Groves v. Alabama State Bd. of Educ., 776 F. Supp. 1518 (M.D. Ala. 1991)

US District Court for the Middle District of Alabama - 776 F. Supp. 1518 (M.D. Ala. 1991)
October 3, 1991

776 F. Supp. 1518 (1991)

Gregory GROVES, et al., Plaintiffs,

Civ. A. No. 88-T-730-N.

United States District Court, M.D. Alabama, N.D.

October 3, 1991.

*1519 Terry G. Davis, Montgomery, Ala., Gregory B. Stein, Mobile, Ala., Donald Watkins, Montgomery, Ala., for plaintiffs.

David Boyd, Balch & Bingham and Richard Meadows, Denise Azar, Office of Gen. Counsel, State Dept. of Educ., Montgomery, Ala., for defendants.



With this litigation, the court has had to venture again into the often complex and controversial area of teacher testing. Plaintiffs Gregory Groves, Floretta L. Coley, and Arnita Holder brought this civilrights action in 1988, challenging the requirement imposed by defendant Alabama State Board of Education that college sophomores seeking admission to undergraduate teacher training programs in Alabama have achieved a minimum score of 16 on the American College Testing Program's ACT exam or a 17 on the new, "enhanced" ACT exam.[1] Plaintiffs charge that the ACT requirement discriminates against them and other African-Americans on the basis of their race, in violation of their rights under Title VI of the Civil Rights Act of 1964, as amended, 42 U.S.C.A. §§ 2000d through 2000d-4a. They assert both "disparate-treatment" and "disparate-impact" claims, and they seek declaratory and injunctive relief against further enforcement of the ACT requirement. The case was tried before the court in April 1991, with each side presenting extensive documentary evidence and expert testimony. For the reasons explained below, the court concludes that the plaintiffs should prevail on their disparate-impact claim and are thus entitled to the requested relief.[2]


Beginning in the late 1960's, public attention in Alabama increasingly focused on what was perceived to be widespread incompetence among school teachers in the state. Then as now, such teachers were the product of training programs offered by institutions of higher learning in Alabama. Much of the popular criticism was directed at the State Board; however, the Board exercised relatively little control over the process of admission to teacher-training programs, to which college students ordinarily applied at the end of their sophomore year, or the content of such programs, and generally granted certification to all successful teacher-training graduates on the recommendation of their respective colleges. However, reacting to public concerns, the State Board signalled its intention to alter this system in a 1972 resolution containing a number of recommendations for improving teacher certification and preparation, including the establishment of certain minimum standards for admission to teacher-education programs.

In the wake of this resolution, the State Board in 1975 created a special committee to study teacher education and propose specific reforms to improve the quality of training programs and the competence of their graduates. The "Poole Committee," named after its chairperson and board *1520 member Victor Poole, contained representatives from various areas of public education in Alabama. Although the committee adopted a requirement that students seeking acceptance into teacher-training courses attain a certain minimum grade-point average in their first two years of college, most of the members agreed that this alone would not be sufficient to improve the quality of students entering such programs; the State Board did not oversee grading practices, and the committee shared a perception held by much of the public that certain institutions awarded passing marks to students regardless of ability or achievement. Accordingly, the committee settled on the idea of also requiring a particular level of performance on a standardized exam as a prerequisite to entering teacher education. The purpose of such a requirement would be to screen out persons unlikely to become adequately capable teachers.[3] Without significant deliberation or dissent, the committee settled on the ACT exam.

The ACT exam is intended and widely used as one of several components in assessing the academic abilities of high school students applying for admission to college. Students ordinarily sit for the exam during their junior year of high school.[4] The exam's value lies in its ability to predict, albeit at best moderately, a student's first-year college grade-point average.[5] However, the ACT test is not designed to be used as an absolute criterion to select students for specialized under-graduate courses, let alone to gauge whether a student possesses the skills necessary to become a competent teacher, the diagnostic purpose for which the exam was adopted by the Poole Committee.[6] Nevertheless, the Poole Committee chose to include a minimum score on the ACT exam as a prerequisite to admission to a teacher training program, ostensibly because it was a readily available, nationally standardized test that already was routinely taken by most college-bound high school students in Alabama. The committee engaged in little discussion and no study, independent or otherwise, of the usefulness of the ACT exam in predicting teaching ability.

Sometime in late 1975, after the committee had completed the remainder of its work and settled on a set of reforms to recommend to the State Board, several of its members turned their attention to deciding on a "cut-off score" the minimum score that students would have to attain on the ACT exam in order to be eligible for teacher-education courses. During one of the committee's final meetings, Poole ordered the half dozen or so members of the "steering committee" to meet by themselves and not to return to the main committee room until they had agreed on a specific cut-off score. Although the steering committee had collected data on the median ACT scores of Alabama students, *1521 broken down along lines such as college, race, and area of academic concentration, it had made no efforts whatsoever to obtain information on the issue of whether and to what extent any particular score or range of scores on the ACT correlated with competence to teach. Moreover, the members were aware of data demonstrating a wide gap between the median scores of black and white students on the ACT exam, and of the fact that anything more than a minimal cut-off score would exclude a significant, disproportionate number of African-Americans from teacher training.[7] Nevertheless, after a meeting that lasted no more than 20 or 30 minutes, the steering committee decided to recommend that a minimum score of 16 on the ACT exam be included as a prerequisite to entrance into teacher education. This score represented the "39th percentile" of Alabama high school students who took the ACT the previous year in other words, 39% of such students scored at or below this level.[8] The members arrived at this figure as a rough compromise between the need, on the one hand, to avoid establishing a standard so stringent that it would prevent such programs from turning out enough graduates to satisfy the demand for new teachers and, on the other hand, their desire for a proposal that the State Board and educators could "sell" to the public as a reform aimed at guaranteeing well-qualified teachers. One member's comments capture perfectly the way in which this attention to public relations guided the meeting and the steering committee's adoption of a minimum ACT score of 16:

So we went in that little room there, and we looked at one another, and we knew we were playing with fire. We had all these pressures.... You knew you were putting both feet, both hands, in the middle of a philosophic war, a media war, a racial war ...
Finally somebody said, well, what can we take to the people? At that point we forgot the university. We forgot everybody.... What kind of argument we can make that the people gon buy? And some soul in there said, well could we make the argument that the teachers ought to be smarter than half the students. And we looked around. We said, them old boys down there in Letohatchee will buy that. Everybody will buy it. We were all Alabamians. We all good old boys.
We said, we can sell that. Folks in Lowndes County will buy it. Folks up in Wilburn will buy it. Even sophisticates up there in them Birmingham Newspapers, that'll make sense that the teacher ought to be as smart as at least half the students she's teaching.
So [one of the steering committee members] was commissioned to go to his office and find out what the average ACT was for graduates, came back and said, I believe it's 16.4. So our big decision was whether to go to 17 or 16. And the only argument I think I recall them arguing for 16. Then we could go back out and say, looka here. Of course, this is also a fallacious argument because the student the teacher never is as smart as half the students.... [But] that was the scientific basis of it gentlemen and lady. It was just that scientific.

The full committee accepted the propsal of a cut-off score of 16 without debate. The State Board, in turn, passed a resolution in March 1977 adopting the admissions criteria recommended by the committee, including the requirement of a score of 16 or higher on the ACT exam. The resolution allowed students to satisfy the ACT requirement *1522 by taking the exam at any time or, if necessary, multiple times within five years of submitting an application to a teacher-training program. However, as most students take the ACT in their junior year in high school, the requirement, in practical terms, has operated on the assumption that scores on the test are capable of predicting the skills a teacher-hopeful will possess five or six years in the future, when he or she graduates from college.

In addition to the ACT requirement, the resolution adopted other prerequisites including a minimum grade-point average of at least 1.2 on a three-point scale on work completed during the first two years of college, a passing grade on a Board-approved "English Language Proficiency Test," and a satisfactory personal interview. Although this resolution called for several "impact studies" of the new standards to be performed in coming years, no such studies nor any other examination of the effect of the minimum score criterion on the quality of teachers or teacher candidates were ever undertaken by the Board. The ACT requirement adopted in 1977 remains in place today. The only alteration in it occurred in 1989, when the American College Testing Program modified the exam, and renamed it the "enhanced" ACT. Because this change meant that scores would no longer correspond to those on the old test, the State Board recently adopted 17 as the minimum score on the enhanced ACT necessary for a student seeking admission to a teacher education program. A score of 17 on the enhanced ACT is roughly equivalent to a score of 16 on the old ACT.

In July 1988, plaintiffs in this lawsuit challenged the ACT requirement as racially discriminatory.[9] Arguing that both its purpose and practical effect was to exclude a large, disproportionate percentage of blacks from teacher training and thus careers in education, plaintiffs charged that the ACT requirement violated Title VI of the Civil Rights Act of 1964, as amended, 42 U.S.C.A. §§ 2000d through 2000d-4a.[10] In October 1988, relying on the parties' joint stipulation, the court certified a plaintiff class consisting of "all past, present, and future black students ... denied admission[] into a teacher education program at institutions of higher education in Alabama solely by reason of failing to attain an ACT score of 16."[11] Moreover, throughout this litigation, the State Board and the plaintiffs have implicitly acknowledged, in the evidence and arguments they have offered in this case, that the switch from a cut-off score of 16 on the ACT to one of 17 on the enhanced ACT is not significant to the court's analysis of the ACT requirement's discriminatory impact or validity.

A. Title VI

Title VI provides that "No person in the United States shall, on the ground of race, ... be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance." 42 U.S.C.A. § 2000d. The evidence is undisputed that all the operations of the State Board, including its imposition of the ACT *1523 requirement on admissions decisions by college teacher education programs, constitute a "program ... receiving Federal financial assistance." See also § 2000d-4a (defining "program" within meaning of Title VI).

Proof of discriminatory intent is not a prerequisite to equitable relief in a private cause of action under Title VI. Franklin v. Gwinnett County Public Schools, 911 F.2d 617, 620 (11th Cir. 1990), cert. granted, ___ U.S. ___, 111 S. Ct. 2795, 115 L. Ed. 2d 969 (1991); Castaneda v. Pickard, 781 F.2d 456, 465 n. 11 (5th Cir.1986). Rather, where a lawsuit is predicated on the federal regulations promulgated under Title VI, redress is available for facially neutral "actions having an unjustifiable disparate impact on minorities." Georgia State Conference of Branches of NAACP v. Georgia, 775 F.2d 1403, 1417 (11th Cir. 1985), quoting Alexander v. Choate, 469 U.S. 287, 293, 105 S. Ct. 712, 717, 83 L. Ed. 2d 661 (1985). Accord Guardians Ass'n v. Civil Service Comm'n the City of New York, 463 U.S. 582, 607 n. 27, 103 S. Ct. 3221, 3235 n. 27, 77 L. Ed. 2d 866 (1983). Plaintiffs in this case rely on the administrative regulations that are keyed to adverse racial impact. For example, 34 C.F.R. § 100.3(b) (2) provides that "A recipient ... may not ... utilize criteria or methods of administration which have the effect of subjecting individuals to discrimination because of their race". Accord 34 C.F.R. § 100, Appendix B., Part K. Because, as shown below, plaintiffs are entitled to prevail on their "disparate-impact" claim, the court need not address their claim of intentional discrimination or disparate treatment.

The elements of a disparate-impact claim under Title VI's regulations are substantially similar to those applicable in an employment discrimination action under Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C.A. §§ 2000e through 2000e-17.[12]Georgia State Conference, 775 F.2d at 1417. Accord Quarles v. Oxford Municipal Separate Sch. Dist., 868 F.2d 750, 754 n. 3 (5th Cir.1989). The regulations require a court to examine two issues: whether a challenged practice has a sufficiently adverse racial impact in other words, whether it falls significantly more harshly on a minority racial group than on the majority and, if so, whether the practice is nevertheless adequately justified. Georgia State Conference, 775 F.2d at 1417; Quarles, 868 F.2d at 754 n. 3.[13] Although both Georgia State Conference and Quarles articulate the proof necessary to sustain a disparate-impact claim under Title VI's regulations, the Title VII law from which they borrow has since been redefined, particularly by the Supreme Court in Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 109 S. Ct. 2115, 104 L. Ed. 2d 733 (1989). Accordingly, the court relies principally on Wards Cove and other, subsequent Title VII decisions in evaluating plaintiffs' challenge to the ACT requirement in the following sections of this opinion.

B. Adverse Impact

In Wards Cove, the Supreme Court carefully recast the elements necessary to make out a prima-facie case on a disparate-impact theory under Title VII. In order to carry the burden of showing adverse racial impact, a plaintiff must initially demonstrate that a specific selection practice engaged in by an employer has caused the exclusion of the plaintiff and similarly situated applicants for jobs or other employment *1524 opportunities.[14]Wards Cove, 490 U.S. at 656-57, 109 S.Ct. at 2124-25; MacPherson v. University of Montevallo, 922 F.2d 766, 771 (11th Cir.1991). See generally Leading Cases, 103 Harv.L.Rev. 137, 350-61 (1989). While observing that "statistical proof can alone make out a prima facie case," the Court in Wards Cove emphasized the importance of limiting any statistical comparison of "the racial composition of the at-issue jobs" to "the racial composition of otherwise qualified applicants." 490 U.S. at 650, 109 S. Ct. at 2121. See also Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 308 n. 13, 97 S. Ct. 2736, 2742 n. 13, 53 L. Ed. 2d 768 (1977); Nash v. Consol. City of Jacksonville, Duval County, Fla., 905 F.2d 355, 358 (11th Cir.1990), cert. denied, ___ U.S. ___, 111 S. Ct. 967, 112 L. Ed. 2d 1054 (1991). Absent such carefully narrowed statistical proof, a factfinder may lack a sufficient basis to conclude that the causation requirement is satisfied in other words, that a greater proportion of members of the minority group would have been selected absent the challenged practice. See Gilty v. Village of Oak Park, 919 F.2d 1247, 1254-55 (7th Cir.1990). See also Watson v. Forth Worth Bank and Trust, 487 U.S. 977, 992-995, 108 S. Ct. 2777, 2787-89, 101 L. Ed. 2d 827 (1988) (O'Connor, J., plurality opinion).[15]

Both sides agree that, since 1977, a significant, far larger percentage of black students than of whites who have taken the ACT in Alabama have failed to meet the minimum-score requirement imposed by the State Board. A 1978 study by the American College Testing Program revealed that 85% of white Alabamians were achieving scores of 16 or above, but that only 21% of African-Americans were satisfying this threshold. More recent figures indicate that this broad disparity has continued; between 1983 and 1988, 73.6% of white test-takers scored 16 or above on the ACT exam, while 32.7% of blacks obtained similar results. And this gap widens when one looks to only those students who indicated an intent to major in education at the time they sat for the exam. Of this group, 19.6% of blacks achieved the minimum cutoff score, while 63.8% of white students reached such a level. Nevertheless, the Board argues, relying on Wards Cove, that in determining whether the ACT requirement has an adverse racial impact, the court should not look to the scores of all Alabama students who took the exam or even all those who declared they would major in education, but rather should rely only on statistics regarding the rates at which black and white students who were otherwise qualified to enter a teacher education program in other words, who also later passed the English Language Proficiency Test and achieved a 1.2 grade point average satisfied the ACT requirement. According to the Board, such statistics for the period between 1977 and 1984 reveal a difference between the black (74.8%) and white pass rates (90.9%) insufficient to demonstrate adverse racial impact under Title VII and, by extension, Title VI. The Board has asked the court to rely on ACT exam scores between 1977 and 1984 and has not argued that subsequent changes in the performance of blacks and whites on *1525 the ACT in any way alters the import or relevance of these statistics.


The State Board is obviously correct in pointing out the overinclusiveness of plaintiffs' statistical samples, one of which embraces all students who sat for the ACT exam and the other those test-takers who indicated an intention to major in education at the time they took the ACT. When it adopted the ACT requirement, the Board also imposed several additional prerequisites to access to teacher-education courses, not challenged by plaintiffs, of which the two most important and selective are that a student maintain a minimum grade-point average of 1.2 in the first two years of college and that the student also achieve a passing grade on the Board-approved English Language Proficiency Test. Significant numbers of students have been rendered ineligible for admission to teacher training not only because they failed to score 16 or higher on the ACT, but because they also did not satisfy the minimum grade requirement or did not pass the English exam. These students were not within the pool of qualified applicants to teacher education programs in other words, for them, the ACT requirement cannot truly be said to be the specific cause of their exclusion from such a program.[16]

On the other hand, the Board's proposed "pool" is notably underinclusive. The Board's own statistics and evidence from its expert witness make clear that many if not a majority of students who would otherwise pursue a career in education but have failed to attain a 16 on the ACT not only neglect to apply to teacher training courses because they are ineligible for admission,[17] but also refrain from taking the English Language Proficiency Test, for the very same reason.[18] Indeed, approximately 74% of all students who took the ACT between 1983 and 1988 and declared they would major in education never subsequently sat for the English exam. Although some of these students may have changed their career plans out of simple preference, it is reasonable to presume that some did so in reaction to failing the ACT. The ACT requirement excluded these students from teacher education just as it did ones who scored below 16 on the ACT but persevered and passed the English exam. The court has no way of knowing whether or to what extent adding these students would alter the racial breakdown of the Board's proposed pool and thus the adverse impact demonstrated by such statistics. However, the fact that a larger percentage of blacks (88%) than of whites (73%) who took the ACT and indicated they would major in education later failed to sit for the English exam, certainly offers reason to question the accuracy of the Board's calculations concerning adverse impact.

It is, therefore, apparent that neither the plaintiffs' nor the Board's set of figures *1526 perfectly captures the "relevant pool" for the purposes of this case. Indeed, the evidence reflects that, because of the several-year delay between the time students sit for the ACT, usually in their junior year of high school, and the time they complete two years of college coursework, take the English Language Proficiency Test, and apply for admission to a teacher training program at the end of their sophomore year of college, it would be infeasible, if not impossible, to attempt to gather information sufficient to compile complete and fully accurate statistics on the extent to which the minimum-score requirement disproportionately excludes black students from eligibility for teacher education. Nevertheless, the court is convinced that each side's evidence has substantial probative or circumstantial value and that, based on their evidence, the statistical "truth" regarding the racial impact of the ACT requirement lies somewhere between the two poles the parties have erected for the court. The court has, in effect, been presented with detailed views of various aspects of the admissions process, and, from these different perspectives, the court is comfortable in concluding that more likely than not the ACT requirement has had a disparate impact on black students.


In any event, even without discounting the Board's statistics, plaintiffs have still successfully demonstrated that the ACT requirement has a significant disparate impact on blacks. Relying on the Board's figures on students who failed the ACT but met the other requirements for entrance into a teacher education program while bearing in mind the limits of these statistics and the partial relevance of plaintiffs' broader sample of black and white ACT scores the court finds that plaintiffs have proven this element of their Title VI claim.

There is no rigid mathematical threshold that must be met to demonstrate a sufficiently adverse impact on blacks in a Title VII suit. Watson, 487 U.S. at 994-95, 108 S. Ct. at 2789 (O'Connor, J., plurality opinion); Richardson v. Lamar County Bd. of Educ., 729 F. Supp. 806, 816 (M.D.Ala.1989) (Thompson, J.), aff'd, 935 F.2d 1240 (11th Cir.1991). The Supreme Court's "formulations ... have consistently stressed that statistical disparities must be sufficiently substantial that they raise ... an inference of causation" in other words, adequate to "show that the practice in question has caused the exclusion of applicants for jobs or promotions because of their membership in a protected group." Watson, 487 U.S. at 994-95, 108 S. Ct. at 2789 (O'Connor, J., plurality opinion).[19] However, the Supreme Court and lower courts have adopted various formulas to measure the degree of disparate impact in a particular case. The Equal Employment Opportunity Commission generally infers adverse racial impact where the members of a particular racial group are selected at a rate that is less than four-fifths, or 80%, of the rate at which the group with the highest rate is selected. Uniform Guidelines on Employee Selection Procedures (hereinafter referred to as "Uniform Guidelines"), 29 C.F.R. § 1607.4(D). See Watson, 487 U.S. at 995 n. 3, 108 S. Ct. at 2789 n. 3 (O'Connor, J., plurality opinion); Nash, 905 F.2d at 358. See also Connecticut v. Teal, 457 U.S. 440, *1527 444 n. 4, 102 S. Ct. 2525, 2529 n. 4, 73 L. Ed. 2d 130 (1982). However, an increasing number of court decisions, particularly from the Eleventh Circuit, have instead or in addition employed one of several forms of statistical analysis to reach reliable inferences about racial disparities in a population based on the performance of a particular sample. See Powers v. Alabama Dep't of Educ., 854 F.2d 1285, 1298 & n. 21 (11th Cir.1988), cert. denied, 490 U.S. 1107, 109 S. Ct. 3158, 104 L. Ed. 2d 1021 (1989); Kilgo v. Bowman Transportation, 789 F.2d 859, 871 & n. 18 (11th Cir.1986); Maddox v. Claytor, 764 F.2d 1539, 1551-52 (11th Cir.1985). See also Bridgeport Guardians, Inc. v. City of Bridgeport, 933 F.2d 1140, 1146 (2nd Cir.1991); Emanuel v. Marsh, 897 F.2d 1435, 1440-41 (8th Cir. 1990). Most of these decisions rely on the Supreme Court's approval of "standard-deviation" analysis in the Title VII context in Hazelwood, 433 U.S. at 309-12, 97 S. Ct. at 2742-44. Accord Watson, 487 U.S. at 995 n. 3, 108 S. Ct. at 2789 n. 3 (O'Connor, J., plurality opinion). Although no particular number of standard deviations automatically translates into a finding of disparate impact, a difference between the observed and expected values of more than three standard deviations is usually recognized as statistically significant in other words, adequate to support an inference that a racially differential selection pattern is a consequence of the challenged practice rather than a random distribution. Hazelwood, 433 U.S. at 308 n. 14, 97 S. Ct. at 2742 n. 14; Powers, 854 F.2d at 1298 n. 21; Maddox, 764 F.2d at 1552; Kilgo, 789 F.2d at 871; Richardson, 729 F. Supp. at 816. Finally, in addition to the "Hazelwood formula," a somewhat different test of statistical significance, the "Shoben formula," has sometimes been employed to assess disparate impact in Title VII cases. Shoben, Differential Pass-Fail Rates in Employment Testing: Statistical Proof Under Title VII, 91 Harv.L.Rev. 793, 797-800 (1978) (hereinafter referred to as "Shoben"); see also Frazier v. Consol. Rail Corp., 851 F.2d 1447, 1450 & n. 5 (D.C.Cir. 1988); Richardson, 729 F. Supp. at 816. According to this standard, referred to in technical terms as "testing the difference between independent proportions," Shoben, at 796, a difference or "Z" value greater than 1.96 standard deviations is ordinarily sufficient to support a finding of adverse impact. Richardson, 729 F. Supp. at 816.

The Board's figures for students who satisfied the minimum grade and English language requirements for entrance into teacher training reveal that out of a total of 3,393 such students, during the years 1977 to 1984, 3024 or 89.1% scored 16 or higher on the ACT. Of the 2,991 white students in the group, 2720 or 90.9% passed the test, while only 285 or 74.8% of the 381 black test-takers performed similarly.[20] Applying the four-fifths rule, these numbers reveal a racially disparate impact of 82.3%. Under the Hazelwood formula, the difference is 10.7 standard deviations. Finally, using the Shoben formula, the "Z" value is 9.5 standard deviations.[21]

From this welter of statistics, the Board predictably fastens upon the result under the four-fifths standard, and concludes that the ACT requirement does not disproportionately exclude blacks from teacher education courses in light of the fact that its adverse impact is 82.3%, slightly short of the 80% threshold. However, the Board's argument does not hold up under a broader analysis. Admittedly, the four-fifths rule has "been criticized on technical grounds" and viewed as "not provid[ing] more than a rule of thumb for the courts." Watson, 487 U.S. at 995 n. 3, 108 S. Ct. at 2789 n. 3 (O'Connor, J., plurality opinion). See also Guardians Ass'n v. Civil Service Comm'n of City of New York, 630 F.2d 79, 91 (2nd *1528 Cir.1980) (Uniform Guidelines "are entitled to deference, not obedience"), cert. denied, 452 U.S. 940, 101 S. Ct. 3083, 69 L. Ed. 2d 954 (1981); Shoben, 812 ("the four-fifths rule produces anomalous results because it does not account for differences in sample size and does not consider the magnitude of differences in pass rates"); B. Schlei & P. Grossman, Employment Discrimination Law, at 40 (2d ed. Supp.1989).[22] Nevertheless, it is not necessary for the court to determine that statistical significance tests such as the Hazelwood and Shoben formulas necessarily carry greater weight than application of the four-fifths rule in order to conclude that plaintiffs have demonstrated a sufficiently racial pattern of selection created by operation of the ACT requirement. It may well be true that a measure of the difference between black and white selection rates, such as the four-fifths rule, incorporates an important test of practical significance not captured by the Hazelwood and Shoben formulas' focus on statistical significance. See P. Meier, J. Sacks, & S. Zabell, What Happened in Hazelwood: Statistics, Employment Discrimination, and the 80% Rule, 1 Am.Bar Ass'n Research J. 139, 163-164 (1984) (hereinafter referred to as "Meier, Sacks, & Zabell"). It may also be true that proof which demonstrates only a small or moderate level of significance under a statistical analysis and falls short of the four-fifths threshold by a significant margin, would be insufficient to evidence disparate impact. See Moore v. Southwestern Bell Telephone Co., 593 F.2d 607, 608 (5th Cir.1979) (sustaining district court finding of no adverse racial impact where statistical significance analysis revealed 3.93 standard deviations difference between expected and actual black pass rates, but pass rate on employment test for black applicants was 93% of that of white applicants); Meier, Sacks, and Zabell, supra, at 163 n. 77. In this case, however, the racial pattern of success and failure with respect to the ACT requirement has both practical and overwhelming statistical significance. The result of analysis under the four-fifths rule is at most a borderline figure while both the Hazelwood and Shoben tests of statistical significance overwhelmingly demonstrate that the disparity between the percentages of "otherwise qualified" black and white students who scored 16 or higher on the ACT is so substantial as to indicate that the test itself, and not merely random sampling, has caused the disproportionate exclusion of blacks. Relying on all three tests, the court is convinced, and so finds, that more likely than not there has been racially disparate impact on black students. See Hill v. Metropolitan Atlanta Rapid Transit Auth., 591 F. Supp. 125, 129 (N.D.Ga.1984) (Forrester, J.) (declining to follow exclusively the four-fifths rule where the selection rate for minorities was 81.55% of that of majority candidates, but there was expert testimony that the disparity nevertheless was statistically significant), aff'd in part and rev'd and remanded in part on other grounds, 841 F.2d 1533, amended on reh'g, 848 F.2d 1522 (11th Cir.1988).

Moreover, the Uniform Guidelines, which are the basis for the four-fifths rule, clearly state that smaller differences between the selection rates for blacks and whites in other words, ratios greater than four-fifths "may nevertheless constitute adverse impact" in circumstances such as those found in this case, "where [the differences] are significant in both statistical and practical terms" or "where the user's actions have discouraged applicants disproportionately on grounds of race." 29 C.F.R. § 1607.4(D). As the court has discussed, there is evidence from which it is reasonable to infer that many students who fail to attain a score of 16 on the ACT and significantly more blacks than whites surrender their hopes for a career in education and never even sit for the English Language Proficiency Test.


Finally, in wrestling with complex, technical evidence in testing cases, courts should keep in mind that "the science of *1529 testing is not as precise as physics or chemistry, nor its conclusion as provable." Guardians Ass'n, 630 F.2d at 89. "While courts should draw upon the findings of experts in the field of testing, they should not hesitate to subject these findings to ... the scrutiny of reason...." Id. (emphasis added). See also Watson, 487 U.S. at 995 n. 3, 108 S. Ct. at 2789 n. 3 (O'Connor, J., plurality opinion) (statistical evidence should be considered on "a case-by-case basis," with the understanding that "statistics `come in infinite variety and their usefulness depends on all of the surrounding facts and circumstances'"), quoting Int'l Brotherhood of Teamsters v. United States, 431 U.S. 324, 340, 97 S. Ct. 1843, 1856-57, 52 L. Ed. 2d 396 (1977). Here, both the plaintiffs and the State Board have wrapped themselves in complex statistical data and terminology. However, this is one of those rare cases where if one stands back and applies reason and common sense the answer is apparent. To be sure, neither the plaintiffs nor the State Board was able to fashion a perfect statistical picture, and, indeed, it appears that the task may be impossible. Nevertheless, the very limited, singular perspective that the Board would have the court adopt as circumstantial proof of no adverse racial impact cannot outweigh the overwhelming circumstantial evidence presented from other perspectives to the effect that the ACT requirement has resulted in substantial adverse racial impact. Indeed, to reach any other conclusion the court would have to close its eyes to the obvious.

Having thus found that the ACT requirement falls more harshly on blacks than whites to a sufficiently substantial degree, the court now turns to the second element of plaintiffs' challenge to the minimum score rule, the issue of whether, despite its adverse impact, this requirement is nevertheless justified as a valid method for screening out persons unqualified to become teachers.

C. Educational Justification

In Wards Cove, the Supreme Court explained that, "at the justification stage of ... a disparate impact case, the dispositive issue is whether a challenged practice serves, in a significant way, the legitimate employment goals of the employer." 490 U.S. at 659, 109 S. Ct. at 2125 (emphases added). Although "there is no requirement that the challenged practice be `essential' or `indispensable' to the employer's business for it to pass muster," the Court made clear that "A mere insubstantial justification in this regard will not suffice."[23]Id. at 659, 109 S. Ct. at 2126. Accord MacPherson v. University of Montevallo, 922 F.2d 766, 771 (11th Cir.1991).[24] In this *1530 phase of a discrimination suit, the employer carries the burden of "producing evidence of a business justification for his employment practice. The burden of persuasion, however, remains with the disparate-impact plaintiff."[25]Wards Cove, 490 U.S. at 659-60, 109 S. Ct. at 2126. Accord MacPherson, 922 F.2d at 771. In evaluating whether the use of an objective test is supported by a business justification, courts have looked to whether the test is "valid" in other words, whether an applicant's score on the test yields an appropriate and meaningful inference about the applicant's successful performance on the job.[26]Richardson, 729 F. Supp. at 820. See Nash v. Consol. City of Jacksonville, Duval County, Fla., 837 F.2d 1534, 1537 (11th Cir. 1988), vacated, 490 U.S. 1103, 109 S. Ct. 3151, 104 L. Ed. 2d 1015 (1989), reinstated, 905 F.2d 355 (11th Cir.1990), cert. denied, ___ U.S. ___, 111 S. Ct. 967, 112 L. Ed. 2d 1054 (1991). As the court has explained, it must borrow from Title VII law in determining whether the ACT requirement is sufficiently justified to withstand plaintiffs' Title VI challenge, substituting the notion of "education" justification for that of business justification as used in the Title VII context.[27]

Because the practice challenged by plaintiffs in this case is the requirement of a minimum score of 16, it is unnecessary for the court to address the broader issue of whether any use of the ACT exam as a criterion in admission to teacher-training programs would pass muster under Title VI despite a showing of adverse racial impact. It is clear that, even if the ACT exam itself is an appropriate measure of teaching ability, the existing ACT requirement would not be educationally justified if the particular cut-off score used by the Board to determine the eligibility of applicants is not itself a valid measure of the minimal ability necessary to become a competent teacher. Richardson, 729 F. Supp. at 821; see also Evans v. City of Evanston, 881 F.2d 382, 383-85 (7th Cir.1989); Gillespie v. Wisconsin, 771 F.2d 1035, 1045 (7th Cir.1985), cert. denied, 474 U.S. 1083, 106 S. Ct. 854, 88 L. Ed. 2d 894 (1986); Guardians Ass'n, 630 F.2d at 105-06; 29 C.F.R. § 1607.5(H); B. Schlei & P. Grossman, Employment Discrimination Law, at 46 & n. 88 (2d ed. Supp.1989).

Because the Board has presented literally no evidence that a score of 16 marks the boundary at which teaching proficiency begins,[28] and because plaintiffs have overwhelmingly *1531 proved that such an absolute level bears no logical let alone significant relationship to minimal competence as a teacher, it is also unnecessary to enter into an extended discussion of the various professionally acceptable methods for validating a cut-off score for an objective exam. See Nash, 837 F.2d at 1537. Here, the Poole Committee, with the approval of the Board, arbitrarily selected a minimum score of 16 essentially as a public relations ploy, so that the Board and education professionals could misrepresent to parents concerned about their children's schooling that it was faithfully ensuring all new teachers would be "as smart" as half or to be more exact, 39% of their students.[29]See York v. Alabama State Bd. of Educ., 581 F. Supp. 779, 788 n. 13 (M.D.Ala.1983) (Thompson, J.) (characterizing as "homemade methodology" county school system's procedure for selecting cut-off score for teacher certification exam by reference to mean scores). There is no rational basis, let alone any professional research or study relevant to the minimum score rule, from which to infer that otherwise qualified students scoring at or above this level will be competent to teach several years in the future, while those failing to achieve a 16 on the ACT will not. Evans, 881 F.2d at 384 (invalidating city's choice of one standard deviation above mean as cut-off score for physical agility for fire-fighting positions). See also United States v. Lulac, 793 F.2d 636, 643 (5th Cir.1986); Gillespie, 771 F.2d at 1045. Indeed, what is known about the validity of the ACT indicates that it is intrinsically unsuited to be used as an absolute criterion, particularly to determine future teaching ability, something far afield from its narrow, intended role in college admissions decisions.[30]

In reaching this determination that the Board's use of the ACT is impermissible under Title VI, the court has been sensitive both to the fact that an appropriate "passing" score on any test cannot be established with mathematical certainty, and to the broader consideration that educational testing practices are properly within the realm of state and local officials rather than appointed judges. Watson, 487 U.S. at 999, 108 S. Ct. at 2791 (O'Connor, J., plurality opinion); Richardson, 729 F. Supp. at 824. However, although a court must "eschew an idealistic view of test validity" and avoid the temptation to second-guess the professional judgments of those responsible for such tests, it must not, as the Board essentially asks it to do in this case, "apply an `anything goes' view." Id. at 825. Here, however, the Board's decision to adopt the ACT requirement "fall[s] far outside the bounds" of even "a good faith exercise of professional judgment." Id. at 823.

Indeed, the State Board's effort to justify the cut-off score is so feeble that it is tempting to declare that the Board did not even carry its burden of production under Wards Cove.[31]See Nash, 905 F.2d at 358. *1532 However, the court need not dwell on this issue, for even if the Board did offer evidence sufficient to shift the inquiry back to whether the plaintiffs have carried their burden of persuasion, the court concludes, as already shown above, that the plaintiffs have more than amply carried this burden by demonstrating that the ACT requirement's racially adverse impact is not significantly justified by any legitimate educational rationale.[32]See Evans, 881 F.2d at 385.

D. Relief

For these reasons, the court holds that Title VI entitles the plaintiffs to relief from the State Board's ACT requirement of a minimum score of 16 on the old exam or 17 on the new, enhanced exam. The court therefore will enjoin the State Board from further use of the requirement. The court will also require that the Board pay reasonable attorney's fees and expenses to plaintiffs' attorneys. 42 U.S.C.A. § 1988. The court will, however, give the parties an opportunity to agree to such fees and expenses before setting them for hearing.


Finally and perhaps most importantly, the court notes that its holding today does not erect an insurmountable, or unreasonably substantial, barrier to the State of Alabama's efforts to reform and improve teacher-training programs. This case, the court hopes, is one of a dying breed; it is akin to those early disparate-impact cases which "involved tests that were so artlessly constructed that they could be judged invalid without extensive inquiry, fine distinctions, or a precise notion of where the line between validity and invalidity was located." Guardians Ass'n, 630 F.2d at 88. The import of the court's holding is that the state can no longer rest on mere "homemade methodologies" but rather must be professional in its approach to the important public issue of teacher competency. Just as the state expects its teachers to measure up to the more exacting professional demands of today's education system it itself must do likewise. Contrast United States v. City of Montgomery, 775 F. Supp. 1450 (M.D.Ala.1991) (Thompson, J.) (finding that the City of Montgomery Police *1533 Department's carefully crafted sergeant's exam is permissible under Title VII even though it has a racially disparate impact on black officers).

An appropriate judgment will be entered in accordance with this opinion.


In accordance with the memorandum opinion entered this date, it is the ORDER, JUDGMENT, and DECREE of the court:

(1) That judgment be and it is hereby entered in favor of plaintiffs Gregory Groves, Floretta L. Coley, and Arnita Holder, and the class of all past, present, and future African-American students who are being, have been, or will be denied admission into a teacher-education program at institutions of higher education in the State of Alabama by reason of failing to attain a minimum score of 16 on the American College Testing Program's ACT exam or a 17 on the new, "enhanced" ACT exam, and against defendants the State Superintendent of Education, the Alabama State Board of Education, and the members of the State Board;

(2) That the policy of denying admission into teacher-education programs at institutions of higher education in the State of Alabama, of all students who have failed to attain a minimum ACT score of 16 or a 17 on the new, "enhanced" ACT exam be and it is hereby DECLARED illegal under Title VI of the Civil Rights Act of 1964, as amended, 42 U.S.C.A. §§ 2000d through 2000d-4a;

(3) That defendants the State Superintendent of Education, the Alabama State Board of Education, and the members of the State Board, and their officers, agents, servants, and employees and those persons in active concert or participation with them who receive actual notice of this order, be and they are hereby ENJOINED and RESTRAINED from denying any of the above-named plaintiffs or members of the plaintiff class admission into teacher-education programs at institutions of higher education in the State of Alabama by reason of such person's failure to attain a minimum ACT score of 16 or an "enhanced" ACT score of 17; and

(4) That the plaintiffs be and they are hereby allowed 28 days from the date of this order to file a request for reasonable attorney's fees and expenses.

It is further ORDERED that this court retains jurisdiction of this case until further order.

It is further ORDERED that all costs of these proceedings be and they are hereby taxed against defendants, for which execution may issue.

The clerk of the court is DIRECTED to issue a writ of injunction.


[1] The lawsuit also names as defendants the State Superintendent of Education and the individual members of the State Board. However, because the issues in this case are identical with respect to all defendants, the court does not treat them separately, but refers to them collectively as the Alabama State Board of Education.

[2] Plaintiffs also charge that the ACT requirement violated the Equal Educational Opportunities Act, 20 U.S.C.A. § 1703, and the equal-protection clause of the fourteenth amendment to the United States Constitution. Because the court concludes that the ACT requirement contravenes Title VI, it need not address plaintiffs' claims under the Equal Educational Opportunities Act or the equal-protection clause. Moreover, these latter provisions do not afford any greater protection to plaintiffs in this case than does Title VI. See Georgia State Conference of Branches of NAACP v. Georgia, 775 F.2d 1403, 1422 n. 34 (11th Cir.1985).

[3] The Board denies that it adopted the ACT requirement for the purpose, in whole or part, of predicting which students would succeed as students in a teacher-education program. Accordingly, in section II(C) of this opinion, the court does not evaluate this possible justification for the requirement. Compare Washington v. Davis, 426 U.S. 229, 250, 96 S. Ct. 2040, 2052, 48 L. Ed. 2d 597 (1976); Ensley Branch of NAACP v. Seibels, 616 F.2d 812, 821-22 (5th Cir.), cert. denied, 449 U.S. 1061, 101 S. Ct. 783, 66 L. Ed. 2d 603 (1980); United States v. South Carolina, 445 F. Supp. 1094, 1115-16 (D.S.C.1977) (per curiam) (three-judge court), aff'd, 434 U.S. 1026, 98 S. Ct. 756, 54 L. Ed. 2d 775 (1978).

[4] However, at those Alabama institutions of higher education like Alabama State University which do not require the ACT as a prerequisite to admission, beginning freshmen are ordinarily required to take the exam during their first week of college.

[5] As the test designer itself acknowledges, the ACT's utility even for this purpose is limited, particularly with respect to minority applicants "whose prior educational opportunities have been limited." See also Ayers v. Allain, 893 F.2d 732, 736 (5th Cir.), reh'g granted, 898 F.2d 1014 (5th Cir.), on reh'g 914 F.2d 676 (5th Cir.1990), cert. granted, ___ U.S. ___, 111 S. Ct. 1579, 113 L. Ed. 2d 644 (1991). Indeed, a 1978 letter to the Board from the American College Testing Program expressing reservations about the ACT requirement warned that "many students who obtain[] low ACT test scores subsequently perform[] adequately in college."

[6] Although most states require or allow for the use of standardized tests to select students for certified teacher training programs, only four states other than Alabama mandate reliance on the ACT exam in such decisions.

[7] Research possessed by the committee on the median ACT scores at various institutions revealed in graphic terms that such a requirement would have a far more severe impact on predominantly black schools such as Alabama State University and Alabama A & M University, than on such traditionally white schools as Auburn University and the University of Alabama at Tuscaloosa.

[8] It is undisputed that, because the ACT is a nationally standardized or "normed" test, a score of 16 or higher in the years subsequent to 1975 has not necessarily continued to represent the 39th percentile of Alabama test-takers, but rather that the proportion of Alabama high school students scoring at or above this level has fluctuated.

[9] Plaintiffs originally sought to intervene in another case pending in this court, Allen v. Alabama State Bd. of Educ., Civil Action No. 81-697-N, which involved a race discrimination claim against Alabama's teacher certification practices. See Richardson v. Lamar County Bd. of Educ., 729 F. Supp. 806, 811-12 (M.D.Ala. 1989) (Thompson, J.) (discussing history of Allen litigation), aff'd, 935 F.2d 1240 (11th Cir. 1991). Several weeks after plaintiffs filed a complaint-in-intervention, the parties agreed and the court ordered that the complaint be treated as a separate lawsuit.

[10] See note 2, supra.

[11] Order entered October 25, 1988. The court also certified a second plaintiff class, comprised of "all past, present and future black students who have matriculated through teacher education programs ... with the expectation that they would be eligible for a teacher certificate, but have been denied certification solely by reason of not having attained an ACT score of 16 or higher." Since that time, however, the Board has granted teacher certificates to all members of the second class; therefore, the claims raised by members of the second class are moot and need not be addressed by the court.

[12] The State Board acknowledges that plaintiffs' Title VI claim should be judged under a disparate-impact standard borrowed from Title VII law.

[13] See also 34 C.F.R. § 100, Appendix B., Part K. ("Recipients may not judge candidates for admission to vocational education programs on the basis of criteria that have the effect of disproportionately excluding persons of a particular race, color, national origin, sex, or handicap. However, if a recipient can demonstrate that such criteria have been validated as essential to participation in a given program and that alternative equally valid criteria that do not have such a disproportionate adverse effect are unavailable, the criteria will be judged nondiscriminatory").

[14] Of course a defendant may rebut a plaintiff's showing of adverse racial impact by challenging the plaintiff's evidence or adducing countervailing proof of its own. Watson v. Fort Worth Bank and Trust, 487 U.S. 977, 996, 108 S. Ct. 2777, 2789, 101 L. Ed. 2d 827 (1988) (O'Connor, J., plurality opinion); Emanuel v. Marsh, 897 F.2d 1435, 1439 (8th Cir.1990). As the court describes below, it has considered both plaintiffs' and the Board's statistical evidence in resolving this issue.

[15] For these reasons, the Court in Wards Cove discounted evidence that a far greater number of non-whites were employed by the defendant employer, a fish cannery, in unskilled, cannery positions than in skilled non-cannery jobs. Noting that the cannery work force did not reflect the "pool of qualified jobs applicants" or the "qualified population in the work force" with respect to the non-cannery jobs, the court concluded that the plaintiffs' evidence did not exclude the possibility that "the absence of minorities holding such skilled positions is due to a dearth of qualified non-white applicants (for reasons that are not [the employer's] fault)," in which case the employer's "selection methods or employment practices cannot be said to have had a `disparate impact' on non-whites." 490 U.S. at 651, 109 S. Ct. at 2122.

[16] Thus the selection process for teacher-education programs distinguishes this from other testing cases in which below-passing performance on an exam is the only basis for rejecting the test-taker. Only in those cases are "actual examinees ... the most logical statistic pool, and ... the appropriate method of comparison to be applied to that pool ... a measurement of the differences in pass-fail rates." Richardson, 729 F. Supp. at 815. See also Nash, 905 F.2d at 358.

[17] Both plaintiffs and the Board acknowledge that data on those students who actually applied to teacher training programs and were accepted or rejected do not provide a valid basis to determine whether the ACT requirement has an adverse racial impact. Many if not most students who fail to meet this prerequisite undoubtedly refrain from the fruitless effort of seeking admission to a teacher training program. See Wards Cove, 490 U.S. at 643, 652 n. 7, 109 S. Ct. at 2117, 2122 n. 7; Int'l Brotherhood of Teamsters v. United States, 431 U.S. 324, 367, 97 S. Ct. 1843, 1871, 52 L. Ed. 2d 396 (1977); Powers v. Alabama Dep't of Educ., 854 F.2d 1285, 1298 (11th Cir.1988), cert. denied, 490 U.S. 1107, 109 S. Ct. 3158, 104 L. Ed. 2d 1021 (1989); Kilgo v. Bowman Transportation, 789 F.2d 859, 868 (11th Cir.1986).

Furthermore, neither side questions the various mathematical calculations performed by the other's expert. Accordingly, all of the figures on disparate impact that the court discusses in this section are ones presented by the parties' experts.

[18] It appears that this particular English exam is used only for the purpose of evaluating applicants to teacher education and is routinely taken by students in their sophomore year of college, subsequent to the ACT for which most sit in high school.

[19] Justice O'Connor's opinion in Watson cited language from Griggs v. Duke Power Co., 401 U.S. 424, 426, 91 S. Ct. 849, 851, 28 L. Ed. 2d 158 (1971) (examining "requirements [that] operate[d] to disqualify Negroes at a substantially higher rate than white applicants"); Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S. Ct. 2362, 2375, 45 L. Ed. 2d 280 (1975) (plaintiffs are required to 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"); Washington v. Davis, 426 U.S. at 246-47, 96 S. Ct. at 2050-51 ("hiring and promotion practices disqualifying substantially disproportionate number of blacks"); Dothard v. Rawlinson, 433 U.S. 321, 329, 97 S. Ct. 2720, 2726-27, 53 L. Ed. 2d 786 (1977) (employment standards that "select applicants for hire in a significantly discriminatory pattern"); and several other recent decisions. Watson, 487 U.S. at 995, 108 S. Ct. at 2789. See B. Schlei & P. Grossman, Employment Discrimination Law, at 98-99 (2d ed. 1983). See also, e.g., Bridgeport Guardians, Inc. v. City of Bridgeport, 933 F.2d 1140, 1146 (2d Cir.1991); Rose v. Wells Fargo & Co., 902 F.2d 1417, 1424 (9th Cir.1990); Emanuel, 897 F.2d at 1441; Green v. USX Corp., 896 F.2d 801, 805 (3rd Cir.), cert. denied, ___ U.S. ___, 111 S. Ct. 53, 112 L. Ed. 2d 29 (1990).

[20] Twenty-one persons in the sample were not classified by race.

[21] When these statistical tests are performed on the figures propounded by plaintiffs, which include those Alabama students who took the ACT between 1983 and 1988 and noted an intention to major in education, the following results are obtained: under the four-fifths test, the adverse impact of the ACT requirement is 45%; using the Hazelwood formula, the difference is 29.2 standard deviations; and applying the Shoben formula, the difference or "Z" value is 26.6 standard deviations.

[22] In Shoben, at 795-96, the author also argues that her approach is superior to the Hazelwood approach in "cases concerning the significance of differences in pass-fail rates in employment tests."

[23] One recent law review commentary explains that, in Wards Cove, the Supreme Court clarified the theoretical base of Title VII law such that a disparate-impact claim is viewed as the "functional equivalent" of a disparate-treatment claim. Leading Cases, 103 Harv.L.Rev. 137, 350-61 (1989). According to the commentary, the Court reshaped disparate-impact law in accordance with a theory of "equal treatment," which "seeks to guarantee fair process," rather than a "theory of equal achievement, which strives for fair results racial parity after years of discrimination." Id. at 356-57 (emphasis in original). In doing this, Wards Cove interprets the theory of equal treatment to embrace a "broader definition of impermissible criteria" than one limited to combatting facial or intentional race discrimination. Id. at 357. "Under this view, courts must prohibit not only the use of racial criteria, but also the use of criteria that are the functional equivalent of race" in other words, those practices that "produce[] a disparate impact on minorities" and which are not "legitimate that is, related to merit and productivity." Id. at 357-58 (emphasis in original). "[T]he `particularized wrong' to be remedied is not disparate impact alone, but the unfair treatment that accompanies the use of a racial criterion or its functional equivalent. Thus, the plaintiff's job is to show this functional equivalence to prove first that a challenged practice produces a disparate impact, and second, that it is unrelated to productivity because it serves no legitimate employment goal of the employer." Id. at 358-59. See also Watson, 487 U.S. at 987, 108 S. Ct. at 2785 (O'Connor, J., plurality opinion) ("the necessary premise of the disparate impact approach is that some employment practices, adopted without a deliberately discriminatory motive, may in operation be functionally equivalent to intentional discrimination").

[24] See EEOC v. J.M. Huber Corp., 927 F.2d 1322, 1329 n. 24 (5th Cir.1991) ("Under an impact theory, however, the employer must show more than some nondiscriminatory reason for the policy to successfully defend the charge of discrimination"). See also Nash, 905 F.2d at 358; Green, 896 F.2d at 801.

[25] If a plaintiff is unable to persuade the trier of fact on the issue of the business justification for an employer's challenged practice, he or she may still prevail by showing that "other tests or selection devices, without a similarly undesirable racial effect, would also serve the employer's legitimate hiring interests ... [thus] prov[ing] that [the employer was] using [the] test[] merely as a `pretext' for discrimination." Wards Cove, 490 U.S. at 660, 109 S. Ct. at 2126. See, e.g., United States v. City of Montgomery, 775 F. Supp. 1450 (M.D.Ala.1991) (Thompson, J.) (finding that plaintiffs' suggested alternative procedures for promotion to sergeant in city police department would not adequately serve city's legitimate promotion interests).

[26] Although there is no single method for appropriately validating employment tests, the Uniform Guidelines and applicable case law recognize three basic methods of validation. The first, "empirical" or "criterion" validity, is demonstrated by correlating test scores with objective measures or "criteria" of successful job performance. The second, "construct" validity, correlates test scores with certain identifiable characteristics of applicants that, in turn, are important to successful job performance. The third and final method measures "content" validity, by examining whether a test's content closely approximates tasks to be performed on the job. Washington v. Davis, 426 U.S. at 247 n. 13, 96 S. Ct. at 2051 n. 13. See 29 C.F.R. § 1607.5(B); Hamer v. City of Atlanta, 872 F.2d 1521, 1525 (11th Cir.1989); Richardson, 729 F. Supp. at 820-21.

[27] The court notes that the administrative regulations promulgated under Title VI itself also look to whether an educational admissions requirement serves a valid, non-discriminatory purpose. See 34 C.F.R. § 100, Appendix B, Part K ("if a recipient can demonstrate that ... criteria [that disproportionately exclude persons of a particular race] have been validated as essential to participation in a given program and that alternative equally valid criteria that do not have such a disproportionate adverse effect are unavailable, the criteria will be judged nondiscriminatory. Examples of admissions criteria that must meet this test are ... standardized tests").

[28] The Board has merely sought to show that the ACT measures academic achievement or aptitude in several disciplines, and that such achievement or aptitude is necessary for a teacher.

[29] As the court previously noted, the ACT requirement even fails to accomplish consistently this meaningless numerical goal. Because the test is nationally standardized, the percentage of Alabama students achieving a score of 16 or greater has fluctuated each year.

[30] It may be true that in a case in which there was a "manifest relationship" between a particular criterion and job performance, an employer or under Title VI, a recipient of federal funds could prevail even absent a formal "validation study." Watson, 487 U.S. at 997-99, 108 S. Ct. at 2790-91 (O'Connor, J., plurality opinion) (citing New York City Transit Auth. v. Beazer, 440 U.S. 568, 587 n. 31, 99 S. Ct. 1355, 1366 n. 31, 59 L. Ed. 2d 587 (1979), but noting that employer least likely to be able to produce evidence of such a "manifest relationship" in the context of a "standardized test"). Suffice it to say, however, that in a case such as this, in which the selection device was adopted for bizarre, irrational reasons and was not designed to measure job performance directly or indirectly, the absence of any supporting validation evidence by the Board is the death knell to their defense on the element of educational justification.

[31] A recent Third Circuit decision, noting that Wards Cove "did not reduce the defendant's burden [of production] to a showing of mere rationality," goes on to conclude that "[t]he employer retains the burden of producing significant evidence that establishes a strong factual showing of manifest relationship between the challenged practice and the defendant's employment goals. That burden, in our view, is met only when the employer is able to adduce some proof that the device serves identified legitimate and substantive goals.... The defendant, therefore, has some burden of presenting objective evidence showing a nexus between the selection device and a particular employment goal." Newark Branch, NAACP v. Harrison, New Jersey, 940 F.2d 792 (3rd Cir.1991)

[32] Indeed, the court is convinced that, in the minds of some members of the Poole Committee and the Board, the strongest "evidence" of the cut-off score's validity was the very data which indicated that a significant, disproportionate number of black students would prove unable to satisfy this requirement. As one commentator has observed,

It may appear `commonsensical' to a white employer that blacks do worse on tests, simply because `commonsensical' racist stereotyping supports the idea that blacks are typically inferior. The failure of the employer to attempt to validate the test systematically results from the fact that bigoted common sense implicitly validates the test's outcome. Perversely, if a text excludes blacks its results may even be thought to confirm the accuracy of the testing mechanism.

M. Kelman, Concepts of Discrimination in "General Ability" Job Testing, 104 Harv.L.Rev. 1158, 1168 (1991) (emphasis in original). The ready manner in which the Poole Committee adopted the requirement, with full knowledge of its probable adverse racial impact and without any reasonable effort at validation, supports this conclusion. The court is convinced, therefore, that the cut-off score was not only the "functional equivalent" of race, see note 23, supra, it was more likely than not an actual product of intentional discrimination.

Furthermore, in considering the challenged actions of the Board, undertaken in the mid-1970's, the court does not write on a clean slate. The court takes judicial notice of the long, not-so-distant history of government race discrimination against black educators and college students in Alabama and particularly the Board's efforts to employ teacher training programs and the teacher certification process as tools to enforce racial segregation in public education. See Lee v. Macon County Bd. of Educ., 267 F. Supp. 458, 473, 481 (M.D.Ala.) (three-judge court) (per curiam), aff'd, 389 U.S. 215, 88 S. Ct. 415, 19 L. Ed. 2d 422 (1967). Particularly to the extent that the ACT requirement may be said to "perpetuate the effects" of such past discrimination, it cannot "be defended simply on the basis of [its] facial neutrality or on the basis of the [Board's] lack of discriminatory intent." Watson, 487 U.S. at 988, 108 S. Ct. at 2785 (O'Connor, J., plurality opinion). Cf. Baker v. Columbus Municipal Separate Sch. Dist., 462 F.2d 1112, 1115 (5th Cir.1972); M. Rebell, Disparate Impact of Teacher Competency Testing of Minorities: Don't Blame the Test-Takers or the Tests, 4 Yale Law & Policy Rev. 375, 377 (1986).