suspension or a discharge. I will weigh the evidence of the lowered attendance marks, despite what appears to be "improved" attendance for Kelly, as circumstantial evidence of discriminatory intent, but I will grant it little probative weight as the context of the performance evaluations and Preveys explanations sufficiently convince me that this "evidence" of discrimination is entitled to little weight.

   Furthermore, the four memoranda placed in Kellys personnel file without his knowledge could not have contributed to making the work environment "so difficult, unpleasant, unattractive, or unsafe that a reasonable person would have felt compelled to resign." As Kelly did not know the memoranda existed, the memoranda could not have contributed to how Kelly perceived his work environment. While the documents are evidence of Preveys problems with Kellys work performance, they cannot serve as direct proof of constructive discharge without Kellys knowledge of them. Cf. Doria v. Cramer Rosenthal McGlynn, Inc., 942 F.Supp. 937, 947 (D.D.N.Y. 1996).

   Kelly argues that Preveys time limitations on his signing of the QAR on February 25, 2000 placed him in the impossible position of either signing what he considered a QAR that covered up problems or, conversely, resigning. It is true that intimidation and coercion brought to bear on a complainant to rescind a health or safety complaint constitutes adverse employment action. Nathaniel v. Westinghouse Hanford Co., 91- SWD-2 (Secy Feb. 1, 1995), slip op. at 13. In the instant case, however, no such intimidation or coercion was wrought upon the complainant. No ultimatums were issued. Complainants job security was not threatened. He was not told that he would be fired or demoted if he did not sign the QAR in the time allotted. Preveys testimony reveals that he was tired of the issues surrounding the GEN QAR, and the objective evidence supports Preveys assessment of the lack of a safety issue in the results sent to GEN, as all other results had comported with the round robin results and were checked in real-time as the

reports were generated. The only report that did not comport was the June sample, and GEN had already been notified of the sample mounting problem with that particular project. Furthermore, Prevey has established that if the existing procedures were followed correctly, there was no need to change them as Kelly advanced. It was simply repetitive. Kelly was confronted with this evidence, but chose to ignore it. Kellys obstinance in maintaining his opinion does not transform Preveys demand to sign the QAR into an adverse employment action.

   Prevey was correct in his assessment of the safety situation, as demonstrated by the NRCs December report. Preveys action of demanding a signature on the report was simply an employer reaching a final decision about a problem and refusing to waste more time on a non-issue. As Kelly was not threatened with any tangible detriment and was merely given a time period to sign the QAR, Preveys time limitations were not adverse employment actions. Furthermore, Preveys actions were no more intimidating or coercive than any employers demand that employees finish certain uncomplicated tasks within a reasonable time frame. As Prevey had on several occasions discussed and considered Kellys recommendations, his demand to sign the QAR was not an adverse employment action, but, rather, was imminently reasonable. See Durham v. Georgia Power Company and Butler Service Group, 86-ERA-9 (ALJ October 24, 1986).

   Any intimation on Kellys behalf that Preveys time constraints on his signing the QAR in February 2000 was a mere continuation of his September 1999 threat of firing Kelly if he wasted any more time on the QAR is 1) attenuated by the intervening five months to such a degree as to render the February confrontation a separate and distinct event, and 2) obviated by the simple fact that Preveys continual allowance of greater time for Kelly to review the QAR indicated a willingness to allow Kelly to spend more time on the issue. Furthermore, I find Preveys assertion that he had no intention of firing Kelly on February 25, 2000, credible.

   I affirm that the record demonstrates Preveys quick temper and abrasiveness, but such choleric outbursts seem to be directed at employees in general and not Kelly in particular. Furthermore, Kelly presents no evidence that comes close to demonstrating that he was subjected to such intolerable conditions that, as a reasonable person, he was left with no other option but to resign. Kelly was never demoted; his salary was not reduced; his title was not changed; and his responsibilities remained constant. There exists no evidence of aggravating factors targeted directly at Kelly which "cornered" Kelly, allowing him no option but to resign. His engagement in heated exchanges with Prevey alone fails to rise to the standard for constructive discharge.

   2. Disciplinary Action

   Four letters were placed in Complainants personnel file during his employment at Lambda, none of which he had any knowledge of their existence. The letters served as "mental notes" for Prevey concerning events that occurred in the laboratory. (Tr. 850). The letters did not result in any suspension or termination of Kelly, but, even though a disciplinary letter does not result in a firing or demotion of a complainant, such drastic action is not required to render such a letter adverse. See, e.g., Self v. Carolina Freight Carriers Corp., 89-STA-9 (Secy Jan. 12, 1990), slip op. at 15 (warning letters that "served to progress [the c]omplainant toward suspension and discharge" adversely affected him even though the letters did not result in suspension or discharge); Helmstetter v. Pacific Gas & Electric Co., 86-SWD-2 (Secy Sept. 9, 1992).

   There is, however, no objective evidence that the letters served to light the path towards termination for Kelly. Prevey testified that he had no intent of firing Kelly or forcing him to resign on the day he resigned, and that to do so would have hurt the company. (Tr. 863). Furthermore, there was no formal system in place at Lambda Research where such a document could objectively serve as ammunition for an impending suspension or termination. In addition, Kellys employment was affected in no way by the letters, as the only individual aware of the presence of the letters was Prevey. A negative performance evaluation, absent tangible job consequences, is not an adverse action. Ilgenfritz v. U.S. Coast Guard Academy, ARB No. 99-066, ALJ No. 1999-WPC-3 (ARB Aug. 28, 2001).

   3. Kelly threatened with termination if he pursued his QAR concerns

   After Prevey recognized that Kelly had failed to implement Preveys edits of the first QAR draft into the second draft, Prevey threatened to fire Kelly if he pursued his concerns or spoke to anyone in the lab concerning the safety issues Kelly believed to be present. This is unequivocally an adverse employment action. See Graf v. Wackenhut Services, L.L.C., 1998-ERA-37 (ALJ Dec. 16, 1999), pet. for review withdrawn Graf v. Wackenhut Services, L.L.C., ARB Nos. 00- 024 and 25 (ARB Feb. 16, 2000)(threats of disciplinary action and attempts to solicit Complainants compliance with Corrective Action Plans were adverse employment action). The governing regulation itself speaks to threats. See 29 C.F.R. § 24.2(b) (prohibiting employer action that "intimidates, threatens, restrains, coerces, blacklists, discharges, or in any other manner discriminates against" an employee). Preveys threat to fire Kelly was an adverse employment action.

   D. Nexus Between Protected Activity and Adverse Employment Action

   The evidence demonstrates that Kelly suffered an adverse employment action after his protected activity of reporting his safety concerns with the GEN project. Once the complainant has demonstrated that 1) he engaged in protected activity and 2) suffered adverse employment action, he must establish a nexus between the protected activity and the adverse employment action. See Bartlik v. United States Dept of Labor, 73 F.3d 100 (6th Cir. 1996). The complainant establishes such a "nexus" if he demonstrates that his engagement in protected activity was a "contributing factor" to the adverse employment action. See 29 C.F.R. Part 24.7(b) ("[A] determination that a violation has occurred may only be made if the complainant has demonstrated that protected behavior or conduct was a contributing factor in the unfavorable personnel action alleged in the complaint."); Paynes v. Gulf States Utilities Co., ARB No. 98-045, ALJ No. 1993-ERA-47 (ARB Aug. 31, 1999). The record, however, evinces no such nexus.

   When the evidence of threat of termination is reviewed, I find that there is no nexus between Kellys submission of the original draft of the GEN QAR and Preveys threat of termination.

   Preveys threat of termination came after Kelly had submitted a second draft of the GEN QAR. Kelly had failed to make the changes Prevey outlined, despite conversations with and explanations from Prevey as to his reasons for making the changes. When Prevey recognized that the changes had not been completed, he was angry and immediately informed Kelly to stop wasting time on the QAR. The evidence indicates that Preveys anger was not directed at the initial filing of the QAR, but rather the refusal of Kelly to make the changes after Prevey had explained the lack of necessity to implement Kellys recommendations. Kelly himself admits this motivation, stating, "[H]e [Prevey] just didnt want me wasting any more time on this QA incident report." (Tr. 271).

   I find it significant that Prevey had no hostile reaction to the QAR when it was first filed by Kelly. When chastising Kelly for his failure to make the changes in the QAR, Prevey at no time questioned the appropriateness of Kellys original concerns and original memorialization of those concerns in the GEN QAR. If Prevey had questioned why Kelly wasted time in filing the initial QAR draft or berated Kelly for causing trouble in the first place, then the existing evidence would indicate by a preponderance of the evidence that Kellys initial filing was a "contributing factor" to Preveys adverse action of threatening Kellys employment. No such evidence exists, however. Accordingly, I must conclude that no nexus exists between Kellys original filing of the QAR and Preveys threat of termination.

VII. CONCLUSION

   The evidence of record reveals that Mark Kelly engaged in protected activity by initially reporting his safety concerns to his employer, Lambda Research, and suffered the adverse employment action of a threat of termination. The evidence of record does not reveal, however, any nexus between Kellys protected activity and the adverse employment action he suffered by a preponderance of the evidence. Kellys filing of the initial QAR was not a contributing factor in Preveys threat of termination. Accordingly, the complainant has failed to carry his burden, and his complaint must fail.

   The complaint by Mr. Mark J. Kelly of discrimination under § 211 of the Energy Reorganization Act of 1974, 42 U.S.C. § 5851, as amended, is DISMISSED.

      JOSEPH E. KANE
      Administrative Law Judge

NOTICE: This Recommended Decision and Order will automatically become the final order of the Secretary unless, pursuant to 29 C.F.R. § 24.8, a petition for review is timely filed with the Administrative Review Board, United States Department of Labor, Room S-4309, Frances Perkins Building, 200 Constitution Avenue, N.W., Washington, DC 20210. Such a petition for review must be received by the Administrative Review Board within ten business days of the date of this Recommended Decision and Order, and shall be served on all parties and on the Chief Administrative Law Judge. See 29 C.F.R. § § 24.8 and 24.9, as amended by 63 Fed. Reg. 6614 (1998).