Paul Carter v. William L. Ball, III

Affirmed by published opinion. Judge RESTANI wrote the opinion, in which Chief Judge ERVIN and Senior Circuit Judge SPROUSE joined. OPINION RESTANI, Judge: This appeal centers on the factual sufficiency of allegations of employment discrimination made by plaintiff-appellant Paul A. Carter, an African-American, against William L. Ball, III, the Secretary of the Navy (“the Navy”). At the end of Carter’s case-in-chief, the district court granted the Navy’s motion for dismissal. Carter argues on appeal that the district court erred in refusing to admit statistical evidence of discrimination and in finding insufficient evidence to support Carter’s claims. We agree with the result reached by the district court as to evidentiary matters and the merits. We therefore affirm. Facts Carter began his career with the Department of the Navy in 1967 as an intelligence analyst. By 1988, Carter had risen to the rank of GS-12 Intelligence Operations Specialist at the Recognition Branch of the Naval Technical Intelligence Center (“NTIC”) in Washington, D.C. In March 1988, a Defense Intelligence Special Career Automated System (“DISCAS”) announcement advertised a vacancy for the position of Supervisory Intelligence Research Specialist. The DISCAS form requested applicants to rate themselves in five areas of position-related experience or knowledge and encouraged the submission of “a brief, unclassified supplementary narrative ” detailing the applicant’s qualifications. J.A. at 380. Carter applied for the position on April 7, 1988. He rated himself as having extensive experience or knowledge in imagery interpretation analysis, preparation of imagery-derived intelligence reports, and intelligence management. Id. at 381. He claimed only moderate experience or knowledge in systematic and non-systematic multi-sensor interpretation techniques and report preparation and presentation. Id. Carter also submitted a three-page resume, explaining that “[w]orking with photography on foreign surface ships and submarine units provided the bulk of [his] Naval imagery interpretation experience.” Id. at 383. The summary of his training listed a Bachelor of Science degree in Industrial Education/Social Science from Tuskegee University, Aabama and several subsequent courses and seminars in management and imagery interpretation. Id. at 384. His military experience, including work in the field of intelligence, comprised more than thirty years in active and reserve duty with the United States Air Force. Id. at 382. Out of 99 candidates who responded to the DISCAS announcement, eleven, including Carter, were determined to be eligible for the position. Three white males were recommended as highly qualified based on scores of 72 or over assigned by the rating panel. The person selected for the position received an average score of 85, while the other two highly qualified candidates received average scores of 76 and 77, respectively. Id. at 398. Carter’s average score was 63. Id. Each of the three highly qualified candidates submitted an SF-171, a standard government form providing detailed employment information, along with his application. Id. at 397. Carter did not submit such a form. Id. at 398. When asked why the three top candidates had received higher scores than Carter, the panel members replied that these candidates had provided more complete decisive information. Id. at 398-99. According to the panel members, Carter’s resume was general, vague and too brief to merit a higher ranking. Id. at 400. As instructed, the panel members determined scores based entirely on the information submitted with the applicants’ resumes and SF-171s. Id. at 399. On July 26, 1988, Carter brought the matter before an Equal Employment Opportuni *455 ty (“EEO”) counselor. Carter maintained that the Navy, in denying him the promotion, discriminated against him on the grounds of age, race, and retaliation for the filing of a prior EEO complaint in 1981. Id. at 397. The EEO counselor determined that Carter “was put at an unfair disadvantage by the actual DISCAS form itself (not really a blame of the rating panel).” Id. at 400. 1 The counselor recommended that either the form specifically mention the availability of SF-171s or the sentence that encouraged the submission of “a brief, unclassified supplementary narrative” be deleted. Id. Carter’s formal EEO complaint, filed on September 26, 1988, repeated the allegations brought before the EEO counselor. During early 1989, in approximately January or February, the previous supervisor of the branch, Lieutenant Kurt Johnson, was replaced by Lieutenant Dave Campbell. Id. at 237, 266. Whereas Lt. Johnson had treated Carter as second in line, Lt. Campbell preferred to leave a GS-5 civilian named Joan Young in charge of the branch during Lt. Campbell’s absences. Id. at 237-39. Young also took over the role of assigning work to Carter’s fellow employee, Dawn Roberts, although Roberts had previously gone to Carter for work. Id. at 241-42. According to Roberts, Lt. Campbell reprimanded Carter publicly, but called other employees into his office to comment on their work performance. Id. at 267-59. In addition, Lt. Campbell prominently displayed a poster containing a picture of a gorilla, and the statement, “I wouldn’t mind being a NOBODY if I could only get A LITTLE RECOGNITION once in awhile.” Id. at 411. Roberts, who is not African-American, considered the poster to be a derogatory reference to African-Americans. Id. at 273. Carter felt similarly. Id. at 174-76. 2 On May 2,1989, the EEO officer for NTIC issued a proposed disposition of the complaint filed by Carter in September 1988, finding no discrimination in the denial of the promotion. 3 On November 1, 1989, Carter was informed that his performance was unacceptable and that he needed to show some improvement within the next 90 days. Carter filed a second EEO complaint on February 9, 1990, alleging harassment and prejudicial assessment of his work performance based on age, race, and retaliation for his 1988 complaint. 4 Lt. Campbell wrote a memorandum to the file on March 9, 1990, stating that during the 90-day probation period Carter had not managed to complete certain tasks assigned to him without significant help from Young. Id. at 401. The Head of the Technology and Integration Department proposed terminating Carter in a letter dated May 21, 1990. Id. at 402. Among the reasons stated for termination were Carter’s inability to finish a photo analysis project assigned to him in 1987, although the deadline had been extended more than two years, and the frequent need for co-workers to assist Carter in performing his duties as computer network administrator. Id. at 402-06. The Navy instead chose to downgrade Carter from a GS-12 to a GS-9, effective July 29, 1990. Carter then decided to retire. Carter brought suit against the Navy on September 17, 1990. A claim of constructive discharge was later added to the allegations of discrimination in promotion and harassment. Carter’s motion to amend his complaint to add a claim under the Civil Rights Act of 1991 was denied. Trial before the *456 United States District Court for the District of Maryland, sitting without a jury, began on April 19, 1993. 5 The court granted the Navy’s motion for dismissal at the end of Carter’s case-in-chief on April 23, 1993, finding “no evidence ... of any discrimination whatsoever.” Id. at 369. 6 Carter now appeals, contending that 1) the district court improperly excluded statistical evidence of discrimination and 2) the evidence introduced by Carter at trial was sufficient to withstand a motion to dismiss. Discussion 1. Exclusion of Statistical Evidence Carter argues that the court erred in enforcing a blanket exclusion against all statistical evidence, rather than examining each proffer to determine its relevance. The dispute centers on Plaintiffs Exhibit 50, an organizational flow chart of the Analysis Directorate of NTIC, which purportedly demonstrates a statistical imbalance in the Navy’s promotional practices. J.A. at 409. Although the exhibit was admitted to show Carter’s relative position in the hierarchy, the judge refused to admit it as statistical evidence. Id. at 57-58, 62. We review the district court’s evidentiary rulings for abuse of discretion. See Martin v. Deiriggi, 985 F.2d 129, 137 (4th Cir.1992). Statistics can provide important proof of employment discrimination. Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 307, 97 S.Ct. 2736, 2741, 58 L.Ed.2d 768 (1977) (considering allegations of a pattern or practice of racial discrimination). In disparate treatment eases, 7 we consider statistical evidence to be “unquestionably relevant.” Ardrey v. UPS, 798 F.2d 679, 684 (4th Cir. 1986) (quoting Diaz v. AT & T, 752 F.2d 1356, 1362 (9th Cir.1985) (finding statistics on hiring and promotion patterns relevant to a claim of discriminatory failure to promote)), cert. denied, 480 U.S. 934, 107 S.Ct. 1575, 94 L.Ed.2d 766 (1987). Such evidence may be used to establish an inference of discrimination as an element of plaintiffs prima facie case, id., or to demonstrate that an employer’s stated nondiscriminatory reason for its action is in reality a pretext. McDonnell Douglas Corp. v. Green, 411 U.S. 792, 804-05, 93 S.Ct. 1817, 1825, 36 L.Ed.2d 668 (1973); Ardrey, 798 F.2d at 684. This court’s broad pronouncement that statistical evidence is “unquestionably relevant” in a Title VII case cannot be read to foreclose the exclusion of evidence with little or no probative value. See Williams v. Cerberonics, Inc., 871 F.2d 452, 455 n. 1 (4th Cir.1989). The usefulness of statistics depends on the surrounding facts and circumstances. International Brotherhood of Teamsters v. United States, 431 U.S. 324, 340, 97 S.Ct. 1843, 1857, 52 L.Ed.2d 396 (1977). In a case of discrimination in hiring or promoting, the.relevant comparison is between the percentage of minority employees and the percentage of potential minority applicants in the qualified labor pool. See Hazelwood, 433 U.S. at 308, 97 S.Ct. at 2742 (comparing proportion of African-American teachers employed by school district to proportion of qualified African-American teachers in surrounding area); Teamsters, 431 U.S. at 337 & n. 17, 97 S.Ct. at 1855 & n. 17 (comparing racial composition of nearby metropolitan area with racial composition of manual workforce). The mere absence of minority employees in upper-level positions does not suffice to prove a prima facie case of discrimination without a comparison to the relevant labor pool. See Moore v. Hughes Helicopters, Inc., 708 F.2d 475, 484 (9th Cir.1983); see also Hagans v. Andrus, 651 F.2d 622, 627 (9th Cir.) (regarding low percentage of women in high-level positions as “meaningless” without evidence of pool of qualified women *457 applicants), cert. denied, 454 U.S. 859, 102 S.Ct. 318, 70 L.Ed.2d 157 (1981). Moreover, if a plaintiff offers a statistical comparison without expert testimony as to methodology or relevance to plaintiffs claim, a judge may be justified in excluding the evidence. Williams, 871 F.2d at 455 n. 1. In the ease at bar, Carter sought to introduce Plaintiffs Exhibit 50 as evidence that none of the thirty managerial positions in the Analysis Directorate of NTIC were filled by African-American personnel. As explained by plaintiff at oral argument, the Analysis Directorate includes some entry-level management positions at the same level as the vacancy for which Carter applied. The exhibit lists names of personnel acting in Analysis Directorate positions, but makes no mention of their race. Even if we accept the assertion that none of the positions was filled by an African-American, we have found no proffer of supporting evidence relating to the pool of African-Americans qualified for those positions. 8 Carter argues that we should infer the presence of qualified African-Americans in the labor pool because Carter himself was qualified. He also points to the “inexorable zero,” ie., the total lack of African-Americans in the Analysis Directorate, which he believes to be infallible proof of discriminatory motive. Carter attributes the absence of proffers on the record to the district court’s “blanket exclusion” of statistical evidence. We are not persuaded by his arguments. First, a plaintiff cannot avoid his obligation to demonstrate the existence of qualified minorities in the labor pool by pointing to his own qualifications. To do so would obviate the necessity for a rigorous statistical comparison, in contravention of established precedent. Second, Carter’s assertion that a zero needs no comparison directly contradicts the proposition that the mere absence of minority employees is not enough to constitute a prima facie case of discrimination. See Moore, 708 F.2d at 484. Third, we can assign no error to the exclusion of evidence, even a “blanket” exclusion, where the plaintiff has not made a specific proffer to the court. See Fed.R.Evid. 103(a)(2). The district court did not abuse its discretion in excluding the proffered statistical evidence from consideration at trial. 9 2. Factual Sufficiency of Alleged Discrimination A district court sitting without a jury may enter judgment as a matter of law against a party on any claim once the party has had a full opportunity to present evidence on that claim. Fed.R.Civ.P. 52(c). 10 The court’s findings of fact will be reviewed under the clearly erroneous standard. Fed.R.Civ.P. 52(a); Holmes v. Bevilacqua, 794 F.2d 142, 147 (4th Cir.1986) (en banc). In determining whether the case at bar was properly dismissed, this court will inquire as to whether Carter proved a prima facie ease of discrimination and, if so, whether the prima facie case was successfully rebutted. See Holmes, 794 F.2d at 147-48 (upholding dis *458 missal on alternative grounds that plaintiff did not prove prima facie case and that record as a whole supported finding of no discrimination). a. Discriminatory Failure to Promote To prove a prima facie case of discriminatory refusal to promote under McDonnell Douglas, 411 U.S. at 802, 93 S.Ct. at 1824, plaintiff must prove that “(1) plaintiff is a member of a protected group; (2) plaintiff applied for the position in question; (3) plaintiff was qualified for the position; and (4) plaintiff was rejected for the position under circumstances giving rise to an inference of unlawful discrimination.” McNaim v. Sullivan, 929 F.2d 974, 977 (4th Cir.1991). The district judge found that Carter satisfied the first three prongs of the McDonnell Douglas test, J.A. at 366. To satisfy the fourth prong, Carter need only show that the position was filled by a white applicant, as he has done. See Patterson v. McLean Credit Union, 491 U.S. 164, 186-87, 109 S.Ct. 2363, 2378, 105 L.Ed.2d 132 (1989); Blue v. United States Dep’t of the Army, 914 F.2d 525, 537 (4th Cir.1990), cert. denied, 499 U.S. 959, 111 S.Ct. 1580, 113 L.Ed.2d 645 (1991). The Navy may rebut Carter’s prima facie case by demonstrating that the white selectee was better qualified for the position. See Patterson, 491 U.S. at 187, 109 S.Ct. at 2378. Carter, who retains the final burden of persuasion on the issue of intentional discrimination, also has the opportunity to prove that the Navy’s articulated legitimate reasons for promoting the white applicant are mere pretexts. See id. In Holmes v. Bevilacqua, this court considered a failure to promote claim by an African-American applicant for the position of deputy commissioner of the Department of Mental Health and Mental Retardation of the Commonwealth of Virginia. 794 F.2d at 143. Plaintiff had a Master’s degree and a doctorate in special education, had served as the director of a mental health facility and had acted as an assistant commissioner to the department for two years. Id. at 143-44. The selectee, a white male, did not have a doctoral degree and had served only a few months as part-time acting deputy commissioner, although he had been executive director of a local community service board. Id. Plaintiff was ranked above the selectee in the screening process. Id. at 144. The department stated that it chose the selectee because of his experience with municipal government programs to provide mental health services. Id. at 144-45. This court found no evidence showing that the failure to promote was based on race and therefore held that plaintiff did not sustain its ultimate burden of proving intentional discrimination. Id. at 147. In the case at bar, Carter did not receive a score that merited consideration as a finalist, and he was ranked well below the selectee. See Favors v. Fisher, 13 F.3d 1235, 1237 (8th Cir.1994) (finding promotion of white male based on higher test score to be legitimate reason for not selecting plaintiff). Carter submitted less complete information than the selectee because he failed to complete an SF-171 form. The EEO counselor determined that the DISCAS advertisement for the position disadvantaged Carter by encouraging a brief narrative submission and not mentioning the availability of an SF-171 form. There is no indication, however, that the advertisement intended to screen out black applicants or to target Carter personally for his filing of an EEO complaint seven years earlier. Therefore, the district court did not err in finding that Carter failed to present evidence that discrimination played a role in the Navy’s failure to promote him. b. Constructive Discharge Based on Race and Retaliation Carter’s second claim on appeal is that the Navy’s decision to demote him, which led to his constructive discharge, was motivated by racial harassment and retaliation for filing the EEO complaints. Carter has the initial burden of establishing a prim a facie case of racial discrimination under McDonnell Douglas. See St. Mary’s Honor Ctr. v. Hicks, — U.S.-,-, 113 S.Ct. 2742, 2746-47, 125 L.Ed.2d 407 (1993). i. Racially Discriminatory Discharge A prima facie case of racially motivated discharge consists of the following four *459 elements: (1) plaintiff is a member of a protected class; (2) plaintiff was qualified for the job and performed the job satisfactorily; (3) in spite of plaintiffs qualifications and performance, plaintiff was demoted and ultimately discharged; and (4) the position remained open to similarly qualified applicants after plaintiffs dismissal. See id. at -, 113 S.Ct. at 2747; Mitchell v. Data Gen. Corp., 12 F.3d 1310, 1315 (4th Cir.1993); Williams, 871 F.2d at 455. The inference of discrimination arising from plaintiffs prima facie ease may be rebutted by evidence of the employer’s legitimate, non-discriminatory reasons for the dismissal. Williams, 871 F.2d at 455-56. The burden then shifts to the plaintiff to prove that the employer’s stated reasons are pretextual. Id. at 456. As an African-American, Carter is clearly a member of a protected class and thus satisfies the first prong of the test. In determining whether Carter has met the second prong, we must consider the satisfactory or unsatisfactory nature of his job performance. The Navy articulated two reasons for its dissatisfaction with Carter: (1) failure to complete a project despite three extensions of final deadlines over two years and (2) inability to perform tasks as computer network administrator without substantial help from others. J.A. at 402, 405. Lt. Campbell was the primary source of information regarding Carter’s abilities as network administrator, as it was he who informed his superior that other co-workers ultimately performed the tasks that Carter could not. Id. at 405-06. At trial, Joan Young confirmed that Carter had fixed certain network problems only by relying on her and her expertise. Id. at 341-42. Although the evidence arguably demonstrated Lt. Campbell’s personal or racial animus towards Carter, there is no indication that Young was in any way biased against Carter. Moreover, Carter’s failure to complete a major photo analysis project within the established deadlines is unrefuted. Carter has thus not proven that his job performance was satisfactory, as mandated by the second prong. The third prong requires that there be an actual or constructive discharge. See Stetson v. NYNEX Serv. Co., 995 F.2d 355, 360 (2d Cir.1993). A constructive discharge occurs when an employer creates intolerable working conditions in a deliberate effort to force the employee to resign. Johnson v. Shalala, 991 F.2d 126, 131 (4th Cir. 1993), cert. denied, — U.S. —, 115 S.Ct. 52, — L.Ed.2d-(1993). Deliberateness can be proven by actual or circumstantial evidence, including evidence of “actions that single out a plaintiff for differential treatment.” Id. The doctrine of constructive discharge protects an employee “from a calculated effort to pressure him into resignation through the imposition of unreasonably harsh conditions, in excess of those faced by his coworkers.” Bristow v. Daily Press, Inc., 770 F.2d 1251, 1255 (4th Cir.1985), cert. denied, 475 U.S. 1082, 106 S.Ct. 1461, 89 L.Ed.2d 718 (1986). The employee “is not, however, guaranteed a working environment free of stress.” Id. Dissatisfaction with work assignments, a feeling of being unfairly criticized, or difficult or unpleasant working conditions are not so intolerable as to compel a reasonable person to resign. See Stetson, 995 F.2d at 360-61. Demotion can constitute a constructive discharge, especially where the demotion is essentially a career-ending action or a harbinger of dismissal. See Jurgens v. EEOC, 903 F.2d 386, 391-92 (5th Cir.1990). “[A] slight decrease in pay coupled with some loss of supervisory responsibilities” is insufficient evidence of constructive discharge, however. Id. at 392. The Fifth Circuit in Jurgens thus found no constructive discharge when plaintiff was demoted from a GS-15 position in which he supervised seven to twelve attorneys to a nonmanagement GS-14 position where he was in charge of only three to five attorneys, even assuming discriminatory intent. Id. at 388, 392. Carter’s evidence that his co-worker Roberts was directed to take assignments from Joan Young rather than from him provides weaker support for an alleged loss of supervisory responsibilities than that offered in Jurgens. Carter complains that the more complete half of his project was assigned to *460 another co-worker, that Lt. Campbell criticized him unfairly, and that he resented Lt. Campbell’s display of a poster containing a picture of a gorilla. These circumstances are weak evidence, at best, of a constructive discharge, but against a background of a demotion based justifiably on unsatisfactory job performance they do not prove constructive discharge. ii. Retaliatory Discharge The test for proving prima facie retaliatory discharge requires that (1) plaintiff engaged in protected activity, such as filing an EEO complaint; (2) the employer took adverse employment action against plaintiff; and (3) a causal connection existed between the protected activity and the adverse action. Williams, 871 F.2d at 457. The employer may rebut plaintiff’s prima facie case by articulating non-retaliatory reasons for its actions. Id. The burden then shifts back to plaintiff to prove the pretextual nature of those reasons. Id. This court found a causal connection between a plaintiffs protected activity and her discharge where the employer, with knowledge of a pending discrimination complaint, fired plaintiff approximately four months after the complaint was filed. Id. at 454, 457. Other courts have concluded that the discharge of an employee soon after the employee engages in protected activity is strongly suggestive of retaliatory motive and thus indirect proof of causation. See, e.g., Oliver v. Digital Equip. Corp., 846 F.2d 103, 110 (1st Cir.1988). Types of indirect proof to be considered in finding a causal nexus may include the temporal proximity of factual hearings regarding discrimination complaints as well as the actual date of filing. See Muehlhausen v. Bath Iron Works, 811 F.Supp. 15, 20 (D.Me.1993) (finding causation where suspension and discharge occurred within six months after complaints were filed and hearings were held). In this case, Carter filed several EEO complaints, including one on September 26, 1988 and another on February 9, 1990, thus engaging in a protected activity. Young, the de facto second-in-command at the branch, knew of at least one complaint because Carter asked for her assistance in typing it. J.A. at 321. On June 14, 1990, a de novo hearing on Carter’s first EEO complaint was held. The decision to downgrade him had an effective date of July 29, 1990. These facts are sufficient to establish a pri-ma facie case of retaliation under Williams. Nevertheless, “mere knowledge on the part of an employer that an employee ... has filed a discrimination charge is not sufficient evidence of retaliation to counter substantial evidence of legitimate reasons” for adverse personnel action against that employee. Williams, 871 F.2d at 457. As indicated, the Navy had substantial reasons for its decision to demote Carter, including his failure to meet deadlines and his unsatisfactory work as network administrator. There is no evidence that these non-diseriminatory reasons were pretextual. Therefore, Carter has not met his ultimate burden of proving retaliatory discharge. c. Racial Harassment or Racially Hostile Work Environment Carter also claims that in early 1989, when the former supervisor of the Recognition Branch was replaced by Lt. Campbell, the work environment degenerated due to the racial animus of Lt. Campbell. The key evidence presented by Carter on this issue includes allegations of disparate disciplinary practices and the display of a poster with a picture of a gorilla and the motto, “I wouldn’t mind being a NOBODY if I could only get A LITTLE RECOGNITION once in awhile.” The written message, with its pun on the name of the Recognition Branch, dominates as much of the poster’s visual layout as the photograph of the gorilla. Prima facie proof of disparate disciplinary practices consists of the following three elements: (1) plaintiff is a member of a protected class; (2) the prohibited conduct in which plaintiff engaged was as serious as the misconduct of employees outside the protected class; and (3) the employer imposed harsher disciplinary measures against plaintiff than against employees outside the protected class. Cook v. CSX Transp. Corp., 988 F.2d 507, 511 (4th Cir.1993). The plain *461 tiff in Cook, an African-American, satisfied the second element of the test by submitting evidence concerning white employees who had been disciplined under the same company operating rule. Id. Plaintiff did not show that the employer disciplined him more severely, however, as the evidence demonstrated that similarly situated employees had received variable treatment, either equally harsh, more severe, or more lenient. Id. at 511-12. Thus, looking at the record as a whole, this court determined that plaintiff had not proven a prima facie case. Id. at 512. In the ease at bar, Carter’s co-worker, Roberts, testified that Lt. Campbell reprimanded Carter publicly but criticized other white employees, including herself, in the privacy of his office. J.A. at 257-59. This evidence satisfies the third element of the test, that plaintiff received harsher treatment. Roberts did not identify the conduct for which Carter or the other employees were reprimanded, however. Thus, there is no evidence in the record to fulfill the second element of the test, that is, a similarly situated employee being disciplined for similar conduct. Thus, Carter has not presented pnma facie proof of disparate disciplinary measures. Carter argues nonetheless that Lt. Campbell’s public criticism of him, combined with the allegedly derogatory poster, gave rise to a hostile work environment. Lt. Campbell displayed the poster on the wall of his office next to his desk. Id. at 220, 269. Because of the arrangement of office space, the branch employees had to walk between Lt. Campbell’s desk and the wall to reach their own offices. Id. at 269. When questioned at trial as to how long the poster remained on the wall, Roberts responded, “I know it was there for a few days of me walking in and out.” Id. at 273. 11 Apparently, Lt. Campbell used the saying as a catch-phrase, often commenting, “If only I could get a little recognition .... ” Id. at 270. Carter interpreted the language on the poster as personally offensive because his hopes for becoming a responsible person in the Recognition Branch were frustrated by Lt. Campbell’s arrival. Id. at 174. He and a white co-worker found the picture of the gorilla to be offensive to African-Americans in general, as they had heard the term “gorilla” used in their past experiences outside the office as a derogatory reference to African-Americans. Id. at 176, 273. Neither Carter nor his co-worker complained about the picture to Lt. Campbell. Id. at 222, 282-83. Carter also failed to bring the poster to the attention of either Lt. Campbell’s direct supervisor or the head of NTIC. Id. at 222. A racially hostile working environment must be sufficiently severe or pervasive to alter the conditions of employment and create an abusive atmosphere. White, 939 F.2d at 160 (quoting Meritor Sav. Bank v. Vinson, 477 U.S. 57, 67, 106 S.Ct. 2399, 2405, 91 L.Ed.2d 49 (1986)). The existence of a hostile environment cannot be predicated upon acts that are isolated or genuinely trivial. Id. An alleged victim of this type of racial harassment must also demonstrate that the employer had actual or constructive knowledge of the hostile working environment. Id. We conclude that the district court did not err in dismissing Carter’s claim of racial harassment. The display of the poster and the joking comments of the Recognition Branch supervisor about “If only I could get a little recognition,” in the factual context discussed here, do not constitute acts of racial harassment severe and pervasive enough to alter the working environment. It is also not clear that Lt. Campbell or any other person in authority had actual or constructive knowledge that the poster was potentially offensive to African-American employees. Furthermore, Roberts’ testimony that Lt. Campbell generally reprimanded Carter publicly but spoke with Carter’s white co-workers in private is not substantiated by accounts of specific dates, times or circumstances. Such general allegations do not suffice to establish an actionable claim of *462 harassment. See Lenoir v. Roll Coater, Inc., 841 F.Supp. 1457, 1462 (N.D.Ind.1992) (finding plaintiffs vague allegations of being reprimanded more severely than white co-workers, without reference to exact dates, to be insufficient to support a harassment claim), aff'd, 13 F.3d 1130 (7th Cir.1994). Looking at the record as a whole, we conclude that the district court did not err in dismissing Carter’s claims of harassment or racially hostile work environment. 12 CONCLUSION In sum, the district court properly excluded Carter’s proffered statistical evidence as to the absence of African-American employees in the Analysis Directorate of NTIC. The district court likewise acted correctly in dismissing Carter’s claims of discrimination based on race and retaliation. Although Carter established a ‘prima facie case of discriminatory failure to promote, the Navy presented adequate rebuttal evidence to show that the selectee was better qualified than Carter. Carter did not succeed in proving that the Navy’s request for a “brief narrative,” like that submitted by Carter, rather than a more detailed standard form, like that submitted by the selectee, was a pretext for discrimination. Carter also did not prove that he was constructively discharged on the basis of either his race or his filing of previous EEO complaints. In addition, although Carter established a prima facie ease of retaliation by demonstrating the temporal proximity between action on the EEO complaints and his demotion, the Navy articulated legitimate, non-discriminatory reasons for taking adverse personnel action against Carter. Finally, the district court did not err in dismissing Carter’s claims of racial harassment or racially hostile work environment. No prima facie proof of disparate disciplinary practices is evident from the record. The display of a poster that may or may not be generally offensive to African-Americans, and about which no complaint was made, under all of the circumstances described here, is not an act severe enough to alter the conditions of employment. For the above reasons, the district court’s judgment is affirmed. AFFIRMED. Footnotes [1] Although Carter alleged that a member of the rating panel discussed the application process with the successful candidate, Carter did not introduce evidence in support of this allegation before the district court. Appellant’s Br. at 6 n. 2. [2] Carter also complained that he was given a version of Pagemaker software to use that was older than any other employee's and that he was not instructed on how to turn on and turn off the network computers, although white coworkers were so instructed. He further alleged that another analyst was assigned to work on the easier portion of a project formerly under Carter’s exclusive control. [3] After a de novo hearing, an administrative law judge concluded on June 14, 1990 that Carter had not suffered from discrimination in being denied the promotion. [4] This administrative complaint was dismissed upon the filing of the judicial complaint. [5] Carter did not pursue his age discrimination claim at trial. [6] Counsel for the Navy began by conceding that Carter had proved a prima facie case of racial discrimination in promotion, but withdrew that concession when it became apparent that the court disagreed. Id. at 345-47. [7] Disparate treatment occurs when an employer treats certain people less favorably than others on the basis of a protected classification such as race. International Brotherhood of Teamsters v. United States, 431 U.S. 324, 335 n. 15, 97 S.Ct. 1843, 1854 n. 15, 52 L.Ed.2d 396 (1977). [8] See Appellant’s Br. at 16 n. 4 ("Applicant flow data, SMSA [Standard Metropolitan Statistical Area] data and other potential information to support the one item before the court was never discussed.”). The only other item of statistical evidence proffered to the district court is Plaintiff’s Exhibit 51, described as “a chart from a newspaper.” J.A. at 180. This exhibit is not reprinted in the record on appeal. [9] Carter contends further that the statistical evidence is relevant to a claim of disparate impact. Carter did not allege disparate impact at the administrative level or in his judicial complaint and should not be allowed to argue it on appeal. Furthermore, the test for a disparate impact claim requires “(1) that there is an underrepre-sentation of the qualified members in a protected class promoted to the positions at issue and (2) that specific elements of the employer's promotion criteria had a significant disparate impact on the protected class.” McNaim v. Sullivan, 929 F.2d 974, 979 (4th Cir.1991) (footnote omitted). Without evidence of how many African-Americans were qualified for positions at the level to which Carter sought to be promoted, there can be no reliable proof of underrepresen-tation. See id. Plaintiff's Exhibit 50 is irrelevant to a disparate impact claim for the same reasons that it is irrelevant to a disparate treatment claim. [10] Rule 52(c), effective December 1, 1991, replaces part of Fed.R.Civ.P. 41(b), which formerly authorized dismissal at the close of plaintiff’s case-in-chief in a non-jury trial if plaintiff failed to carry an essential burden of proof. Fed. R.CÍV.P. 52(c) advisory committee’s note. [11] Young did not see the poster in Lt. Campbell’s office until after Carter had retired, when it was pointed out to her. J.A. at 318-19. [12] Because the district court properly dismissed all of Carter's claims, there is no need to consider Carter's argument that the case should be assigned to a different judge on remand. We also decline parties’ suggestion to stay this appeal pending the Supreme Court's decision on the retroactivity of certain provisions of the Civil Rights Act of 1991 (“the Act"). The Supreme Court recently held that those provisions of the Act are not retroactive. Rivers v. Roadway Express, Inc., -U.S. -, -, 114 S.Ct. 1510, 1513, 128 L.Ed.2d 274 (1994) (§ 101 of the Act); Landgraf v. USI Film Prods.,-U.S.-,-, 114 S.Ct. 1483, 1488, 128 L.Ed.2d 229 (1994) (§ 102 of the Act). Therefore, the district court properly denied Carter's motion to amend his complaint to add a claim under the Act.