Isabel v. City of Memphis

BATCHELDER, Circuit Judge,

dissenting.

I dissent because the plaintiffs have failed to demonstrate that the City’s use of a cut-off score, which required applicants to answer 66 of the 100 questions on the written portion of the examination correctly, had an adverse impact on African-Americans in violation of Title VII. At the insistence of the Police Union, the City agreed to impose a cut-off score on the written part of the examination, which applicants would be required to achieve in order to be considered for promotion to lieutenant. Before administering the test, Dr. Mark Jones, an industrial psychologist hired by the City to create the promotional exam, set the cut-off score at 70. With a cut-off score of 70, the pass rate of African-American candidates was less than 45% of that of whites. In order to achieve compliance with the EEOC’s “four-fifths rule,” which says that a selection rate for any race which is more than four-fifths of *417 the rate for the group with the highest rate will generally not be regarded as evidence of adverse impact, see 29 C.F.R. § 1607 .4D (2004), the City adopted Dr. Jones’s recommendation that the cut-off score be lowered to 66. With a cut-off score of 66, the pass rate of African-Americans was 83.4% of that of whites.

Despite the fact that the written exam yielded a pass rate that satisfied the four-fifths rule, the Majority, relying on alternative forms of statistical analyses such as the T— and Z-Tests, concludes that the City’s use of a minimum cut-off score creates an adverse impact in violation of Title VII. In my view, however, the law of this Circuit requires that when the results of the employment action in question satisfy the four-fifths rule, courts are not to look for other statistical analyses that might reveal an adverse impact. Black v. City of Akron, 831 F.2d 131 (6th Cir.1987) a published Sixth Circuit case that makes a cameo appearance in the first footnote of the Majority’s opinion, analyzed whether an examination that satisfied the four-fifths rule still created an adverse impact, and addressed the concerns of some commentators and courts that the four-fifths rule is not a useful indicator of adverse impact when a small sample of employees is being tested. The court stated that “the small size of the sample might be a reason to rule that there was no prima facie case even if the 4/5 rule were violated, but it is certainly not a reason for finding a prima facie case when the 4/5 rule is not violated” and held that the plaintiffs’ statistical evidence was not suggestive of discrimination. Id. at 134 . The Majority distinguishes Black , which it says “did not consider the question of whether it would rely on alternative analyses to find an adverse impact.” That Black did not consider alternative forms of statistical analyses is exactly the point: courts in this Circuit do not consider alternative analyses when the employment practice in question satisfies the four-fifths rule. The Majority’s opinion does not cite, and my research failed to uncover, any cases from this Circuit where the court looked to other statistical analy-ses to find an adverse impact when the results of the employment action in question satisfied the four-fifths rule.

The Majority’s misunderstanding of Black’s rule is evidenced by its citation to N.A.A.C.P v. City Mansfield, 866 F.2d 162 (6th Cir.1989), which it says “implies] that other statistical analyses may be considered,” in support of the proposition that Black “has been offset by subsequent rulings.” The City of Mansfield court did indeed consider alternative forms of analy-ses, such as the “chi-square,” which, according to the plaintiffs expert witness, suggested that the “differing examination pass ratios for whites and blacks on the police test were not attributable to chance.” Id. at 167 . But the results of the test at issue in that case did not satisfy the four-fifths rule, and City of Mansfield, therefore, does not permit the conclusion that we are free to seek out and rely on alternative forms of statistical analysis to find an adverse impact in cases where the four-fifths rule has otherwise been satisfied. 1

*418 For the past 20 years, employers like the City of Memphis have relied on the bright line four-fifths rule to assess whether their employment practices comport with Title VII’s requirements. As of today, however, employers in this Circuit no longer have that assurance that their scrupulous adherence to the four-fifths rule will insulate them from an adverse impact suit. The Majority’s opinion opens the door for plaintiffs to file a Title VII suit based on the newest statistical flavor of the month, but provides absolutely no guidance as to which tests are to be used in assessing whether an employment practice results in an adverse impact, how any tests are to be applied, or how to determine their statistical significance.

Even considering the alternative statistical evidence that the Majority relies on, the disparity between the scores of white and African-American applicants, while statistically significant, does not rise to the level of a Title VII infraction. In Watson v. Forth Worth Bank & Trust, 487 U.S. 977, 994-95 , 108 S.Ct. 2777 , 101 L.Ed.2d 827 (1988) the Supreme Court indicated that a disparity must be substantial: “[o]ur formulations, which have never been framed in terms of any rigid mathematical formula, have consistently stressed that statistical disparities must be sufficiently substantial that they raise such an inference of causation.” The Court in Albe-marle Paper Co. v. Moody stated that, to establish an adverse impact, the complaining party must show “that the tests in question select applicants for hire or promotion in a racial pattern significantly different from that of the pool of applicants.” 422 U.S. 405, 425 , 95 S.Ct. 2362 , 45 L.Ed.2d 280 (1975). In the case at bar, 98 out of the 120 applicants (81.7%) passed the City’s test when a cut-off score of 66 was imposed; 51 out of 57 (89.5%) whites and 47 out of 63 (74.6%) African-Americans passed. I do not think that this small disparity is significant in “practical terms,” see 29 C.F.R. § 1607.4 (D) (“Smaller differences in selection rate may nonetheless constitute adverse impact, where they are significant in both statistical and practical terms ... ”) and therefore, I do not believe it is cognizable under Title VII. What’s more, there is absolutely nothing in the record to suggest that the City purposefully discriminated against African-Americans or ignored a risk that the written test would cause an adverse impact. To the contrary, the City retained an industrial psychologist to create a test that would not cause an adverse impact and even lowered the cut-off score to ensure that more racial minorities, who otherwise would have not passed the test, remained in consideration for promotion to lieutenant.

Accordingly, I would reverse the decision of the district court because the Ap-pellees have failed to show that the City’s use of a cut-off score created an adverse impact in violation of Title VII.

. Despite the fact that the Majority notes that Wards Cove Packing Co., Inc. v. Atonio, 490 U.S. 642, 656-57 , 109 S.Ct. 2115 , 104 L.Ed.2d 733 (1989) "did not say what kind of statistical evidence should be relied on,” it argues that Wards Cove "implfies] that other statistical analyses may be considered.” Even assuming that it supports this proposition, Wards Cove does not "offset” Black's holding. As in City of Mansfield, the Court in Wards Cove held that the plaintiffs’ statistical evidence did not establish a disparate impact in violation of Title VII. 490 U.S. at 661 , 109 S.Ct. 2115 . Wards Cove is not a case where the Court relied on alternative statistical anal- *418 yses to find an adverse impact when the four-fifths rule was otherwise satisfied.