McAVOY, Chief District Judge:

Plaintiff Joyce Bickerstaff appeals from a final judgment of the United States District Court for the Southern District of New York, Charles L. Brieant, Judge, dismissing her complaint alleging that defendant Vassar College (“Vassar”) denied her request for promotion to full professor because of her race and sex, in violation of Title VII of the Civil Rights Act of 1964 (“Title VII”), 42 U.S.C. § 2000e et seq. (1994). The district court granted summary judgment dismissing the complaint on the ground that Vassar had presented a sufficiently supported nondiscriminatory reason for denying Bickerstaff promotion and Bickerstaff had not produced evidence that the reason advanced was pretextual. On appeal, Bickerstaff contends that summary judgment was improper because the district court overlooked and misconstrued a vast array of evidence establishing genuine issues of material fact as to whether Vassar’s decision to deny her promotion to full professor was race and sex-based. Finding no basis for reversal, we affirm.

*441 I. BACKGROUND

The facts of this case, taken in the light most favorable to Bickerstaff as the party against whom summary judgment was granted, are as follows.

Vassar is a private educational institution, chartered in 1863, and located in Poughkeepsie, New York. Dr. Bickerstaff, an African-American female, was hired by Vassar as a lecturer with a joint appointment in the Africana Studies Program and the Education Department in 1971. 1 Her joint appointment was originally allocated two-thirds to the Department of Education and one-third to the Africana Studies Program. That allocation was later reversed. Upon earning a Ph.D. in 1975, Bickerstaff received the rank of Assistant Professor at Vassar. In 1978, Vassar promoted Bicker-staff to the rank of Associate Professor and granted her tenure.

In 1989, Bickerstaff sought promotion to full professor, which Vassar denied. Bick-erstaff appealed to Vassar’s Appeal Committee (“VAC”), which rejected her challenge. No litigation ensued. Between 1989 and 1994, Bickerstaff published no scholarly articles. Bickerstaff spent the entire 1990-91 and 1991-92 academic years on leave as a visiting professor at Berea College.

In 1994, Bickerstaff again sought promotion to full professor, which Vassar denied. The present litigation ensued concerning the denial in 1994 only. We thus review Vassar’s procedures for promotion and the events surrounding Bickerstaffs application in 1994 for full professor.

A. Vassar’s Criteria and Procedures for Promotion to Full Professor

To achieve promotion to full professor, the Vassar Faculty Handbook requires a candidate to meet the following posted criteria:

Continued demonstration of sound scholarship or significant artistic activity and teaching of a high quality will be required. It is necessary that marked distinction will have been reached in scholarship or teaching, preferably in both.

An additional important consideration will be academic leadership, which may be evidenced by participation in professional activities outside the College, service on committees within the College, or contributions to educational innovation or policy making at both the departmental and college levels.

Vassar’s procedures for promotion are established in its bylaws and are set forth in the Faculty Handbook. For a candidate such as Bickerstaff with a joint appointment, the review is a diffusive process that involves several steps and multiple recom-menders. First, “two members of rank higher than that of the member under consideration each from the home department (e.g., the chair and one other) and from the multidisciplinary program ... meet to evaluate the professional qualifications of the candidate.” Second, members of the program and the department confer and “make a written report of their deliberations,” which is transmitted to the program, the department, the college-wide Faculty Appointments and Salary Committee (“FASC”), the Dean of the Faculty, and the President. Third, “the program and the department ... take this report into consideration in making their own separate recommendations to FASC, the Dean and the President.” Fourth, FASC and the Dean, upon consideration of the departmental and program recommendations, the teaching evaluation of the Student Advisory Committee (“SAC”), the committee reports of the department and the program, and the reports of the outside evaluators, 2 make separate recommen *442 dations to the President. Lastly, the President submits her final recommendation to the Board of Trustees.

In instances “[w]hen the department or the program has fewer than two members of rank higher then [sic ] that of the person under consideration, an ad hoc committee [is] formed in each case.” These procedures “are designed to accommodate recommendations from both [the] department and program.” An appeal is available to VAC, comprised exclusively of members of the Vassar faculty.

B. Bickerstaffs Review for Promotion to Full Professor in 1994-1995

At issue is Bickerstaffs review for promotion to full professor over the 1994-95 academic year.

On November 14, 1994, SAC, which is charged with reviewing student Course Evaluation Questionnaires (“CEQs”) 3 and issuing its analysis and recommendation on promotional applications, issued a 5-0 (with one abstention) recommendation against promotion for Biekerstaff. SAC stated that it was alarmed by Bickerstaffs recently “remarkably low” CEQs and that “[t]he only identified trend is a steady decline in evaluations.” SAC added that it “is so concerned at [Bickerstaffs] performance that it feels her present status as an associate Professor deserves reexamination by the College.”

On November 28, 1994, a group consisting of the ad hoc Education Committee (“Education Committee”) and the ad hoc Africana Studies Committee (“AS Committee”) met to discuss Bickerstaffs qualifications for promotion. Vassar asserts that, in accordance with its bylaws, it convened the two ad hoc committees to review Biekerstaff for promotion because the Africana Studies Program and the Education Department each had fewer than two members of rank higher than hers. The meeting between the two committees was memorialized in a written report dated November 30, 1994, which generally discussed her candidacy, without reaching any firm conclusions, under the criteria of scholarship, teaching, and service.

On December 2, 1994, the three-member AS Committee unanimously recommended Biekerstaff for promotion to full professor. As past committees had done when reviewing other faculty for promotion, the AS Committee examined Bickerstaffs scholarship and teaching under a “broader” lens that it deemed more reflective of her real contributions, which included “nontraditional scholarship” and her role outside the classroom as an “educator.” The AS Committee stated, in pertinent part, that:

[Biekerstaff] more than any other faculty member in the Africana Studies Program provided the constant vision and guidance, regardless of whether she was the program director. The toll on her time and energy in establishing a new area of study at [Vassar] is another reason for viewing her scholarly contributions in a different light.

In terms of Bickerstaffs classroom teaching, however, the AS Committee did find “problems” as reflected in her CEQs. Thus, it urged her to “refocus and reinvigorate her efforts in the classroom.” In the end, the AS Committee concluded that Biekerstaff had made “distinctive contributions to scholarship” and that she had “a constant and consistent educational vision through all of her work as educator, scholar, and consultant.”

On December 5, 1994, the three-member Education Committee issued its report. In assessing her scholarship, it too viewed Bickerstaffs work in a “broader” context, and remarked positively on a number of *443 Bickerstaffs specific achievements. But as to teaching, the Education Committee stated that Bickerstaffs CEQs placed her “considerably lower than what would be expected of a faculty member at Vassar in order to receive a marked distinction in teaching.” In unanimously recommending against promotion, the Education Committee concluded that Bickerstaffs scholarly activities, while “creditable,” did not exhibit “marked distinction” and that her teaching fell short of “marked distinction by a rather wide margin.”

The three outside evaluators asked to assess Bickerstaffs scholarship and service (they were not asked to assess her teaching) unanimously recommended her for promotion to full professor. The outside evaluators commended Bickerstaffs commitment to the Africana Studies Program at Vassar and, more generally, recognized her as a national leader in her field. Representative comments included that Bickerstaff has “an extraordinary record of curriculum development, program innovation, and administration leadership’ ” that she “has excelled in ... many areas while taking her turn as Chair of the Education Department, as Director of the Program in Africana Studies, developing a number of innovative courses at Vassar, and serving on her fair share of committees’ ” and that “her service record is outstanding.” Notwithstanding, one outside evaluator acknowledged that she did not know Vassar’s criteria for promotion. She then remarked:

[T]he University of Maryland would find Dr. Bickerstaffs publication record to be limited. I am unclear as to how limited publications would be viewed at Vassar.

On December 7, 1994, the three-member FASC issued its report unanimously recommending against Bickerstaff for promotion, stating that the members found that “she did not meet the stated criteria for promotion.” With regard to Bicker-staffs scholarship, FASC noted that there was just one piece of traditional scholarship in her dossier. “Overall, [it] found Ms. Bickerstaffs scholarship to be exceptionally thin.” With respect to her teaching, FASC noted that Bickerstaffs CEQs were declining and that, of recent, were “dismally low, among the very worst” that the FASC members had seen. It concluded that her teaching did not exhibit “high quality, much less ... marked distinction.”

On December 22, 1994, the Acting Dean, Jesse Kalin, wrote a letter to Bickerstaff informing her that “inadvertently” some of her CEQs had not been provided to the AS Committee. The letter then told her that the AS Committee had been provided the missing CEQs and that it would be issuing an addendum report.

On January 10, 1995, the AS Committee issued its Addendum Report, which acknowledged the “mixed record” of Bicker-staffs CEQs as a whole. While noting that Bickerstaff received fours and fives 4 in sixty percent (i.e., 25/42) of the courses she had taught, it also remarked that:

[S]ince 1988 and her last review of promotion, we see a pattern of decline in course ratings, which became more precipitous in the last two years, 1992-1994. For example, 8 out of 16 courses that were ranked with a majority in category 3 occurred from 1988 on. Six out of those eight courses were taught from 1992 to 1994.

On January 23, 1995, a joint meeting was held in which FASC, the Dean and the President met with the AS Committee. The meeting took the form of questions from FASC, the Dean and the President, to the AS Committee. The first question posed was how the AS Committee appraised Bickerstaffs complete set of CEQs, to which the AS Committee responded that it viewed Bickerstaffs CEQs as “uneven.” Specifically, the AS Committee stated that her CEQs were “strongest in the mid-to-late 1980s and showed a *444 marked decline in the last two years.” It concluded that “overall, [her CEQs were] neither strong nor poor.”

Dean Kalin recommended against promotion for Bickerstaff because she did not meet the posted criteria of marked distinction in teaching. In his affidavit, he states that “[h]er teaching evaluations are among the poorest [he has] seen in a promotional review for full professor.”

On February 8, 1995, the President of Vassar wrote to Bickerstaff and informed her that she would not be recommending her for promotion to the rank of full professor. While acknowledging Bickerstaffs contribution to the Africana Studies Program at Vassar, the President expressed concerns relating to her teaching and scholarship. The President wrote that her classroom teaching was “uneven” and that “the most recent reports indicate a significant decline in classroom effectiveness.” The President also stated that while she agreed with the outside evaluators that Bickerstaffs service to the educational community and to Vassar has been significant, Vassar “requires the achievement of ‘marked distinction’ in at least teaching or scholarship, a standard which has not yet been met.” The President concluded that “[w]hile service is certainly important, it cannot alone compensate for limitations in both these areas.”

Bickerstaff filed an appeal of the denial of promotion to VAC, claiming that Vassar’s appointment of two ad hoc committees violated its written procedures. VAC denied the appeal, stating that in the event of a joint appointment, its written procedures provide for a recommendation from both the Africana Studies Program and the Education Department.

In the academic year 1994-95, Vassar considered eight applications for promotion to full professor. Of the six successful candidates, five were women.

At present, Bickerstaff continues to teach at Vassar as a tenured associate professor.

C. The Present Action

Following the denial of her request in 1994 for promotion to full professor, Bick-erstaff filed charges of race and sex-based discrimination with the Equal Employment Opportunity Commission. After receiving a right-to-sue letter, on November 29, 1996, Bickerstaff timely filed a complaint against Vassar in the United States District Court for the Southern District of New York. The complaint claimed that she was denied promotion to full professor in 1994 because of her race and sex, in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., and 42 U.S.C. § 1981. It also alleged that Vassar denied her promotion in 1994 in retaliation for her internal appeal of the denial of her promotion to full professor; and that Vassar paid her less than male professors of comparable rank, in violation of the Equal Pay Act, 29 U.S.C. § 623(a). The complaint sought monetary damages and an order granting her the rank of full professor at Vassar.

After discovery, Vassar moved for summary judgment on all of Bickerstaffs claims. On January 26, 1998, the district court granted Vassar’s motion for summary judgment and dismissed the complaint in its entirety. See Bickerstaff v. Vassar College, 992 F.Supp. 372 (S.D.N.Y.1998). In granting summary judgment, the district court assumed arguendo that Bickerstaff had made a prima facie case of race and sex-based discrimination. It then found Vassar had met its burden of production by articulating a sufficiently supported nondiscriminatory reason for denying Bickerstaff promotion to full professor — namely, that she did not satisfy the posted criteria for promotion. Ultimately, the district court concluded that Bickerstaff had failed to present sufficient evidence from which a reasonable juror could infer that Vassar’s stated reason for not promoting her was pretextual. In reaching this conclusion, the district court noted that most of Vassar’s posted criteria *445 for promotion, like teaching of “marked distinction,” are amorphous and subjective—“just so much bafflegab.” Id. at 375. It found that the one exception to this “extreme subjectivity” was that of the criterion of “teaching of a high quality.” Id. Citing Bickerstaffs CEQs as an “objective” gauge of teaching ability, the district court held that no reasonable jury could find Vassar’s stated reason for denying her promotion was pretextual. Id. at 377. Accordingly, the district court dismissed the Title VII claim for sex and race-based discrimination. For the same reason—as well as because it found that a failure to promote was not a refusal to contract—it dismissed the section 1981 claim. It also dismissed the retaliation claim for insufficiency of evidence, and the Equal Pay Act claim—to the extent it was not time barred—for insufficiency of evidence.

Bickerstaff moved for reconsideration, asserting, inter alia, that the district court had failed to draw the proper inferences from the evidence presented. In denying the motion, the district court stated that it had considered all the evidence but found that it did not raise a jury question whether Vassar’s reason to deny Bickerstaff promotion was pretextual.

This appeal followed.

II. DISCUSSION

On appeal, Bickerstaff asserts that the district court erred in granting Vassar’s motion for summary judgment, asserting that it both overlooked and misconstrued a vast array of evidence establishing genuine issues of material fact as to whether race and sex were motivating factors in Vassar’s decision to deny her promotion to full professor.

We review a grant of summary judgment de novo. See Lowrance v. Achtyl, 20 F.3d 529, 534 (2d Cir.1994). Under Federal Rule of Civil Procedure 56(c), summary judgment may be granted only when there is no genuine issue as to any material fact and the movant is entitled to judgment as a matter of law. See Celotex Corp. v. Catrett, 477 U.S. 317, 322, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986). “While genuineness runs to whether disputed factual issues can ‘reasonably be resolved in favor of either party,’ ... materiality runs to whether the dispute matters, i.e., whether it concerns facts that can affect the outcome under the applicable substantive law.... A reasonably disputed, legally essential issue is both genuine and material and must be resolved at trial.” Graham v. Henderson, 89 F.3d 75, 79 (2d Cir.1996) (quoting Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248-50, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986)).

Title VII forbids an employer from intentionally discriminating against an employee because of that employee’s “race, color, religion, sex, or national origin.” 42 U.S.C. § 2000e-2(a) (West 1997). A violation of Title VII can be shown by either direct, statistical or circumstantial evidence. See, e.g., Luciano v. Olsten Corp., 110 F.3d 210, 215-16 (2d Cir.1997).

Title VII suits fall into two basic categories: “single issue motivation” and “dual issue motivation” cases. Renz v. Grey Adver., Inc., 135 F.3d 217, 221 (2d Cir.1997); Fields v. New York State Office of Mental Retardation and Developmental Disabilities, 115 F.3d 116, 119-20, 124 and n. 4 (2d Cir.1997); see also Raskin v. Wyatt Co., 125 F.3d 55, 60 (2d Cir.1997) (using the less preferable labels “pretext” and “mixed motive”). In single issue motivation cases, “the single issue [is] whether an impermissible reason motivated the adverse action,” Fields, 115 F.3d at 119-20, which courts analyze under the framework first set forth in McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802-03, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). In dual issue motivation cases, the determination involves “both the issue of whether the plaintiff has proved that an impermissible reason motivated the adverse action and the additional issue of whether the defendant has proved that it would have taken the same action for a permissible rea *446 son[,]” Fields, 115 F.3d at 120, which is analyzed under the framework set forth in Price Waterhouse v. Hopkins, 490 U.S. 228, 258, 109 S.Ct. 1775, 104 L.Ed.2d 268 (1989) (plurality opinion).

In the instant matter, while Bickerstaff principally treats this as a single issue motivation case pursuant to McDonnell Douglas, 411 U.S. at 802-03, 93 S.Ct. 1817, she does make passing argument for dual issue motivation treatment under Price Waterhouse, 490 U.S. at 258, 109 S.Ct. 1775. If there had been “policy documents [or] evidence of statements or actions by' decisionmakers [at Vassar] that may be viewed as directly reflecting the alleged discriminatory attitude,” i.e., a “smoking gun,” we would agree that this is a dual issue motivation case. See Raskin, 125 F.3d at 60 (emphasis in original) (internal quotations and citation omitted). But there is not, as will be shown.

As a single issue motivation case, we apply the three-step, burden-shifting paradigm set forth in McDonnell Douglas, 411 U.S. at 802-03, 93 S.Ct. 1817, and its progeny. See St. Mary’s Honor Ctr. v. Hicks, 509 U.S. 502, 506-510, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993); Texas Dep’t of Community Affairs v. Burdine, 450 U.S. 248, 254, 101 S.Ct. 1089, 67 L.Ed.2d 207 (1981). Under this framework, a plaintiff bears the initial burden of proving, by a preponderance of the evidence, a prima facie case of discrimination. St. Mary’s, 509 U.S. at 506, 113 S.Ct. 2742. Plaintiffs burden of establishing a prima facie ease is “minimal.”. Id. “Establishment of the prima facie case in effect creates a presumption that the employer unlawfully discriminated against the employee.” Burdine, 450 U.S. at 254, 101 S.Ct. 1089. “To establish a ‘presumption’ is to say that a finding of the predicate fact (here, the prima facie case) produces a required conclusion in the absence of explanation.” St. Mary’s, 509 U.S. at 506, 113 S.Ct. 2742 (internal quotations and citation omitted).

Once a plaintiff has established a prima facie case, the burden of production shifts to the employer to rebut this presumption by articulating a legitimate, nondiscriminatory reason for its actions. Fisher v. Vassar College, 114 F.3d 1332, 1335-36 (2d Cir.1997) (en banc) (“Fisher III”), cert. denied, — U.S. -, 118 S.Ct. 851, 139 L.Ed.2d 752 (1998). The defendant’s burden of production also is not a demanding one; she need only offer such an explanation for the employment decision. Fisher III, 114 F.3d at 1336. Although the burden of production shifts to the defendant, the ultimate burden of persuasion remains always with the plaintiff. St. Mary’s, 509 U.S. at 507, 511, 113 S.Ct. 2742; see also Norton v. Sam’s Club, 145 F.3d 114, 118 (2d Cir.) (“The thick accretion of cases interpreting this burden shifting framework should not obscure the simple principle that lies at the core of anti-discrimination cases. In these, as in most other cases, the plaintiff has the ultimate burden of persuasion.”), cert. denied, — U.S. -, 119 S.Ct. 511, 142 L.Ed.2d 424 (1998).

If the defendant proffers a legitimate, nondiscriminatory reason for a challenged employment action, “the presumption raised by the prima facie case is rebutted, and drops from the case.” St. Mary’s, 509 U.S. at 507, 113 S.Ct. 2742 (internal quotations and citations omitted). “The plaintiff then has the opportunity to demonstrate ‘that the proffered reason 'was not the true reason for the employment decision,’ and that race was.” Fisher III, 114 F.3d at 1336 (quoting St. Mary’s, 509 U.S. at 507-08, 113 S.Ct. 2742). The plaintiffs opportunity to demonstrate that the employer’s proffered reason was false now merges with her ultimate burden to persuade the trier of fact that she has been the victim of intentional discrimination (i.e., that an illegal discriminatory reason played a motivating role in *447 the adverse employment decision). 5 See St. Mary’s, 509 U.S. at 516-17, 113 S.Ct. 2742, Burdine, 450 U.S. at 256, 101 S.Ct. 1089; Fisher III, 114 F.3d at 1337; see also Hollander v. American Cyanamid Co., 172 F.3d 192, 200 (2d Cir.1999).

We have previously held that:

Though there are sentences in some opinions to the effect that a Title VII plaintiff must prove “both that the [defendant’s proffered] reason was false, and that discrimination was the real reason,” St. Mary’s, 509 U.S. at 515, 113 S.Ct. at 2752, these decisions do not require a finding of [falsity] in addition to a finding of discrimination; they make the quite different point that a Title VII plaintiff may not prevail by establishing only [falsity], but must prove, in addition, that a motivating reason was discrimination. See id. at 511, 113 S.Ct. 2742.

Fields, 115 F.3d at 120 (emphasis added); see also Cronin v. Aetna Life Ins. Co., 46 F.3d 196, 203 (2d Cir.1995) (“[T]he plaintiff is not required to show that the employer’s proffered reasons were false or played no role in the employment decision, but only that they were not the only reasons and that the prohibited factor was at least one of the ‘motivating’ factors.”); Sutera v. Schering Corp., 73 F.3d 13, 17 (2d Cir.1995) (same); Renz, 135 F.3d at 222 n. 3 (“Though a discrimination plaintiff may not succeed by proving only that a proffered explanation is false but must prove that discrimination motivated the adverse action, the plaintiff is entitled to succeed by proving discriminatory motivation without proving that a proffered explanation is false.”) (citing Fields, 115 F.3d at 121). The opportunity presented to the plaintiff under the paradigm—to show the employer’s proffered reason false—is designed simply “to sharpen the inquiry into the elusive factual question of intentional discrimination.” Burdine, 450 U.S. at 255 n. 8, 101 S.Ct. 1089. The consequence of a plaintiff engaging this opportunity and demonstrating falsity is that it may (and, in most circumstances, will) advance her greater enterprise of showing discriminatory intent. See Fisher III, 114 F.3d at 1338-40. Thus properly understood, as the paradigm hones and sequentially narrows the factual inquiry, it extends to the plaintiff the opportunity to demonstrate that the proffered reason was false, as one means to support her ultimate burden of proving discrimination. See, e.g., St. Mary’s, 509 U.S. at 511, 517, 113 S.Ct. 2742 (“[R]ejection of the defendant’s proffered reasons will permit the trier of fact to infer the ultimate fact of intentional discrimination ... [but does not] compel[ ] judgment for the plaintiff’).

Applying the paradigm to the present case, we too shall assume in Bick-erstaff s favor a prima facie case of sex and race discrimination. Vassar, in turn, has satisfied its burden of production by proffering admissible evidence of a legitimate, nondiscriminatory reason for denying Bickerstaff promotion to full professor— namely, that she did not meet the posted criteria. Thus, the central question presented on this appeal is whether Bicker-staff has presented sufficient admissible evidence from which a rational finder of fact could infer that more likely than not she was the victim of intentional discrimination. See St. Mary’s, 509 U.S. at 507, 511, 113 S.Ct. 2742; Fisher III, 114 F.3d at 1337; Stern v. Trustees of Columbia Univ. in the City of New York, 131 F.3d 305, 312 (2d Cir.1997); Renz, 135 F.3d at 221-23 & n. 5.

*448 In assessing the record to determine whether there is a genuine issue to be tried as to any material fact, courts resolve all ambiguities and draw all reasonable factual inferences in favor of the party opposing the motion. See Nora Beverages, Inc. v. Perrier Group of Am., Inc., 164 F.3d 736, 742 (2d Cir.1998). Moreover, as discrimination will seldom manifest itself overtly, courts must “be alert to the fact that [e]mployers are rarely so cooperative as to include a notation in the personnel file that the firing is for a reason expressly forbidden by law.” Ramseur v. Chase Manhattan Bank, 865 F.2d 460, 464-65 (2d Cir.1989) (internal quotation marks omitted). However, they must also carefully distinguish between evidence that allows for a reasonable inference of discrimination and evidence that gives rise to mere speculation and conjecture. This undertaking is not one of guesswork or theo-rization. After all, “[a]n inference is not a suspicion or a guess. It is a reasoned, logical decision to conclude that a disputed fact exists on the basis of another fact [that is known to exist].” 1 Leonard B. Sand, et al., Modern Federal Jury Instructions ¶ 6.01, instr. 6-1 (1997). Thus, the question is whether the evidence can reasonably and logically give rise to an inference of discrimination under all of the circumstances. As a jury would be entitled to review the evidence as a whole, courts must not view the evidence in piecemeal fashion in determining whether there is a trial-worthy issue. Stern, 131 F.3d at 314; Danzer v. Norden Sys., Inc., 151 F.3d 50, 57 (2d Cir.1998).

We now turn to consider whether Bick-erstaff s evidence is sufficient to create an issue of material fact as to whether she was the victim of discrimination.

A. Statistical Evidence

Bickerstaff submitted to the district court a statistical report from Professor Mary Gray, which addressed two basic questions: whether salaries at Vassar varied due to race or sex, and whether Bick-erstaff s CEQs varied based on the racial makeup of her classes. With respect to the matter of salary variance, Gray employed a multiple regression analysis, a commonly accepted method of statistical analysis for examining the effect of independent variables on a dependent variable. See Bazemore v. Friday, 478 U.S. 385, 400, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986) (Brennan, J., joined by all other Members of the Court, concurring in part). Gray’s regressions controlled for the independent variables of experience, rank, productivity, and discipline. She concluded that “there is evidence of consistent differences in salaries based on sex and race.” She also found that Bickerstaff s salary “is consistently below that predicted for her by a statistically significant amount.”

As to the question of race bias affecting CEQs, Gray relied on information provided to her by Bickerstaff that “the Africana Studies courses are approximately % to African American students whereas the courses listed in the Education Department have few, if any, African American students.” After performing a “preliminary analysis” based on that information, Gray found that in eighty-three percent of the Africana Studies courses Bickerstaff had taught, she had CEQ scores in the two highest categories, compared to only forty-two percent of the Education Department courses. Based on this evidence, Gray’s Report concluded that “race is a component in [CEQs].”

It is apodictic that “[statistical reports may ... be admissible to support an inference of discrimination.” Raskin, 125 F.3d at 67 (citing Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 307-08, 97 S.Ct. 2736, 53 L.Ed.2d 768 (1977)) (further citations omitted). The district court found, however, that Gray’s statistical analysis did not raise such an inference because it failed to control for nondiscriminatory causes. The net import of the district court’s opinion is that this flaw in the statistical evidence rendered it of no pro *449 bative value to the issue of race or sex discrimination and, thus, was inadmissible.

As such a ruling is evidentiary in character, we review it for abuse of discretion. See Hollander, 172 F.3d at 202 (citing General Elec. Co. v. Joiner, 522 U.S. 136, 118 S.Ct. 512, 517, 139 L.Ed.2d 508 (1997)). A district court has broad discretion in choosing whether to admit evidence. Id. “Although disputes as to the validity of the underlying data go to the weight of the evidence, and are for the fact finder to resolve, see McCullock v. H.B. Fuller Co., 61 F.3d 1038, 1044 (2d Cir.1995), questions of admissibility are properly resolved by the court.” Raskin, 125 F.3d at 66 (further citations omitted). The court’s role is one of “a gatekeeper to exclude invalid and unreliable expert testimony.” Hollander, 172 F.3d at 202. Further, an expert’s opinion is no different than any other evidence in that it is subject to the relevancy requirement of Rule 401 of the Federal Rules of Evidence. Id. (citing United States v. Khan, 787 F.2d 28, 34 (2d Cir.1986)).

Bickerstaff asserts that notwithstanding any shortcomings in Gray’s statistical report, the district court erred in according it no probative weight. For this proposition, Bickerstaff relies on Bazemore, 478 U.S. at 400, 106 S.Ct. 3000, in which the Supreme Court held that the lower courts erred in finding that a multiple regression analysis “which accountfed] for the major factors” lacked evidentiary value simply because the analysis failed to account for “all measurable variables.” Bazemore counseled that “[n]ormally, failure to include variables will affect the analysis’ probativeness, not its admissibility.” Id.

The critical distinction that Bickerstaff misses, however, is that in the present case the district court did not reject Gray’s regression analysis because it included less than all the relevant variables; rather, it found the evidence not probative because it omitted the major variables. See, e.g., Koger v. Reno, 98 F.3d 631, 637 (D.C.Cir.1996). Thus, as we explain below, Gray’s statistical analysis falls within the Baze-more exception—that “[t]here may ... be some regressions so incomplete as to be inadmissible as irrelevant.” Bazemore, 478 U.S. at 400 n. 10, 106 S.Ct. 3000.

The starting point is to understand that salary increases at Vassar are based upon merit and seniority. Merit increases are determined on a scale of 0-8 points, based on the criteria of scholarship (0-3 points), teaching (0-3 points), and service (0-2 points). Gray’s regression analysis, however, does not even purport to account for two of these major variables—teaching and service. See Raskin, 125 F.3d at 67-68 (citing Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 109 S.Ct. 2115, 104 L.Ed.2d 733 (1989) (without a showing of causation between the challenged practice and the alleged disparities, “employers [would be] potentially liable for the myriad of innocent causes that may lead to statistical imbalances in the composition of their work forces”) (internal quotations and citations omitted)); see also Hollander, 172 F.3d at 204. These variables are too significant not to be accounted for in the regression analysis in this case. Further, in light of the point system that Vassar employs, it cannot be said that these variables are not quantifiable or controllable. As to the remaining major variable of scholarship, though Gray attempted to account for it, she candidly admits that this variable “is not totally reliable” because she had incomplete background information on other faculty members. Even to the extent she had complete data, Gray neither explained how she ranked the publications nor whether these were the rankings that Vassar employs. See Fisher III, 114 F.3d at 1361 n. 13 (Calabresi, J., concurring in part and dissenting in part) (“[W]hat are first class reviews is not for the district court [or a statistician] to determine. That is, Vassar has a perfect right to decide for itself what its standards are and it may premise promotion on a publication in a journal that the district *450 court thinks is awful.”)- We thus conclude that, because the regression analysis failed to account for the major factors that Vassar considers in evaluating salary increases, the district court did not abuse its discretion in according the regression analysis no probative weight.

Gray’s statistical evidence that race bias tainted Bickerstaffs CEQs is equally infirm. First, much like her regression analysis, her so-called “preliminary” statistical analysis of race taint in the CEQs makes no effort to account for nondiscriminatory explanations for the disparity, such as class size or class level. See Raskin, 125 F.3d at 67-68. Without attempting to control for such other causes, her assumption that race bias affected the CEQs is untenable. See Hollander, 172 F.3d at 202; Raskin, 125 F.3d at 67-68. Second, Gray’s statistical evidence is based upon the review of an inadequate sample of one—the plaintiff. See, e.g., Pollis v. New Sch. for Soc. Research, 132 F.3d 115, 121-22 (2d Cir.1997) (collecting cases in which small sample size was found as a matter of law to render statistical evidence insufficient to support inference of discrimination). Gray did not review race as a factor tainting CEQs for the Vassar faculty as a whole, nor did she attempt an analysis of the six African-American associate professors who were reviewed for tenure on joint appointments in Africana Studies. In short, because Gray’s statistical evidence does not make it more or less likely that Vassar discriminated against Bickerstaff because of her race or sex, the district court did not abuse its discretion in finding the statistical evidence of no probative value.

B. Shamba Donovan

Bickerstaff next contends that Shamba Donovan, a student member of SAC who was pursuing secondary social studies in the Department of Education, is racially prejudiced. She thus asserts that the unfavorable SAC Report, which formed part of the materials reviewed by Vassar’s decision makers in evaluating her teaching, was impermissibly tainted with race bias. As support, Bickerstaff cites to her own affidavit, which alleged that weeks before SAC issued its report, Donovan had confronted her and expressed “a hostile opinion about a new Africana Studies course that was brought before the all-college Committee on Curriculum Policy on which he was a student representative.” She also cites to the deposition testimony of Daniel Perlstein, an Assistant Professor in Vassar’s Education Department, who taught a course that Donovan took. Perl-stein characterized Donovan as “very confrontational,” one who “liked to criticize.” In response to questions about Donovan’s view of race, Perlstein commented that:

[Donovan] consistently objected to an attentiveness to race, to discussing it, to students arguments of a need to combat racism.... His position was that the interests and issues of social and cultural racial diversity which are common at Vassar were misplaced and wrong.... That students and faculty ... were too concerned of [sic ] matters of difference and in particular matters of racial difference.

We recognize that the impermissible bias of a single individual at any stage of the promoting process may taint the ultimate employment decision in violation of Title VII. See Lam v. University of Hawaii, 40 F.3d 1551, 1560 (9th Cir.1994). This is true even absent evidence of illegitimate bias on the part of the ultimate decision maker, so long as the individual shown to have the impermissible bias played a meaningful role in the promotion process. For two independent reasons, we find that there is insufficient evidence to raise a reasonable inference that such occurred in this case.

First, Bickerstaffs evidence does not allow for the reasonable inference that Donovan was biased against women or African-Americans. Donovan’s mere opposition to the introduction of a new Afri-cana Studies course he was charged with *451 reviewing, without more, cannot rationally be considered tantamount to an expression of race animus. Neither do Perl-stein’s comments provide any footing to make the inferential leap that Bickerstaff desires. That Donovan was critical of the emphasis at Vassar on issues of cultural and racial diversity does not reasonably lead to the conclusion that he was in the grip of race bias. Indeed, Perlstein’s testimony does not say as much.

Second, even assuming arguendo that Bickerstaffs evidence could give rise to the reasonable inference that Donovan harbored discriminatory bias, there is no evidence that Donovan played a meaningful role in the review process such that any race animus he harbored could have tainted SAC’s recommendation or the ultimate promotional decision. Donovan was only one of six student members of SAC, an entity created for the narrow purpose of compiling and analyzing the CEQ data. SAC hardly had unfettered discretion in evaluating Bickerstaff for promotion; its source of information was limited to the CEQ data. The SAC Report in this case is no different than any other. SAC arrived at its recommendation after analyzing the raw data of Bickerstaffs CEQs. Based on Bickerstaffs CEQs, SAC recommended 5-0 (with one abstention) against promotion, concluding that Bickerstaffs CEQs were remarkably low in all but two categories in recent years. In varying degrees, at least five levels of Vassar decision makers expressed the same concern over Bickerstaffs CEQs. Even the positive internal recommendation she received from the AS Committee acknowledged the precipitous decline in her CEQS.

It is noteworthy as well that Bickerstaff does not contest the validity of SAC’s compilation of the raw CEQ data it reviewed. Nor has Bickerstaff otherwise explained or presented evidence addressing how Donovan’s alleged race bias could have influenced SAC’s recommendation. She has presented no evidence that discussions relating to her race or sex took place at a SAC meeting. Cf. Lam, 40 F.3d at 1560 (finding , that discrimination could be inferred when race-biased committee member disparaged plaintiffs abilities at committee meeting). She also has presented no argument that the SAC Report is evidence of discrimination. While Bickerstaff objects to its critical tone, Title VII does not insulate an individual from criticism that is not based on an impermissible reason.

In short, given that Bickerstaffs evidence does not support the reasonable inference that Donovan harbored discriminatory bias, and the absence of evidence establishing any causal link between Donovan’s alleged discriminatory bias and Vassar’s decision to deny her promotion, we find Bickerstaffs allegation that Donovan infected the promotional review process with racism insufficient to raise a question of material fact.

C. Affidavit of Lawrence Mamiya

Bickerstaff also submitted the affidavit of Professor Lawrence Mamiya, who chaired the AS Committee that reviewed Bickerstaff for promotion. Mami-ya opined that Bickerstaff “is in every way qualified for the promotion to Full Professor of Africana Studies and the Department of Education according to promotion criteria as outlined in the Faculty Handbook.” His affidavit further stated that: “It is my opinion that the reason Joyce Bickerstaff was turned down for promotion to full professor is on account of race and the prejudice against her based on attitudes about African Americans at Vassar.”

We have previously held that “[t]o satisfy Rule 56(e), affidavits must be based upon ‘concrete particulars,’ not conclusory allegations.” Schwapp v. Town of Avon, 118 F.3d 106, 111 (2d Cir.1997) (quoting Meiri v. Dacon, 759 F.2d 989, 998 (2d Cir.1985)); BellSouth Telecomms., Inc. v. W.R. Grace & Co., 77 F.3d 603, 615 (2d Cir.1996). The problem with Mamiya’s affidavit is that it contains no “concrete particulars.” While his affidavit informs us *452 that he has known Bickerstaff for more than 20 years as a colleague in the Africa-na Studies Program at Vassar, Mamiya fails to offer a single act, statement, or admission by any Vassar decision maker or decision making entity—or by anyone else at Vassar for that matter—to support his allegation that Bickerstaff was denied promotion to full professor because of her race. Thus, the district court properly disregarded Mamiya’s allegations of race animus as conclusory.

Mamiya’s affidavit is also notable for what it does not contain. His affidavit does not assert that the CEQs were tainted by race; he does not retract his Committee’s statements concerning Bicker-staffs teaching that there was “a pattern of decline in [her CEQs], which became more precipitous in the last two years;” and he does not contradict Vassar’s assertion that CEQs are the chief evaluative instrument employed by Vassar to assess teaching.

D. Professor Robin Trainor

Bickerstaff next asserts that Professor Robin Trainor, the chair of the Education Committee that reviewed her for promotion, openly declared her racially prejudiced opposition to Bickerstaff prior to her 1994 candidacy for promotion. As evidence, she cites to the deposition testimony of Professor Perlstein, who testified, in pertinent part, that:

At more than one department meeting when Miss Bickerstaff was there, [Trai-nor] said that it would be a crime if Miss Bickerstaff were made full professor and a disgrace.... [Trainor] frequently said as well that Miss Bickerstaff contributed nothing to the department or the college.

Perlstein also testified “that matters of race were, it seemed to me, relevant to her low opinion [of Bickerstaff].”

Even assuming that Trainor made the statements Perlstein attributed to her, Perlstein’s testimony provides no hint as to any racial reason for Trainor’s opposition. Attributing the reason to race would be based entirely on speculation, not logic. See Fisher v. Vassar College, 70 F.3d 1420, 1442 (2d Cir.1995) (“Fisher I”) (noting that an inference could not be drawn that “vehement” opposition “arose from impermissible bias as opposed to, for example, strong opposition on principled grounds”). Indeed, during oral argument, Bickerstaff s counsel conceded as much but argued that Perlsteiris testimony attributing Trainor’s remarks to “matters of race” provides the missing link. We disagree. Statements that are devoid of any specifics, but replete with conclusions, are insufficient to defeat a properly supported motion for summary judgment. See, e.g., BellSouth Telecomms., Inc., 77 F.3d at 615 (“ ‘It is not sufficient merely to assert a conclusion without supplying supporting arguments or facts.’ ”) (quoting SEC v. Research Automation Corp., 585 F.2d 31, 33 (2d Cir.1978)). Perlstein’s testimony is just that—devoid of any concrete particulars and wholly conclusory. While Trai-nor’s alleged comments may be rude and derogatory, Title VII is not a “general civility code.” Oncale v. Sundowner Offshore Servs., Inc., 523 U.S. 75, 118 S.Ct. 998, 1002, 140 L.Ed.2d 201 (1998).

E. Report of the Visiting Committee and the Affidavit of Professor Rhett Jones

Bickerstaff next cites to the report of the Visiting Committee for Africa-na Studies (“Visiting Committee”) that reviewed the Africana Studies Program at Vassar in 1988. Specifically, Bickerstaff relies on the statement contained in the Report that “Africana Studies has been subjected to hostilities that make it unique among academic departments and programs [at Vassar].” The Visiting Committee also remarked that it had been informed that both the former president of Vassar and faculty advisors had been unsympathetic and unsupportive of Africana Studies. It further stated that it had been told that certain departments at Vassar, such as the Art Department and the Dra *453 ma Department, had expressed “some hostility toward Africana Studies.”

A careful reading of the entire report reveals, however, that the Visiting Committee also noted that it “is in no position to ascertain the truth of such charges.” Additionally, the Visiting Committee remarked positively in other regards, stating that “the support of the administration has been consistent, and that faculty on campus have come to favorable terms with Africana Studies.” Similarly, in another section, the Visiting Committee wrote that it is “delighted to report [about] mutual admiration and support on the part of department chairs and Africana Faculty” regarding the Africana Studies Program. In the final analysis, the Visiting Committee found both positive and negative reaction to and support for Africana Studies in 1988.

But more to the point, the connection that Bickerstaff suggests—that perceived resistance to Africana Studies in 1988 is evidence that she was denied promotion in 1994 because of her race—cannot reasonably be made. The affidavit of Professor Rhett Jones, a member of the Visiting Committee, as well as an outside evaluator of Bickerstaff for promotion in 1994, does not alter this conclusion. In her affidavit, Jones asserts that Bickerstaff should have been promoted based on her achievements in Africana Studies. Yet, Jones admits that she has never observed Bickerstaff in the classroom. Jones also does not assert that she has any personal knowledge whether the attitudes observed by the Visiting Committee in 1988 persisted in 1994, much less whether they might have influenced Vassar’s decision to deny Bickerstaff promotion in 1994. 6 While Vassar ultimately declined to follow Jones’ outside recommendation that Bickerstaff be promoted, Jones’ statement that, “I can only assume [my recommendation] was ignored for reasons of race,” is both speculative and contrary to fact. In truth, both the FASC Report and the President’s letter to Bickerstaff specifically mention the consideration given to the recommendations of the outside evaluators. In declining to follow the recommendations of the outside evaluators, the President’s letter stated:

The outside reviewers place much emphasis on your sendee to the educational community and within the College, and we agree that this has been significant. The Governance, however, requires the achievement of “marked distinction” in at least teaching or scholarship, a standard which has not yet been met. While service is certainly important, it cannot alone compensate for the limitations in both these areas.

Thus, contrary to Jones’ belief, her recommendation was not “ignored,” it was just not followed by Vassar in light of its posted criteria for promotion.

In short, neither the Visiting Committee’s Report in 1988 nor Jones’ affidavit is sufficient to create a question of material fact whether Bickerstaff was denied promotion in 1994 because of her race.

F. Alleged Incidences of Disparate Treatment

1. The Appointment of Two ad hoc Committees

Bickerstaff next claims that the appointment of two ad hoc committees is in violation of Vassar’s written procedures and is without precedent and constituted retaliation against her for having filed an appeal from her denial of tenure in 1989.

While it is true that “[departures from procedural regularity ... can raise a question as to the good faith of the process where the departure may reasonably affect the decision,” Stern, 131 F.3d at 313 (internal quotations and citation omitted), Bick-erstaff has not shown that the creation of two ad hoc committees in her case was a procedural irregularity or, more impor *454 tantly, that it was either race-related or motivated by retaliation. First, Bicker-staff misunderstands Vassar’s written procedures for promotion in connection with joint appointments. The Faculty Handbook addresses multidisciplinary programs and the process for promotions for persons holding joint appointments, and unambiguously provides for the program and department “making their own separate recommendations to FASC, the Dean and the President.” Further, the Faculty Handbook plainly states that “[w]hen the department or the program has fewer than two senior members then [sic] that of the candidate for promotion, an ad hoc committee [is] formed in each case.” The Faculty Handbook continues that FASC will consider “the departmental and program recommendations,” and that these “procedures are designed to accommodate recommendations from both [the] department and program.” Thus, in establishing two ad hoc committees and receiving separate recommendations from each, Vassar complied with its posted procedures.

Second, Bickerstaff has not presented evidence that the appointment of two ad hoc committees is unprecedented at Vassar. In her deposition, Bickerstaff admitted that there was a recommendation from the Education Department and a recommendation from the Africana Studies Program for Patricia Kauroma. Similarly, she conceded that there were separate recommendations from the Department and the Africana Studies Program for two other faculty members—Moses Nkondo and Tess Onwueme. She also admitted that the basis for her belief that the appointment of two ad hoc committees was unprecedented was limited to herself and a few others. For example, Bickerstaff points out that only one ad hoc committee was appointed to review her application for promotion in 1989. What Bickerstaff has identified, however, is an apparent procedural irregularity in her 1989 application for promotion (which application is not presently before us), rather than an irregularity in 1994. Because there were no faculty members of rank higher than hers in 1989, the director of the Africana Studies Program should have been consulted in the appointment of a committee. Thus, consistent with our previous analysis, the Faculty Handbook required the formation of two ad hoc committees in 1989. The remaining examples that Bickerstaff cites in which Vassar appointed only one ad hoc committee are inapposite; the faculty members under consideration did not hold joint appointments. Accordingly, the use of two ad hoc committees in 1994 was in conformance with the procedures outlined in the Faculty Handbook and, therefore, does not reasonably or logically form the basis of an inference of impermissible discrimination or retaliation for having filed an appeal from her denial of tenure in 1989.

2. Vassar’s Treatment of CEQs

Bickerstaff next argues that Vassar departed from procedure by placing undue emphasis on her CEQs as an instrument for evaluating her teaching ability.

In support of summary judgment, Vassar submitted the affidavit of Dean Kalin, who opined that CEQs have been the primary evaluative tool used to measure teaching for the purpose of promotional decisions. According to Dean Kalin, though there is no “bright line divide” at which an applicant’s CEQs evidence either marked distinction or teaching of high quality, successful applicants regularly receive seventy to eighty percent of their scores in the top two categories on the key questions of overall effectiveness of the course and overall effectiveness of the instructor. In opposition to summary judgment, Bickerstaff submitted her own affidavit stating that CEQs are not the primary instrument used to evaluate teaching and that Vassar’s undue emphasis on her CEQs is evidence that she was treated differently than others.

In her deposition testimony, however, Bickerstaff admitted that Vassar *455 relies almost exclusively on CEQs in evaluating teaching. It is beyond cavil that “a party may not create an issue of fact by submitting an affidavit in opposition to a summary judgment motion that ... contradicts the affiant’s previous deposition testimony.” See, e.g., Hayes v. New York City Dep’t of Corrections, 84 F.3d 614, 619 (2d Cir.1996) (citations omitted). For this reason alone, Bickerstaffs affidavit does not create a genuine issue of material fact. We also note another admission by Bickerstaff confirming the consistent use of CEQs in evaluating teaching ability. Namely, just six months before her own promotional review, Bickerstaff authored a recommendation against tenure for an African-American woman with a joint appointment, in which she herself relied considerably on CEQs in evaluating the candidate’s teaching ability.

Furthermore, Bickerstaff has not otherwise shown that her CEQs were unduly emphasized in comparison to those of other candidates. Rather, the numerous promotional review materials in the record support Vassar’s position that it relies predominantly on CEQs in evaluating teaching ability. The record thus shows that Bickerstaff was treated no differently than other candidates in using CEQs as the chief instrument to assess her qualifications to teach. While it is plain that Bick-erstaff and others genuinely take issue with Vassar’s emphasis on CEQs as the chief evaluative tool for teaching ability, such disagreement is not relevant to Bick-erstaffs discrimination claim.

Bickerstaff also contends that Vassar’s failure to have her full set of CEQs considered by FASC and the Education Committee is evidence of possible CEQ manipulation and bad faith. We do not believe this is the case for two reasons: first, the evidence does not support a finding of bad faith on the part of Vassar; and second, the initial omission of some of her CEQs from her review packet could not have affected the promotional decision. The allegation of bad faith finds scant support in light of the fact that once Vassar discovered the error, it took corrective action and requested the AS Committee to review Bickerstaffs complete set of CEQs and issue an addendum report. Vassar took additional remedial measures in holding a joint meeting between the President, the Dean, FASC and the AS Committee, in which the question was posed to the AS Committee how it interpreted Bickerstaffs complete set of CEQs.

Bickerstaff also has not shown how Vassar’s initial failure to include her full set of CEQs in her evaluation packet had any effect on the promotion decision. See Stern, 131 F.3d at 313. Putting aside the curative measures just mentioned, the simple fact is that the missing CEQs did not change the bottom line finding that all six of the decision makers at Vassar agreed upon: Bickerstaffs CEQs were at best neither “good nor bad” and were, of recent, in marked decline. Bickerstaff herself does not assert—as she cannot—that her CEQs were demonstrably superior to past successful candidates. Bickerstaffs CEQs, in fact, were appreciably lower than those of the six successful candidates in 1994, indicating that Vassar did not point falsely to her teaching ability in denying her promotion. At her deposition, Bicker-staff herself did not dispute that since 1988 there had been a pattern of decline in her CEQs that had become more precipitous in the last two years, 1992-94.

At most, Bickerstaff asserts that her overall CEQs are equal to those of a select few past successful candidates. This argument, however, overlooks that Vassar’s posted criteria for promotion require “[c ]ontinued demonstration of ... teaching of high quality” (emphasis supplied). Vassar alone has the right to set its own criteria for promotion and then to evaluate a candidate’s fitness for promotion under them. 7 See Gant, 630 F.2d *456 at 67 (“A university’s prerogative ‘to determine for itself on academic grounds who may teach’ is an important part of our long tradition of academic freedom.”) (quoting Sweezy v. New Hampshire, 354 U.S. 234, 263, 77 S.Ct. 1203, 1 L.Ed.2d 1311 (1957) (Frankfurter, J., joined by Harlan, J., concurring in the result)). Our role is narrowly limited to determining whether an illegitimate discriminatory reason played a motivating role in the employment decision. See Gant, 630 F.2d at 67. Bicker-staff has not presented evidence that an illegal discriminatory motive played a motivating role in Vassar’s decision to deny her promotion. Instead, the evidence overwhelmingly supports Vassar’s explanation that it denied Bickerstaff promotion for the legitimate, nondiscriminatory reason that she did not satisfy the posted criteria for promotion.

G. Bickerstaffs Affidavit

Bickerstaff next relies on her own affidavit and deposition testimony in which she stated that she had experienced feelings of isolation, hostility, and disinterest in her work in Africana Studies. We also find this tissue-thin contention unpersuasive. Bickerstaffs “[feelings and perceptions] of being discriminated against [are] not evidence” of discrimination. See Fisher I, 70 F.3d at 1439 (“[Plaintiffs] ‘sense of being discriminated against’ is not evidence.”).

H. Professor Ann Constantinople

The last alleged evidence of discrimination that deserves mention concerns Professor Ann Constantinople, the Chair of the FASC that recommended against Bick-erstaff for promotion. According to Bick-erstaff, Constantinople joined in an expression of race bias against her. The basis for this allegation is a written statement in a 1993 FASC Report, of which Constantinople was one of four members, which addressed the tenure application of Teresa Onwueme, another African American joint appointee seeking tenure. Specifically, when reviewing Onwueme for promotion, FASC wrote: “We can ill afford to tenure as associate professor yet another black faculty member who seems destined to be stuck in that rank forever.” Bickerstaff asserts that this statement refers to her and that it is a clear expression of race animus.

Bickerstaff did not, however, present this argument to the attention of the district court. See Bickerstaff, 992 F.Supp. at 375. Although Bickerstaff curiously annexed the FASC Report concerning Onwueme as an exhibit to her affidavit in opposition to summary judgment, she did not cite the FASC Report as evidence of an expression of race bias by Constantinople. Rather, Bickerstaff cited it for the very different proposition that Vassar denied tenure to Onwueme, despite positive recommendations from the Africana Studies Program and the Department of English. “It is the general rule ... that a federal appellate court does not consider an issue not passed upon below ... [unless *457 in] the discretion of the courts of appeals ... the proper resolution is beyond any doubt or ... injustice might otherwise result.” Singleton v. Wulff, 428 U.S. 106, 120-21, 96 S.Ct. 2868, 49 L.Ed.2d 826 (1976) (citations omitted); see also In re Cuyahoga Equip. Corp., 980 F.2d 110, 117 (2d Cir.1992). As Bickerstaff did not present this statement to the district court and has made no showing of manifest injustice to permit consideration of the alleged discriminatory statement, this argument is not properly before us.

Moreover, the Constantinople statement does not lend itself to Bicker-staffs interpretation on appeal. First, despite Bickerstaffs contention that “[i]n context the other ‘black faculty member ... [’] was plaintiff,” Bickerstaff Brief on Appeal at 14 n. 10 (emphasis added), Constantinople’s statement likely did not refer solely to Bickerstaff. Vassar’s two most senior African American faculty members were associate professors. Bickerstaff had held the rank of associate professor since 1979 and Norman Hodges had been an associate professor for seven years longer than Bickerstaff. Thus, there is no reason to infer that the Constantinople statement singled out Bickerstaff.

Second, although the Constantinople statement made reference to race, the statement is a negative prediction, based on an analysis of Onwueme’s work, as to Onwueme’s prospects for promotion. To the extent that the phrase “seems ... to be stuck” referred to individuals other that Onwueme, it appears to be an accurate reference to the then-status of at least Vassar’s two most senior black faculty members. Constantinople testified at her deposition that in order to become a full professor, “[o]ne is theoretically expected to [make an application] after six years [as an associate professor] and every three years thereafter.” Bickerstaff was not pursuing the usual timetable. She first applied for full professorship in 1989, approximately eleven years after becoming an associate professor in 1978. Further, at the time Constantinople made the statement in 1993, Bickerstaff had already been denied promotion to full professor four years earlier, had not yet made another application for promotion at the time Of the statement, and had not published any scholarly articles in the interim. In light of the foregoing and the statement’s use of the word “seems,” it does not appear to be a pre-judgmental prediction of an unmade application by Bickerstaff.

Third, that the Constantinople statement does not lend itself to Bickerstaffs contention on appeal is further supported by the fact that, although she made numerous other arguments in her forty page, eighty paragraph affidavit and thirty-two page memorandum of law in opposition to Vassar’s motion for summary judgment, she did not think to advance this interpretation to the district court. As previously discussed, Bickerstaff submitted the FASC Report concerning Onwueme to the district court for a very different reason than that asserted here on appeal.

In the final analysis, in response to the evidence that Vassar denied Bickerstaff promotion to full professor because she did not satisfy the criteria for promotion, Bick-erstaff does not present sufficient rebuttal evidence from which a rational finder of fact could infer that more likely than not Vassar intentionally discriminated against Bickerstaff because of her race or sex. 8 The district court thus properly granted *458 summary judgment to Vassar College and dismissed Bickerstaffs Title VII and section 1981 claims. Bickerstaffs Equal Pay Act claim, which relies on the statistical evidence that we have already discussed and found that the district court properly discounted, was also properly dismissed.

CONCLUSION

We have considered all of Bickerstaffs arguments on appeal and have found them to be without merit. The judgment of the district court dismissing the complaint is affirmed.

. A "joint appointment” is described in Vassar’s Faculty Handbook as a "contractual agreement to divide a faculty member's re-sponsibililies between a department and a multidisciplinary program.”

. The Faculty Handbook provides that the Dean chooses three outside evaluators, with *442 one evaluator being chosen from each of the separate lists of nominations for outside evaluators submitted by the program, the department and the candidate.

. The CEQs are completed by the students in each class at Vassar, and the questions are responded to with numerical designations on a scale from one to five, in ascendant order based upon excellence.

. As noted, the questions on the CEQs were responded to with numerical designations— one being the lowest and five being the highest. See, e.g., supra at n. 3.

. As noted by the District of Columbia Circuit, “[t]he term 'pretext' can be slippery; sometimes it means that an employer's explanation is incorrect, and sometimes it means both that the explanation is incorrect and that the employer's real reason was discriminatory.” Aka v. Washington Hosp. Ctr., 156 F.3d 1284, 1288 n. 3 (D.C.Cir.1998) (emphasis in original). We too "will avoid using the term 'pretext,' and instead refer (as appropriate) to evidence that the employer's explanation is false, ... or that the employer’s ... motivation was discriminatory.” Id.

. Notably, while the Visiting Committee mentions departments that had been hostile to Africana Studies, the Department of Education is not one of them.

. As the Third Circuit has aptly articulated:

[C]ourts ... should not substitute their *456 judgment for that of the college with respect to the qualifications of faculty members for promotion and tenure. Determinations about such matters as teaching ability, research scholarship, and professional stature are subjective, and unless they can be shown to have been used as the mechanism to obscure discrimination, they must be left for evaluation by the professionals, particularly since they often involve inquiry into aspects of arcane scholarship beyond the competence of individual judges.

Kunda v. Muhlenberg College, 621 F.2d 532, 548 (3d Cir.1980); accord Fields v. Clark Univ., 966 F.2d 49, 54 (1st Cir.1992), cert. denied, 506 U.S. 1052, 113 S.Ct. 976, 122 L.Ed.2d 130 (1993); Brousard-Norcross v. Augustana College Ass’n, 935 F.2d 974, 975-76 (8th Cir.1991).

Judge Calabresi has similarly noted that it is “always pernicious” for a court to set standards for the employer, and "[i]t is particularly pernicious in academic contexts, where, second only to athletics, we all think that we are better coaches than the coaches themselves.” Fisher III, 114 F.3d at 1361 n. 13 (Calabresi, J., concurring in part and dissenting in part). The Fisher III majority joined him on this point.

. We have considered all of the evidence in the totality in reaching our conclusion. Although the individual pieces of evidence proffered by Bickerstaff have been examined in turn, we have done so only as a matter of necessity and practicality. As many of the pieces of evidence do not allow for the reasonable inference of discrimination, they assume in this case no super-inferential powers when viewed together. Similarly, Bicker-staffs remaining pieces of evidence that are of no probative value singularly have no greater probative value collectively.