Opinion
96 CV 8414 (KMW)
12-04-2012
OPINION & ORDER
WOOD, U.S.D.J.:
Plaintiffs, who represent a class of African-American and Latino teachers in the New York City public school system, brought the above-captioned action in 1996. Plaintiffs allege that the Board of Education of the City School District of the City of New York ("the Board"), currently a Defendant, and former Defendant the New York State Education Department ("SED") discriminated against Plaintiffs in violation of Title VII of the Civil Rights Act of 1964 ("Title VII"), as amended, 42 U.S.C. § 2000e et seq. Specifically, Plaintiffs claim the Board engaged in discrimination by requiring Plaintiffs to pass certain standardized tests—the National Teacher Core Battery exam ("Core Battery exam") and the Liberal Arts and Sciences Test ("LAST"), the successor to the Core Battery exam—in order to be licensed to teach in New York City public schools. In 2001, Judge Constance Baker Motley, before whom this case was originally pending, certified the class pursuant to Federal Rules of Civil Procedure 23(a) and 23(b)(2).
In 2003, after five month bench trial, Judge Motley entered judgment in favor of the Board and SED, finding that their use of the Core Battery exam and the LAST did not violate Title VII. In 2006, the Second Circuit Court of Appeals vacated the District Court's judgment with respect to the LAST, and remanded the case. The Second Circuit also dismissed all claims against SED, leaving the Board as the sole defendant. While the case was pending before the Court on remand, the Board moved to decertify the class in light of the Supreme Court's decision in Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011).
This Opinion (1) first considers the Board's decertification motion, and then addresses the three remaining post-remand issues pertaining to the Board's Title VII liability: (2) whether the Board can be subject to Title VII liability for its use of the LAST; (3) whether the Board violated Title VII by requiring Plaintiffs to pass the LAST in order to receive a teaching license; and (4) whether the Board violated Title VII by reducing Plaintiffs' salaries, benefits, and seniority if they failed to pass the Core Battery exam and the LAST.
For the reasons set forth below, the Court holds that: (1) the Board's decertification motion should be granted—and the class decertified—with respect to all of Plaintiffs' requests for relief except a declaratory judgment as to the Board's liability under Title VII and injunctive relief benefitting the class as whole; (2) the Board can be subject to Title VII liability for its use of the LAST; (3) the Board violated Title VII by requiring Plaintiffs to pass the LAST because it was not properly validated; and (4) the Board did not violate Title VII by reducing Plaintiffs' salaries, benefits, and seniority if they failed to pass the Core Battery exam.
I. BACKGROUND
A. Licensing of New York City Teachers
In order to teach in New York State's public school systems, teachers must be certified by the state. Trial Tr. 1668-69. SED, which supervises the state public school system, is responsible for state certification of teachers. Until 1991, the Board was responsible for setting licensing requirements for teachers in the New York City ("City") school system. Although the State and the City used different terminology ("certificate" versus "license"), both State certification and City licensing serve the same purpose: to ensure that new teachers met certain requirements, primarily with respect to education and experience, deemed necessary for successful teaching. In order to comply with state law, City licensing standards had to be "substantially equivalent" to state certification standards. Bd. Ex. N. This meant that the Board had to adopt any certification requirements imposed by SED. Trial Tr. 230-33. SED reviewed and approved the licensing requirements set by the Board to ensure equivalence. Bd. Ex. N; Trial Tr. 243-44, 1005, 1725.
City teachers could receive conditional teaching licenses if they passed a licensing exam and met certain minimal requirements. A conditional license lasted for five years and became a permanent license if the teacher met additional licensing requirements within those five years. Trial Tr. 230, 238, 868; Pls. Ex. 1. A teacher with a conditional license could teach full time in a City public school; after a probation period, the teacher received tenure, with a higher salary and more generous benefits and seniority rights. Pls. Ex. 1. If, however, the teacher failed to meet the requirements for a permanent license within five years, her conditional license could be revoked. Pls. Ex. 1. Once the conditional license was revoked, the teacher could teach only as a substitute; substitutes were permitted to meet less stringent licensing requirements. Pls. Ex. 1. Because of teacher shortages, many substitute teachers worked full time; however, substitute teachers had lower salaries, fewer benefits, and no seniority rights. SED Ex. 58(c).
A teacher could obtain a conditional license through two different methods. First, a teacher could pass the "open examination," a series of exams and interviews intended to assess a teacher's knowledge of relevant subjects. Alternatively, a teacher could begin teaching immediately as a substitute with a temporary certificate. The requirements for obtaining a temporary certificate were minimal, but the teacher received a lower salary and fewer benefits than a teacher with a conditional license and the certificate had to be renewed on a yearly basis. After teaching for two years, a teacher holding a temporary certificate could obtain a conditional license by passing the "closed examination," which was essentially just an interview. See Gulino v. Bd. of Educ. of the City Sch. Dist. of the City of N.Y., No. 96 Civ. 8414, 2003 WL 25764041, at *-2-3 ¶¶ 2-11 (S.D.N.Y. Sept. 4, 2003) (Motley, J.) ("Gulino III").
In 1984, SED issued a new regulation that required teachers to pass the Core Battery exam in order to receive state certification. SED Ex. 17. The Core Battery exam was a set of standardized tests that measured teachers' communication skills; general knowledge of social studies, math, science, literature, and the fine arts; and knowledge of pedagogy. Gulino v. Bd. of Educ. of the City Sch. Dist. of the City of N.Y., No. 96 Civ. 8414, 2003 WL 25764041, at *13 ¶ 69 (S.D.N.Y. Sept. 4, 2003) (Motley, J.) ("Gulino III").
Around the time that SED introduced the Core Battery exam, SED informed the Board that, in order for City licensing standards to be equivalent to state standards, City teachers must also pass the Core Battery exam. Trial Tr. 1006-07, 2349, 3483-84. SED and the Board agreed, however, that the Board could phase in the Core Battery exam requirement gradually: beginning in 1985, City teachers would be required to pass the Core Battery exam in order to receive a permanent license; eventually, the Board would make the Core Battery exam a requirement to obtain a conditional license. Pls. Ex. 30; 31. City teachers who received permanent licenses prior to 1985 were not required to take the exam. SED Ex. 17.
Under this plan, the Board continued to issue conditional licenses to teachers who had not passed the Core Battery exam, and, because of teacher shortages and administrative problems, the Board gave many teachers with conditional licenses more than five years to pass the Core Battery exam. Pls. Exs. 1, 273. Many teachers, therefore, were able to continue teaching in full-time, non-substitute positions with a conditional license, even though they did not satisfy the permanent licensing requirements (including passing the Core Battery exam) within five years.
In 1991, the New York State legislature passed a new law standardizing licensing requirements across the state, including a mandate that all New York teachers—including City teachers—obtain state certification. SED Ex. 58(c). Pursuant to this law, the Board could not issue a conditional license until the teacher obtained state certification, which required passing the Core Battery exam. Trial Tr. at 925. Teachers who had not passed the Core Battery exam could still teach in full-time, non-substitute positions by obtaining a state temporary license, Trial Tr. 925-26, which could be renewed yearly for up to three years, or longer in certain circumstances. Trial Tr. 889. Even after the 1991 law came into force, teachers who obtained conditional licenses prior to 1991 could continue teaching with those licenses, subject, as before, to the condition that they pass the Core Battery exam within five years. Trial Tr. 242-44.
To ensure compliance with the 1991 law, SED pressured the Board to enforce the Core Battery exam requirement and to revoke the conditional and temporary licenses of teachers who had not passed the exam within five years. Trial Tr. 979-80, 2350. Accordingly, after 1991, the Board began informing delinquent teachers that their conditional or temporary licenses would be revoked if they did not pass the Core Battery exam by a certain date. Many teachers who were unable to pass the Core Battery exam subsequently lost their conditional and temporary licenses, and could work only as substitutes. Due to continuing teacher shortages, however, many of these substitutes remained in the same classrooms and continued to teach full time, but at a lower salary, with a reduced benefit level, and without seniority.
Beginning in 1993, SED phased out the Core Battery exam and phased in the LAST, a new test developed by a professional test development company, National Evaluation Systems ("NES"). Trial Tr. at 2316. It was one of several new requirements to obtain a permanent teaching license, which also included requiring teachers to obtain a master's degree, gain two years of teaching experience, and pass content-specific tests. Trial Tr. At 1710-2315. From that point forward, most new teachers were required to pass the LAST in order to obtain state certification or a conditional license, and teachers with a conditional license had to pass either the Core Battery exam or the LAST in order to receive a permanent license. Trial Tr. at 2316. By 1996, SED had completely eliminated the Core Battery exam, and all teachers had to pass the LAST in order to be licensed to teach in New York.
For more detailed background on the development and use of the LAST, see infra Part V.A.
B. Procedural History
Plaintiffs represent a class of African-American and Latino teachers who were teaching in City public schools with temporary or conditional licenses ("experienced teachers"), but were unable to obtain permanent licenses or had their licenses revoked because they could not pass the Core Battery exam or the LAST. Gulino v. Bd. of Educ. of the City Sch. Dist. of the City of N.Y., 201 F.R.D. 326, 330 (S.D.N.Y. 2001) (Motley, J.) ("Gulino I"). Plaintiffs brought this action against SED and the Board in 1996, pursuant to Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq. Title VII prohibits an employer from requiring job applicants to pass an employment exam that: (1) has a disparate impact on a protected class, and (2) is not "job related," meaning that the exam does not have a "manifest relationship to the employment in question." See 42 U.S.C. 2000e-2(a)(1); Albemarle Paper Co. v. Moody, 422 U.S. 405, 425 (1975). Plaintiffs alleged that: (1) the LAST and Core Battery exam had a disparate impact because Caucasian test-takers passed the exams at statistically significantly higher rates than African-American and Latino test-takers; and (2) the exams were not job related because they did not measure whether experienced teachers such as Plaintiffs were qualified to teach. Accordingly, Plaintiffs claimed that the Board and SED violated Title VII by requiring Plaintiffs to pass the Core Battery exam and the LAST in order to receive permanent licenses, and by revoking Plaintiffs' temporary or conditional licenses and reducing their salaries, benefits, and seniority for failing to pass. Plaintiffs sought three types of relief: monetary relief in the form of backpay; a declaratory judgment as to Defendants' liability; and injunctive relief, including the appointment of a monitor to ensure that the current version of the LAST did not violate Title VII, and an award of licenses and seniority rights to teachers who were denied them because of their performance on the LAST.
1. Initial District Court Proceedings
This case was originally assigned to the Honorable Constance Baker Motley. In 2002, after extensive discovery, Defendants moved for summary judgment. SED argued that it was not subject to Title VII because it was not Plaintiffs' employer, and the Board argued that it could not be held liable under Title VII because it was following the mandates of state law when it required teachers to pass the Core Battery exam and the LAST in order to receive a license. Judge Motley denied both of these motions, and held that SED was an employer for the purposes of Title VII. She then held that the Board could be liable under Title VII, regardless of whether it was following state law, because (1) the Board, not SED, had decertified teachers who failed to pass the Core Battery exam and the LAST; and (2) Title VII preempts any state laws in conflict with it. See Gulino v. Bd. of Ed. of City Sch. Dist. of N. Y., 236 F. Supp. 2d 314, 332-37 (S.D.N.Y. 2002) (Motley, J.) ("Gulino II").
The case proceeded to "an epic bench trial that lasted more than eight weeks and filled over 3,600 pages of trial transcript." Gulino III, 2003 WL 25764041, at *1. In 2003, following the trial, Judge Motley ruled that SED and the Board had not violated Title VII by requiring teachers to pass the Core Battery exam or the LAST in order to receive a permanent license. Id. at *30-31 ¶¶ 161-64. Although Judge Motley held that Plaintiffs had established a prima facie case of disparate impact, id. at *30 ¶ 160, she ultimately found that the Board and SED were not liable under Title VII because the evidence proved that both exams were job related, a defense to Plaintiffs' showing of disparate impact. Id. at *30-31 ¶¶ 161-63 (citing Albemarle Paper, 422 U.S. at 425).
Judge Motley held that two alternative standards existed to determine whether an employment exam is job related. Under the first standard, an exam is job related if the exam is properly "validated." Gulino III, 2003 WL 25764041, at *30 ¶ 161. Validation requires a showing "by professionally acceptable methods, [that the exam is] 'predictive of or significantly correlated with important elements of work behavior which comprise or are relevant to the job or jobs for which candidates are being evaluated.'" Albemarle Paper, 422 U.S. at 431 (quoting 29 C.F.R. § 1607.4(c)). In the Second Circuit, validation is assessed using a five-part test established in Guardians Ass'n v. Civil Service Commission of New York, 630 F.2d 79 (2d Cir. 1980). Judge Motley drew the second standard from the Supreme Court's decision in Watson v. Fort Worth Bank & Trust, 487 U.S. 977 (1988), and found that an employment exam is job related if it is "manifestly related to legitimate employment goals." Gulino III, 2003 WL 25764041, at *30 ¶ 162.
Applying the Guardians test, Judge Motley held that Defendants had proven that the Core Battery exam was properly validated, and thus was job related. Id. at *30 ¶ 161. However, Judge Motley held that the LAST was not properly validated under the Guardians standard; there was insufficient documentary evidence in the record regarding validation of the LAST; and, because of this "pervasive lack of documentation," Defendants had not met their burden of proving validation. Id. at *29 ¶ 153. However, applying the alternative Watson standard, Judge Motley held that the LAST was "manifestly related to legitimate employment goals" and thus was job related. Id. at *31 ¶ 163.
2. Second Circuit Proceedings
The parties cross-appealed Judge Motley's rulings. Defendants appealed Judge Motley's decision that they could be subject to Title VII liability. SED renewed its argument that it was not an employer for the purposes of Title VII. Gulino v. N.Y. State Educ. Dep't, 460 F.3d 361, 372 (2d Cir. 2006) ("Gulino IV"). The Board, however, abandoned its prior argument that it could not be held liable under Title VII because it was following state law when it required teachers to pass the LAST. Id. at 380. Instead, the Board contended that it could not be subject to Title VII liability because it was acting as a licensor, not an employer, when it licensed Plaintiffs to teach in City schools. Because Title VII does not apply to licensors, the Board argued, it was not subject to liability. Id.
Plaintiffs appealed Judge Motley's ruling that the LAST was job related. Id. at 381. Plaintiffs also argued that Judge Motley had failed to address their argument that the Board "misused" the Core Battery exam and the LAST by requiring experienced teachers like Plaintiffs to take the exams, and demoting them to substitute positions if they failed. Pls. Ex. H at 57-61; Gulino IV, 460 F.3d at 370. Plaintiffs argued that, to the extent the Core Battery exam and the LAST were properly validated, they were validated for use only in licensing new teachers who had not yet entered a classroom. Plaintiffs argued that the Board should not have required experienced teachers to pass the Core Battery exam and the LAST, and that this "misuse" of the exams violated Title VII. Id.
Plaintiffs did not appeal Judge Motley's opinion as to the validity of the Core Battery exam, so the Second Circuit did not address it and it is not at issue on remand.
The Second Circuit issued its opinion in 2006. First, the Second Circuit dismissed all claims against SED, finding that it was not an "employer" within the meaning of Title VII. Gulino IV, 460 F.3d at 379-80. With respect to the Board, however, the Second Circuit decided that Title VII did apply because the Board was both a licensor and Plaintiffs' employer, and employers are subject to Title VII. Id. at 380-81. Although the Board had not renewed its argument that it could not be liable because it was following state law, the Second Circuit nonetheless found that Judge Motley was correct to reject this argument on summary judgment because Title VII preempts conflicting state laws. Id. at 380.
Finally, the Second Circuit held that Judge Motley had erred in finding that the LAST was job related. The Second Circuit disagreed with Judge Motley's reading of Watson, and held that the only standard for determining whether an employment exam is job related is whether the test was validated, assessed using the five-part Guardians test. Gulino IV, 460 F.3d at 385-86. The Second Circuit also found several factual errors in Judge Motley's analysis, including her focus on the essay portion of the test. Id. at 387. Finally, the Second Circuit questioned Judge Motley's finding that there was "a pervasive lack of documentation" in the case, and noted that a defendant need not present documentary evidence of the validation process in order to prove that an exam was properly validated; rather, testimony from those involved in the validation process and "studied opinions of certified experts" could also prove that the LAST had been properly validated. Id. at 387-88. Accordingly, the Second Circuit remanded the case to determine in the first instance whether the LAST was job-related under the Guardians test using these evidentiary guideposts. Id. at 388. The Second Circuit also directed the District Court to address Plaintiffs' "misuse" argument. Id. at 370 n.9.
3. Current Proceedings
On remand, the case was reassigned to Judge Sydney H. Stein, and then transferred to this Court in February 2009. In December 2009, the Court ordered the parties to brief the two issues remaining on remand: (1) whether the LAST was properly validated, and thus job related; and (2) whether the Board misused the LAST and the Core Battery exam to make decisions regarding the conditions of Plaintiffs' employment. See Order dated Dec. 8, 2009 (Dkt. 241). Along with their briefs, the parties submitted all relevant trial evidence for the Court's review. (Dkt. 247; Dkt. 276; Dkt. 253). SED, although no longer a party to the case, submitted an amicus brief arguing that the LAST was properly validated. Remand Mem. by Amicus Curiae N.Y. State Educ. Dep't (Dkt. 251) ("SED Remand Mem."); see also Order dated Sept. 17, 2009 (Dkt. 237) (denying SED's motion to intervene but allowing it to participate as amicus curiae).
In August 2010, the Court requested additional briefing on the Board's involvement in developing the LAST; what information the Board had about the development and administration of the LAST; and what action the Board could have taken if it suspected the LAST violated Title VII. See Order dated Aug. 13, 2010 (Dkt. 294). Plaintiffs, the Board, and SED (as amicus) all submitted supplemental briefs addressing these issues. (Dkt. 299; Dkt. 302; Dkt. 304). The parties also addressed what effect, if any, the Supreme Court's holding in Ricci v. DeStefano, 557 U.S. 557 (2011), had on the pending issues.
In these briefs, the Board and SED argue that (1) the LAST is not subject to challenge under Title VII because it is a licensing test, not an employment test; and (2) the Board was mandated by state law to use the LAST, and following state law is a "business necessity" that exempts the Board from Title VII liability. The Court notes that it did not ask the Board and SED to brief these issues on remand. However, because the Court understands the importance of these issues, it addresses them in this Opinion.
In July 2011, while the issues on remand were still pending, the Board moved to decertify the class based on the Supreme Court's decision in Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011).
For the reasons set forth below, the Court finds that: (1) the Board's decertification motion should be granted—and the class decertified—with respect to all of Plaintiffs' requests for relief except a declaratory judgment as to the Board's liability under Title VII and injunctive relief benefitting the class as whole; (2) the Second Circuit has established that the LAST may be challenged under Title VII and that the Board is subject to Title VII liability; (3) the Board violated Title VII because the LAST was not properly validated; and (4) the Board did not misuse the Core Battery exam to make decisions regarding the conditions of Plaintiffs' employment. Based on these findings, the Court holds that the Board violated Title VII by requiring Plaintiffs to pass the LAST in order to receive a permanent license.
II. CLASS CERTIFICATION ISSUE
Before considering the merits of Plaintiffs' Title VII claims, the Court must first address the Board's motion to decertify Plaintiffs' class in light of Wal-Mart Stores, Inc. v. Dukes, 131 S. Ct. 2541 (2011) ("Wal-Mart").For the following reasons, the Court grants in part and denies in part the Board's motion to decertify the class.
A. Class Certification Order
In 2001, pursuant to Rule 23, Plaintiffs sought to certify the class, defined as "All African-American and Latino individuals employed as New York City public school teachers by Defendants, on or after June 29, 1995, who failed to achieve a qualifying score on either the [Core Battery exam] or the LAST, and as a result either lost or were denied a permanent teaching appointment." Gulino I, 201 F.R.D. at 330. Rule 23 stipulates that a class action may be maintained only if the action satisfies the requirements of Rule 23(a) and at least one subsection of Rule 23(b). Judge Motley certified the class pursuant to Rules 23(a) and (b)(2).
Judge Motley held that the four requirements of Rule 23(a)—numerosity, commonality, typicality, and adequacy of representation—were satisfied because, respectively: (1) the approximately 2,000-member class was numerous enough to render joinder impractical; (2) the class members had in common the legal question "at the heart of this suit"—that is, whether Defendants' use of the Core Battery exam and the LAST had a disparate impact on African-American and Latino teachers; (3) Plaintiffs' claims were typical of the other class members, especially in light of the fact that all members would necessarily benefit from any injunctive relief requiring Defendants to cease using the LAST; and (4) the interests of Plaintiffs and the class members were sufficiently aligned, and class counsel was adequately qualified and experienced to conduct the representation, such that Plaintiffs fairly and adequately protected the interests of the class. Gulino I, 201 F.R.D. at 331-33.
Judge Motley then determined that the class should be certified under Rule 23(b)(2), which is satisfied where "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." Fed. R. Civ. P. 23(b)(2). Judge Motley held that Rule 23(b)(2) was invoked appropriately because Plaintiffs sought to enjoin Defendants from using the LAST to engage in an alleged pattern of discriminatory conduct that caused injury to the entire class. Gulino I, 201 F.R.D. at 333. Judge Motley further noted that "[w]here injunctive relief and damages are sought under Rule 23(b)(2), the injunctive relief must be the predominant issue," id., and found that, because Plaintiffs had dropped their claims for compensatory and punitive damages, their "remaining claims for monetary relief [in the form of backpay] do not predominate over [their] claims for injunctive and declaratory relief." Id. at 334.
The Board, joined by SED as amicus curiae, now moves to decertify the class based on Wal-Mart, arguing that the Supreme Court's holding that individualized claims may not be certified under Rule 23(b)(2) precludes Plaintiffs from maintaining their class certification under that provision. Def. Letter dated July 8, 2011, at 2 (Dkt. 306).
B. Legal Standard
Pursuant to Rule 23(c)(1)(C), "[a]n order that grants or denies class certification may be altered or amended before final judgment." Fed. R. Civ. P. 23(c)(1)(C); see also Boucher v. Syracuse Univ., 164 F.3d 113, 118 (2d Cir. 1999) ("[U]nder Rule 23(c)(1), courts are required to reassess their class rulings as the case develops." (internal quotation omitted)). A court is permitted to "decertify a class if it appears that the requirements of Rule 23 are not in fact met." Sirota v. Solitron Devices, Inc., 673 F.2d 566, 572 (2d Cir. 1982). However, a court "may not disturb its prior findings absent some significant intervening event or a showing of compelling reasons to reexamine the question." Doe v. Karadzic, 192 F.R.D. 133, 136-37 (S.D.N.Y. 2000) (Leisure, J.) (internal quotation and citations omitted). Compelling reasons "include an intervening change of controlling law, the availability of new evidence, or the need to correct a clear error or prevent manifest injustice." Id. at 137 (internal quotation omitted).
Decertification is an "extreme step," particularly at a late stage in the litigation, "where a potentially proper class exists and can easily be created." Woe v. Cuomo, 729 F.2d 96, 107 (2d Cir. 1984) (internal quotation omitted). A defendant seeking to decertify a class "bear[s] a heavy burden to prove the necessity of either the drastic step of decertification or the less draconian but still serious step of limiting the scope of the class." Gordon v. Hunt, 117 F.R.D. 58, 61 (S.D.N.Y. 1987) (Lasker, J.).
C. Analysis
Defendants contend that the Wal-Mart decision requires decertifying the Plaintiff class because Plaintiffs seek monetary damages and individualized injunctive relief, both of which the Supreme Court deemed inappropriate for certification under Rule 23(b)(2). Although the Court agrees that Plaintiffs' claims for individualized injunctive relief and backpay must be decertified in light of Wal-Mart, the Court finds that the class survives as to Plaintiffs' claims for classwide relief, including a declaratory judgment as to Defendant's liability, and injunctive relief benefitting the class as whole.
1. Class Certification Under Rules 23(b)(2) and 23(c)(4)
Rule 23(b)(2) authorizes class certification when "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." Fed. R. Civ. P. 23(b)(2). The history of the Rule indicates that it was designed to address "civil rights cases against parties charged with unlawful, class-based discrimination," where "a single injunction or declaratory judgment would provide relief to each member of the class." Wal-Mart, 131 S. Ct. at 2557.
In Wal-Mart, the Supreme Court held that claims for non-incidental monetary relief, such as backpay, are not appropriate for certification under Rule 23(b)(2). Id. While certification under Rule 23(b)(2) is appropriate where the plaintiffs seek "an indivisible injunction benefitting all its members at once," id. at 2258, it is inappropriate where each class member "would be entitled to a different injunction or declaratory judgment against the defendant." Id. at 2256. In order to obtain individualized relief, a putative class must satisfy the requirements of Rule 23(b)(3), which includes greater procedural protections, such as notice and opportunity for members to opt out of the litigation. Fed. R. Civ. P. 23(c)(2)(B). Because Rule 23(b)(2) lacks the procedural safeguards necessary to ensure that all class members' due process rights are protected, the Supreme Court concluded that it cannot be used to certify class actions requiring individual determinations and awards of individualized relief. Wal-Mart, 131 S. Ct. at 2558-59.
The Court left open the question of whether claims for "incidental" monetary relief—or relief "flow[ing] directly" from injunctive or declaratory claims properly certified under Rule 23(b)(2) which "should not require additional hearings to resolve"—could still be certified under Rule 23(b)(2). Wal-Mart, 131 S. Ct. at 2560 (quoting Allison v. Citgo Petroleum Corp., 151 F.3d 402, 415 (5th Cir. 1998)). Plaintiffs have not argued that the monetary relief they seek qualifies for such treatment, and, given that their claims would require individualized hearings to adjudicate, could not succeed on such a theory.
In so holding, the Supreme Court overruled the Second Circuit's test for certification under Rule 23(b)(2), which had required only that classwide injunctive or declaratory relief predominate over individual relief. See Wal-Mart, 131 S. Ct. at 2559; see also Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 164 (2d Cir. 2001) (establishing the "predominance" test for certifying damages under (b)(2)). After Wal-Mart, simple predominance is insufficient; only claims for injunctive or declaratory relief that benefit the class as a whole may be certified under (b)(2).
The Wal-Mart Court did not, however, address certification under Rule 23(c)(4)(A), which provides that "an action may be brought or maintained as a class action with respect to particular issues." Fed. R. Civ. P. 23(c)(4)(A). In the Second Circuit, district courts are directed to "take full advantage of this provision to certify separate issues," Robinson, 267 F.3d at 167, and to "certify those portions of a claim that satisfy (b)(2) even if the claim as a whole does not." United States v. City of N.Y., 276 F.R.D. 22, 33 (E.D.N.Y. 2011) ("Vulcan III"). District courts frequently use Rule 23(c)(4) to bifurcate proceedings by first certifying an "injunctive" class under (c)(4) to determine liability, and then certifying a "remedial" class under (b)(3) to determine damages. See, e.g., Chen-Oster v. Goldman, Sachs & Co., 10 Civ. 6950, 2012 WL 205875, at *7-8 (S.D.N.Y. Jan. 19, 2012) (Francis, J.) (using (c)(4) and (b)(2) to certify a class "at the liability stage only"); Jermyn v. Best Buy Stores, L.P., No. 08-CV-214, 2011 WL 4336664, at *8 (S.D.N.Y. Sept. 15, 2011) (McMahon, J.) (certifying liability, but not monetary claims, under Rule 23(b)(2) in an action where both were brought); Vulcan III, 276 F.R.D. at 33. Allowing certification of a class for liability purposes, separate from remedies, "reduce[s] the range of issues in dispute and promote[s] judicial economy," and, in the event that defendants succeed at the liability stage, eliminates the need for a remedial stage inquiry altogether. Robinson, 267 F.3d at 168.
Further, using issue certification under (c)(4) to bifurcate class certification into liability and remedial phases, is "fully consistent with Wal-Mart[]." Vulcan III, 276 F.R.D. at 34; see also Janes v. Triborough Bridge & Tunnel Auth., No. 06-Civ-1427, 2011 U.S. Dist. LEXIS 115831, at *6 (S.D.N.Y. Oct. 4, 2011) (Engelmayer, J.) (finding that Wal-Mart had left issue certification under (c)(4) "intact"). Indeed, in liability determinations under Title VII, "the class seeks an indivisible declaration" that affects the class as a whole, precisely the type of "indivisible" relief the Wal-Mart Court decided fit squarely within Rule 23(b)(2). Vulcan III, 276 F.R.D. at 35; see also Wal-Mart, 131 S. Ct. at 2558.
2. Application to Plaintiff's Claims
Plaintiffs seek three types of relief: declaratory, monetary, and injunctive. Pls.'s Letter, dated Oct. 14, 2011 (Dkt. 320) ("Pls.'s Letter"). The Court will address each in turn.
a. Declaratory Relief
Plaintiffs seek declaratory relief as to Defendant's liability—specifically, a finding that the Board violated Title VII by requiring teachers to pass the LAST in order to obtain teaching licenses, and a finding that the Board misused scores from the Core Battery exam and the LAST to rescind teaching licenses. Pls.'s Letter at 4. Questions of liability for acts that affected a group are inherently classwide, and are commonly certified under Rules 23(b)(2) and (c)(4) for resolution at the liability phase of a trial. See Advisory Committee Note to Rule 23(c)(4) ("[T]he action may retain its 'class' character only through the adjudication of liability to the class; the members of the class may thereafter be required to come in individually and prove the amounts of their respective claims."); see, e.g., Robinson, 267 F.3d at 167-69 (encouraging certification of liability under (c)(4)(A) and (b)(2)); In re Nassau Cnty. Strip Search Cases, 461 F.3d 219, 227 (2d Cir. 2004) (approving use of (c)(4) to single out issues of liability even where action as a whole does not satisfy (b)(3)). Moreover, certifying a class claiming that Defendant is liable for having used the LAST comports with the reasoning of Wal-Mart in that resolution of that claim will not involve individual determinations, but rather involves a claim common to the class, the resolution of which will streamline later proceedings for damages and other individual relief. See Vulcan III, 276 F.R.D. at 34-35 (explaining that, if a class action is bifurcated, individual issues arise only if liability is proven, and noting that this is "fully consistent" with Wal-Mart); Chen-Oster, 2012 WL 205875, at * 8 (noting that "disparate treatment cases are especially appropriate for bifurcation" because "individualized issues arise only if the class establishe[s] the employer's liability").
The Court thus denies Defendant's motion for decertification as to Plaintiffs' request for declaratory relief.
b. Monetary Damages
Second, Plaintiffs request monetary relief in the form of backpay, on both a classwide and an individual basis. Pls.'s Letter at 7. After Wal-Mart, this type of relief unquestionably may be certified only under Rule 23(b)(3), and is inappropriate for certification under (b)(2). Wal-Mart, 131 S. Ct. at 2558 (concluding that "individualized monetary claims belong in Rule 23(b)(3)"). Plaintiffs thus cannot maintain their class certification under Rule 23(b)(2) as to their claim for monetary damages.
However, Plaintiffs have proposed bifurcating the proceeding into two phases: a liability phase certified under Rule 23(b)(2) and (c)(4), and a remedial phase in which Plaintiffs will seek class certification under Rule 23(b)(3). Pls.'s Letter at 7. The Second Circuit has encouraged such bifurcation to promote judicial flexibility in managing complex class actions. E.g. Robinson, 267 F.3d at 167 (urging district courts to take "full advantage" of Rule 23(c)(4)(A) to certify issues separately and "reduce the range of disputed issues in complex litigation and achieve judicial efficiencies" (internal quotations omitted)). This mandate expressly includes bifurcating class actions with respect to liability and damages. See id. at 167-69 (finding that the district court abused its discretion when it failed to certify class only as to liability); Nassau Cnty., 461 F.3d at 226 (noting that the advisory committee notes to Rule 23(c)(4)(A) support separate adjudication of liability and damages).
The reasons for bifurcating liability from remedial issues apply with no less force now than they did prior to Wal-Mart, particularly since individual issues will arise "only if the class establishe[s] the employer's liability." Vulcan III, 276 F.R.D. at 34; see also McReynolds v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 672 F.3d 482, 490-91 (7th Cir. 2012) (explaining that judicial efficiency favors resolving certifying a class on the separate issue of liability under (c)(4)(A)); Chen-Oster, 2012 WL 205875, at *6 (same). Given this framework, the Court grants Defendant's motion to decertify the Plaintiff class to the extent Plaintiffs seek backpay under Rule 23(b)(2). Because the Court in this decision concludes the liability phase of the proceedings, and determines that the Board did in fact violate Title VII, Plaintiffs should address potential Rule 23(b)(3) certification in their submission concerning remedies. See infra Part VII.
c. Injunctive Relief
Finally, Plaintiffs seek three types of injunctive relief: (1) an injunction providing teaching certificates to class members who wish to be considered for permanent teaching positions in New York City public schools; (2) an injunction affording seniority rights and other non-monetary benefits to class members still teaching in New York, that they would have received had they not failed the allegedly discriminatory examination; and (3) the appointment of a monitor to ensure that Defendant's current testing and licensing procedures do not violate Title VII. Pls.'s Letter at 5-6.
The Supreme Court's holding in Wal-Mart requires decertification of the class as to the first and second injunctive remedies requested. Certification pursuant to Rule 23(b)(2) is not appropriate "when each member of the class would be entitled to a different injunction or declaratory judgment against the defendant." Wal-Mart, 131 S. Ct. at 2557. In holding that individualized claims for backpay were not cognizable as a class action under Rule 23(b)(2), the Supreme Court noted that, under Title VII, Wal-Mart would be entitled to "individualized determinations of each employee's eligibility for backpay," and would have the opportunity to show that "it took an adverse employment action against an employee for any reason other than discrimination." Id. at 2560-61.
Although Plaintiffs characterize these requested injunctions as classwide, the injunctions they seek—including the provision of teaching certificates and seniority rights—are precisely the type of individualized relief the Supreme Court found to be outside the ambit of class certification under (b)(2). Under Title VII, once an employee shows that a particular test has a disparate impact, the burden shifts to the employer to show that it had legitimate, job-related reasons for denying a particular individual the benefits claimed. See 42 U.S.C. § 2000e-2(k)(1)(A)(i); see also Gulino IV, 460 F.3d at 382. Here, just as in Wal-Mart, the Board should have the opportunity to rebut individual plaintiff's claims for seniority rights and teaching licenses by presenting legitimate, job-related reasons why a particular individual was not promoted or did not receive a teaching license.
The Court did not address whether some individualized injunctive relief could proceed under (b)(2) as "incidental" to a classwide injunction. Cf. Wal-Mart, 131 S. Ct. at 2560 (noting that the holding did not address whether incidental monetary relief could be consistent with Rule 23(b)(2)). Although Plaintiffs have not argued that the individual injunctions requested here could qualify as incidental, the Court notes that the relief requested is more than incidental. Rather, the requested injunctions require individualized determinations of each class member's eligibility for seniority, a license, or other non-monetary benefits, with respect to which Defendant should be able to present evidence regarding each member's qualifications or other legitimate, job-related reasons to deny benefits. See 42 U.S.C. § 2000e-5(g). This type of relief closely resembles the backpay held inappropriate for (b)(2) certification in Wal-Mart, and thus cannot be considered "incidental" to a broader injunction.
However, the Plaintiffs' third request for relief—asking for a monitor to ensure that the current version of the LAST excises the discriminatory portions of the old version—is "an indivisible injunction benefitting all [the class] members at once." Wal-Mart, 131 S. Ct. at 2558. The Board argues that "there is nothing in the record to support a conclusion that the version of the LAST that is currently in use by the State of New York has an unlawful disparate impact on minority test takers." Def.'s Letter, dated Oct. 28, 2011, at 3 (Dkt. 316). The Board also contends that this relief is "inappropriate" because the Board has no role in the development or oversight of the LAST, but rather accepts certification exams developed by the SED. Id.
The Court disagrees. First, the Board's argument that it cannot be liable under Title VII because it played no role in developing the LAST, and just accepted the version of the test required by SED, has already been rejected by the Second Circuit. Gulino IV, 460 F.3d at 380 ("[T]he mandates of state law are no defense to Title VII liability."). If the LAST had a discriminatory impact in violation of Title VII, "Title VII explicitly relieves [the Board] from any duty to observe a state hiring provision which purports to require or permit any discriminatory employment practice." Id. (quoting Guardians, 630 F.2d at 105).
Second, Title VII provides the Court with "broad equitable powers" in order to provide victims of employment with "complete relief." Loeffler v. Frank, 486 U.S. 549, 558 n.6 (1988); United States v. Brennan, 650 F.3d 65, 90 (2d Cir. 2011); see also Albemarle Paper Co., 422 U.S. at 419-21 (discussing generally Title VII's remedial provisions). This discretion includes reinstatement, backpay, "or any other equitable relief as the court deems appropriate." 42 U.S.C. § 2000e-5(g)(1). The Court finds that Plaintiffs' request for a monitor to evaluate whether the current version of the LAST excises the portions of previous version that violate Title VII is necessary to provide Plaintiffs with "complete relief."
Finally, even if the Board is correct that the current version of the LAST is not discriminatory, voluntary cessation of illegal conduct moots an equitable claim only if the defendant can show that "(1) there is no reasonable expectation that the alleged violation will recur and (2) interim relief or events have completely and irrevocably eradicated the effects of the alleged violation." Seidemann v. Bowen, 499 F.3d 119, 128 (2d Cir. 2007); see also Easterling v. Connecticut, 265 F.R.D. 45, 53-54 (2d Cir. 2010) (applying Bowen to Title VII disparate impact claim and finding that defendants had failed to make the requisite showing). Defendants have not made the required showing in this case.
In short, the arguments the Board now offers to avoid the appointment of a monitor are identical to those the Second Circuit rejected in Gulino VI, and the Court rejects them. The Plaintiff class survives as to its request for a monitor to assess whether the current version of the LAST incorporates any of the invalid provisions of the version of the LAST used from 1996 to 2000 (the "Old LAST"). Although the Court previously held that evidence related to newer versions of the LAST is not admissible to prove Defendant's liability with respect to the Old LAST, see December 8, 2009 Order at 3-4 (Dkt. 243), the Court finds that evidence relating to the current version of the LAST and its development are admissible insofar as they relate to crafting an appropriate remedy.
D. Conclusion
For the foregoing reasons, the Defendant's motion to decertify the class is granted in part and denied in part. The class action will proceed as to Plaintiffs' request for declaratory relief, and as to Plaintiffs' request for a monitor to ensure the Board's current testing and licensing procedures do not violate Title VII. The remainder of this Opinion addresses the Board's liability under Title VII; Plaintiffs are directed to address specifics of their requested injunctive relief in a submission to the Court concerning remedies. See infra Part VII.
III. OTHER PRELIMINARY ISSUES
Before considering the parties' main arguments, the Court must address whether Judge Motley's findings are binding on this Court on remand. In her decision, Judge Motley made a number of factual findings regarding the use and development of the LAST, including that there was a "pervasive lack of documentation" in the record regarding validation. Gulino III, 2003 WL 25764041, at *26-27 ¶¶ 143-46. Plaintiffs argue that the Second Circuit largely affirmed Judge Motley's findings on appeal, and thus that her findings are binding on this Court on remand as the law of the case. See United States v. Quintieri, 306 F.3d 1217, 1225 (2d Cir. 2002) (explaining that a trial court must "follow an appellate court's previous ruling on an issue in the same case"). The Court rejects Plaintiffs' argument.
The Court asked the parties to address whether new evidentiary hearings were necessary in their remand briefings. The parties agreed that no new hearings were warranted, and the Court concurs. The Board asked the Court to allow it to submit new evidence regarding the validation of the current version of the LAST (the "New LAST"), which took place from 2000-2004. As noted above, the Court denied this request on the ground that evidence related to the validation of the New LAST is irrelevant. Order dated Dec. 8, 2009 (Dkt. 243). Plaintiffs were required to pass the Old LAST—used prior to 2004—and their claims are based entirely on that exam. The Court notes that the findings in this Opinion apply exclusively to the Old LAST.
On remand, a district court generally has discretion to reconsider rulings that it or another district court made in the same case, subject to two exceptions: district courts (1) must follow the decision of the appellate court "where issues have been explicitly or implicitly decided on appeal," Burrell v. United States, 467 F.3d 160, 165 (2d Cir. 2006), and (2) may not revisit an issue "that was ripe for review at the time of an initial appeal but was nonetheless foregone," United States v. Ben Zvi, 242 F.3d 89, 95 (2d. Cir. 2001). Neither exception applies here.
First, the Second Circuit did not affirm Judge Motley's findings. With respect to the documentation (or lack thereof), the Second Circuit explained the types of evidence that a defendant might use to prove validation, and stated that documentary evidence was not necessarily required. See Gulino IV, 460 F.3d at 387. The appellate court did not state that it had reviewed the record and concurred with Judge Motley's finding of a "pervasive lack of documentation," nor was such a statement implicit in the court's ruling. The Second Circuit simply never addressed the vast majority of Judge Motley's other factual findings.
The Second Circuit stated:
[t]he district court concluded that the LAST could not be properly validated under the Guardians standard at least in part because of a "pervasive lack of documentation." Were we to agree, it might be appropriate to direct the entry of judgment on remand in favor of appellants. In fact, however, we are not convinced, as a matter of law, that such judgment is warranted in this case.
The Second Circuit did review and overturn certain factual findings Judge Motley made regarding the essay portion of the LAST. Gulino IV, 460 F.3d at 387. The court's decision on these findings is the law of the case and thus binding on remand. Plaintiffs argue that by explicitly overturning Judge Motley's findings regarding the essay portion of the test, the Second Circuit implicitly affirmed the rest of her factual findings. The Court rejects this argument. Judge Motley made numerous factual findings regarding the LAST, but the fact that the Second Circuit addressed some of these findings does not support the conclusion that the Second Circuit thoroughly reviewed and implicitly accepted the remaining findings.
Second, Defendants did not forgo their right to appeal Judge Motley's findings. If a party fails to challenge a particular decision on appeal, it is deemed to have waived the right to challenge the decision. See Cnty. of Suffolk v. Stone & Webster Eng'g Corp., 106 F.3d 1112, 1117 (2d Cir. 1997). However, an issue is not considered waived "if a party did not, at the time of the purported waiver, have both an opportunity and an incentive to raise it." Quintieri, 306 F.3d at 1229. Because Defendants prevailed in the district court, they had no incentive to challenge Judge Motley's findings as to the use and development of the LAST in order to avoid a waiver. Id. at 1229-30. In these circumstances, Defendants cannot be deemed to have waived their ability to challenge Judge Motley's findings.
Accordingly, on remand the Court is not bound by Judge Motley's finding regarding the lack of documentation, nor by her factual findings regarding the development and use of the LAST. The Court notes, however, that because Judge Motley presided over this case through trial, she undoubtedly had a deep understanding of the facts. The Court thus gives deference to her findings so long as they are supported by the record. See Quintieri, 306 F.3d at 1230 (stating that a district court may reconsider previous rulings on remand, but is not obligated to do so).
* * * * *
The remainder of this Opinion addresses the merits of the issues remaining on remand: whether the Board is subject to Title VII liability; whether the LAST was validated; and whether the Board misused the LAST and the Core Battery exam to make employment decisions.
IV. APPLICABILITY OF TITLE VII
The Board and SED argue that: (1) the LAST cannot be subject to a Title VII challenge because it is a licensing exam, not an employment exam, and Title VII does not apply to licensing exams; (2) the Board cannot be liable under Title VII because it was required by state law to use the LAST, and following state law is a "business necessity" that exempts the Board from Title VII liability; and (3) the Supreme Court's decision in Ricci v. DeStefano provides the Board with a defense to Title VII liability in this case. The Court rejects the Board's and SED's arguments.
The Second Circuit acknowledged that the LAST is a licensing exam, and that the Board is a licensor, but nonetheless held that the Board can be sued under Title VII for requiring Plaintiffs to pass the LAST, because the Board is both a licensor and Plaintiffs' employer, and employers are subject to liability under Title VII. Gulino IV, 460 F.3d at 380-81. According to the Second Circuit, therefore, Plaintiffs can challenge the LAST under Title VII by suing the Board, their employer, for requiring them to pass the exam.
On remand, the Board and SED cite to a number of decisions holding that a state licensing authority cannot be liable under Title VII for requiring licensure candidates to take a licensing exam, even if the exam has a disparate impact. See Comacho v. Puerto Rico Ports Auth., 369 F.3d 570, 578 (1st Cir. 2004); Fields v. Hallsville Indep. Sch. Dist., 906 F.2d 1017, 1020 (5th Cir. 1990); George v. Bd. of Veterinary Med. Exam'rs, 635 F. Supp. 953 (D.N.J. 1985). The Board cited the same and similar decisions in its briefing on appeal, however, and the Second Circuit stated that the decisions were inapplicable because the defendants in those cases were acting solely as licensors, and not as employers. Gulino IV, 460 F.3d at 381. The Board and SED have also provided the Court with an amicus brief submitted by the U.S. Solicitor General opposing the Board's petition for a writ of certiorari. See McHale Decl. Ex. 3 (Dkt. 252). In addition to opposing the writ because of the case's interlocutory posture, the Solicitor General argued that the Second Circuit had erred in finding that the Board could be liable under Title VII for requiring teachers to pass the LAST in order to receive a teaching license and suggested that Title VII does not apply to state licensing requirements. On remand, the Court follows the Second Circuit's clear holding that the Board can be liable under Title VII.
With respect to the Board's and SED's second argument, the Second Circuit clearly stated that the Board could be liable under Title VII even though it followed state law by requiring teachers to pass the LAST. The Board and SED attempt to avoid the Second Circuit's holding by invoking the concept of "business necessity," and allege that complying with a state licensing requirement is "a business necessity to which there is no reasonable alternative." SED Remand Mem., at 36 (Dkt. 251). The Second Circuit explicitly held, however, that the Board may be liable under Title VII for requiring teachers to pass the LAST, even though it did so in accordance with state law. Gulino IV, 460 F.3d at 380 ("[T]he district court was correct in holding that the mandates of state law are no defense to Title VII liability...Title VII explicitly relieves employers from any duty to observe a state hiring provision which purports to require or permit any discriminatory hiring practice." (internal quotation omitted)). The Court conclusively rejects the Board's business necessity argument.
Indeed, the Board and SED admit that the Second Circuit addressed their argument in its decision, but argue that this Court should reconsider the Second Circuit's decision because it was only "two sentences long" and thus "conclusory." The Court, however, will not ignore a clear decision by the Second Circuit simply because it was stated concisely. In addition, the Court notes that the Board's argument is not properly before the Court on remand. The Board argued before Judge Motley that it could not be held liable under Title VII because it was required by state law to use the LAST, but failed to raise this argument on appeal. On remand, the Court cannot consider an issue that was "ripe for review at the time of an initial appeal, but nonetheless forgone." Quintieri, 306 F.3d at 1230.
Finally, the Board argues that the Supreme Court's decision in Ricci v. DeStefano, as interpreted by the Second Circuit in United States v. Brennan, 650 F.3d 65 (2d Cir. 2011), absolves it of Title VII liability. In Ricci, white and Hispanic firefighters challenged the City of New Haven's decision to discard the results of an employment exam, fearing Title VII disparate impact liability, after white firefighters outperformed minority candidates on the exam. Ricci, 557 U.S. at 562-63. The Supreme Court held that concern about disparate impact liability can excuse an otherwise impermissible action only if supported by a "strong basis in evidence" that the employer would face such liability. Id. at 584. The Board now argues that, for Plaintiffs to prevail, the Board would have been required to take a race-conscious action by disregarding the LAST, which, the Board argues, could be done only with a "strong basis in evidence." See Letter from Charles E. Enloe, dated May 13, 2011. Because there was no such evidence, the Board contends, the Board could not disregard the LAST.
The Board's argument is incorrect. The Second Circuit foreclosed this application of Ricci, and explicitly held that the "sound basis in evidence" standard does not apply to disparate impact claims. See Briscoe v. City of New Haven, 654 F.3d 200, 205-06 (2d Cir. 2011) (rejecting city's argument that "an employer may defeat a disparate impact claim if it had a strong basis in evidence that it would have been subject to disparate treatment liability"). Among other reasons, the Second Circuit grounded this holding on the fact that disparate impact law is well-settled and adding this new requirement would only "muddle that which is already clear," and that employers would have difficulty establishing a "sound basis in evidence" for their own hypothetical future discriminatory actions. Id. at 207-08. Given this holding, the Court rejects the Board's reading of Ricci and agrees with Plaintiffs that the Guardians standard controls in this case.
V. WHETHER THE LAST WAS PROPERLY VALIDATED
The Court finds that the LAST was not properly validated and thus is not job related.
A. Background: Development and Use of the LAST
In 1988, a New York State task force studying teacher qualifications determined that all teachers should have a basic understanding of the liberal arts and sciences in order to be competent to teach. The task force recommended that the state require teachers to pass a liberal arts and sciences exam prior to receiving state certification in order to prevent any harm to students that would be caused by having incompetent teachers in the classroom. SED Ex. 35 pp. 617-18; SED Ex. 800 p. 11. SED was responsible for implementing the task force's recommendation. In 1990, SED entered into a contract with NES, a professional test development company, to develop the LAST.
NES established two committees to assist with the test development process: the Bias Review Committee ("BRC") and the Content Advisory Committee ("CAC"). Gulino III, 2003 WL 25764041, at *21 ¶ 108. The BRC reviewed all proposed materials for the LAST to ensure that they were free from bias. Id. at *21 ¶ 111. The BRC members included twenty-two New York public school teachers and education college faculty, including ten African Americans, three Asian Americans, one Native American, and two individuals with disabilities. Id. The CAC reviewed all LAST materials to ensure that they were both free from bias and relevant to the job of a New York public school teacher. Id. at *21 ¶ 110. The CAC consisted of twenty-two teachers and education college faculty members with specialized expertise in the liberal arts and sciences. Id.
To create the LAST, NES first developed a draft framework. The framework identified five general areas that would be tested on the exam: (1) Scientific and Mathematical Processes; (2) Historical and Social Scientific Awareness; (3) Artistic Expression and the Humanities; (4) Communication Skills; and (5) Written Analysis and Expression. SED Ex. 395. For each area, NES developed four or five subtopics that would be tested. For example, the first subtopic identified under "Scientific and Mathematical Processes," was the "use [of] mathematical reasoning in problem-solving situations to arrive at logical conclusions and to analyze the problem-solving process." SED Ex. 395. The first subtopic identified under "Historical and Social Scientific Awareness" was understanding "the interrelatedness of economic, geographic, political, social, and cultural issues and factors." SED Ex. 395.
The parties use the word "objective" to indicate a subset of knowledge targeted by the examination. For clarity, the Court will use the word "subtopic" instead of "objective."
The BRC and the CAC reviewed the draft framework and subtopics. Following this review, NES sent the framework and subtopics to eight hundred New York public school teachers and four hundred education college faculty members, asking these individuals to rate the importance of each subtopic on a scale from 1 to 5, with 1 being of "no importance" and 5 being of "very great importance." SED Exs. 401, 406, 408-10. Five hundred and fifty-four teachers and two hundred and thirty faculty members responded. Each subtopic received a separate rating, with the lowest rating being a 3.08 and the highest a 4.51; the average importance rating for the subtopics was 3.83 (a 4 being of "great importance"). SED Ex. 410.
NES next developed a "question bank" of three hundred and fifty test questions relating to the subtopics; the BRC and the CAC reviewed a small sample of these questions. Trial Tr. 1157, 2612-15; SED Exs. 418-19, 412-30; Gulino III, 2003 WL 25764041, at *28 ¶ 149 (noting the committees reviewed only eighty out of three hundred and fifty test questions). NES conducted pilot tests of some of the questions by having students at various New York education colleges take shortened versions of the LAST, each with thirty-six questions from the question bank. Trial Tr. 1164-65, 1171-73. The pilot testing was intended to ensure that the questions were understandable, relevant, and unbiased.
After the questions were refined by the above-described processes, NES and SED began administering the LAST to teacher candidates. The exam had eighty multiple choice questions (selected from the question bank), only sixty-four of which were scored, and an essay question intended to assess a candidate's reading comprehension, writing skills, and analytical ability. Gulino III, 2003 WL 25764041, at *22 ¶ 116. The multiple-choice section comprised 80% of a test-taker's total score, and the essay question comprised 20%. Id. A test-taker received one point for each multiple choice question answered correctly. The essay question was scored on a scale of 1 to 4, with 1 being the lowest score and 4 the highest. Test-takers were required to achieve a passing score on the LAST in order to receive state certification.
The BRC and the CAC were charged with determining what should constitute a passing score on the LAST. To do so, committee members reviewed an early version of the LAST and estimated the percentage of test-takers that would answer each multiple-choice question correctly and the average essay score. Based on this review, the CAC recommended a passing score of 48, and the BRC recommended a passing score of 38. Id. at *28 ¶ 149. The New York State Education Commissioner ("Commissioner") set the passing score at 38. Id.
This method of determining a passing score is known as the "Modified Angoff Method." Using this Method, committee members reviewed each multiple-choice question on the exam, and asked the question:
Imagine a hypothetical group of individuals who have the minimally acceptable level of knowledge and skills required to perform the job of an educator receiving a provisional certificate in New York State. What percent of this group would answer the item correctly?For the essay question, the committee members asked a similar question to determine what score (1-4) a minimally-competent test-taker would receive on the essay. SED Ex. 494. The committees performed this analysis on only eighty questions; an Angoff analysis was never conducted on the remaining two hundred and seventy questions. Gulino, 2003 WL 25764041, at *28 ¶ 149.
In 1997, the Commissioner decided to update the certification standards for public school teachers. From 1997 to 1998, SED and NES reviewed and updated the LAST, using procedures similar to those described above. In addition, a new committee of teachers reviewed the LAST to determine the appropriate passing score. The committee recommended raising the passing score to 44. SED Exs. 588-89, 591-93, 782, 810. Based on the committee's recommendation, the Commissioner ultimately raised the passing score to 43. Gulino III, 2003 WL 25764041, at *28 ¶ 150.
At trial, Judge Motley found that Caucasian test-takers passed the LAST at a statistically significant higher rate than African-American and Latino test-takers. Id. at *9-10 ¶¶ 46-49. Accordingly, requiring teachers to pass the LAST in order to receive a City license had a disparate impact on African-American and Latino teachers. Id. at *11 ¶ 57. The Board did not challenge this ruling on appeal, and it is not at issue on remand.
B. Legal Standard for Validation
Under Title VII, an exam is job related—a statutory defense for an employer using an exam with a disparate impact—if it has been properly validated. Validation requires showing, "by professionally acceptable methods, [that the exam is] 'predictive of or significantly correlated with important elements of work behavior which comprise or are relevant to the job or jobs for which candidates are being evaluated.'" Gulino IV, 460 F.3d at 383 (quoting Albemarle Paper, 422 U.S. at 431).
The Second Circuit uses a five-part test for determining whether a content-related employment exam, such as the LAST, has been properly validated and is thus job related for the purposes of Title VII: (1) the test-makers must have conducted a suitable job analysis; (2) the test-makers must have used reasonable competence in constructing the test; (3) the content of the test must be related to the content of the job; (4) the content of the test must be representative of the content of the job; and (5) there must be a scoring system that usefully selects those applicants who can better perform the job. Guardians, 630 F.2d at 95; see also Gulino IV, 460 F.3d at 384. The first two elements of this test concern the quality of the test's development. Guardians, 630 F.2d at 95. These parts are "particularly crucial" because "validity is determined by a set of operations, and one evaluates...validity by the thoroughness and care with which these operations have been conducted." Id. at 95 n.14 (internal citation and quotation omitted). The last three factors establish standards that that an exam, "as produced and used, must be shown to have met." Id. The "essence of content validation" is in the third requirement: "that the content of the test be related to the content of the job." M.O.C.H.A. Soc'y, Inc. v. City of Buffalo, No. 98 Civ. 99, 2009 WL 604898, at *14-15 (S.D.N.Y. Mar. 9, 2009) (Curtin, J.) ("M.O.C.H.A. Soc'y I"), aff'd, 689 F.3d 263 (2d Cir. 2012).
SED argues that the LAST is a "licensing exam," not an employment exam, and thus that the Court should modify the Guardians standard in this case. SED asserts that a licensing exam is different from an employment exam because a licensing exam is intended to assess whether a test-taker "has the knowledge or skills required of a minimally competent person in an occupation," whereas an employment test is intended "to select from among [test-takers] those who can perform better." See SED Remand Mem., at 5. However, the Second Circuit noted that the LAST was a licensing exam, but explicitly stated that the Court should apply the Guardians test to the LAST on remand. Gulino IV, 460 F.3d at 384-85.
Validation is technical and complex, and "not primarily a legal subject." Guardians, 630 F.2d at 89. Accordingly, when determining whether an employment exam is properly validated, a court "must take into account the expertise of test validation professionals." Gulino IV, 460 F.3d at 383. There are two primary bodies of expertise on which courts rely to assess validation: (1) the testimony of experts in the field of test validation; and (2) the Equal Employment Opportunity Commission's Uniform Guidelines on Employee Selection Procedures ("Guidelines"), which establish standards for properly validating an employment test. See 29 C.F.R. § 1607.1-1607.18 (2006). Courts are not bound by the Guidelines, but the Supreme Court has stated that they are "entitled to great deference" because they represent "the administrative interpretation of [Title VII] by the enforcing agency." Griggs v. Duke Power Co., 401 U.S. 424, 433-34 (1971). Although the Second Circuit applies the Guidelines with "the appropriate mixture of 'deference and wariness,'" Gulino IV, 460 F.3d at 384 (quoting Guardians, 630 F.2d at 91), the Guidelines should be the "primary yardstick by which [to] measure [D]efendant['s] attempt to validate the LAST" because of their longstanding use in the field. Gulino IV, 460 F.3d at 384.
The defendant bears the burden of proving that a challenged employment exam is properly validated. See Albemarle Paper Co., 422 U.S. at 425; Gulino IV, 460 F.3d at 388. Even if an exam appears on its face to be job related, a defendant must submit sufficient evidence to evaluate the procedures used to construct the exam and the exam's content so that the Court can determine whether the exam is "significantly correlated with important elements of work behavior which comprise or are relevant to the job or jobs for which candidates are being evaluated." See Gulino IV, 460 F.3d at 383 (quoting Albemarle Paper Co., 422 U.S. at 431); 29 C.F.R. § 1607.15(A)(3) ("[E]vidence should . . . permit direct evaluation of the validity of the [employment exam]."); see also Guardians, 630 F.2d at 92 (stating that a test is job related only if "the abilities being tested for can be determined, by direct, verifiable observation, to be required or desirable for the job" (emphasis added)).
The Guidelines state that a defendant must submit documentary evidence of the process used to develop a challenged exam in order to prove that the exam was properly validated. 29 C.F.R. § 1607.15(A)(3). As discussed above, however, the Second Circuit has stated that such documentary evidence is not required to prove validation. Gulino IV, 460 F.3d at 387-88. Rather, "first-hand accounts of those involved in the test validation process, [and] studied opinions of certified experts, may be sufficient, in some circumstances, to establish" that an exam was properly validated. Id. at 388. A defendant, however, bears "a heavy burden . . . in seeking to validate a test without relying on documentary evidence." Id. Validation cannot be established "'through vague and unsubstantiated hearsay.'" Id. (quoting Albemarle Paper Co., 422 U.S. at 428 n.23).
C. Application
Based on the evidence presented by the parties at trial and in their remand submissions, the Court finds that the LAST was not properly validated, because (1) the test-maker (NES) did not conduct a suitable job analysis; (2) the test-maker did not use reasonable competence in constructing the LAST; (3) the content of the LAST is not related to the content of teaching; (4) the content of the LAST is not representative of the content of teaching; and (5) the scoring system did not usefully select applicants who could better perform as teachers. See Guardians, 630 F.2d at 95; see also Gulino IV, 460 F.3d at 384. Thus, the LAST is not job-related and violates Title VII.
1. NES Did Not Conduct a Suitable Job Analysis
A job analysis is an "assessment 'of the important work behaviors required for successful performance" of the job in question and the "'relative importance'" of these behaviors. Guardians, 630 F.2d at 95 (quoting 29 C.F.R. § 1607.14(C)(2)). The purpose of a job analysis is to ensure that the exam adequately tests for knowledge, skills, and abilities that are actually needed to perform the daily tasks of the job in question. See United States v. City of N.Y., 637 F. Supp. 2d 77, 111 (E.D.N.Y. 2009) (Vulcan II). To perform a suitable job analysis, a test developer must: (1) identify the tasks involved in performing the job; (2) determine the relative importance of these tasks; and (3) identify the knowledge, skills, and abilities necessary to complete the tasks. Guardians, 630 F.2d at 95-96; see also M.O.C.H.A. Soc'y, Inc. v. City of Buffalo, 689 F.3d 263, 278 (2d Cir. 2012) ("M.O.C.H.A. Soc'y II") (explaining that a proper job analysis "includ[es] a thorough survey of the relative importance of the various skills involved in the job in question and the degree of competency required in regard to each skill" (internal quotation omitted)). The test developer must be able to explain the relationship between the subject matter being assessed by the exam and the job tasks identified. Compare Vulcan II, 637 F. Supp. 2d at 111 (finding that defendant's job analysis for a test given to firefighter candidates was inadequate because no effort had been made to explain the relationship between the knowledge, skills, and abilities being tested on the exam and the tasks involved in being a firefighter), and M.O.C.H.A. Soc'y I, 2009 WL 604898, at *14 (finding "comprehensive" job analysis adequate where employer had conducted multiple surveys, statistical analyses, and solicited committee input to ensure subjects evaluated by exam related to the tasks involved in being a firefighter). This requirement ensures that "the pertinent abilities have been selected for measurement." Vulcan II, 637 F. Supp. 2d at 111.
The Board does not attempt to argue that NES performed the first two steps required by Guardians (identifying tasks involved in the job and determining the relative importance of those tasks). Without performing these crucial threshold steps, the Court doubts that NES could have performed a suitable job analysis. Nonetheless, the Board asserts the process NES used to develop the LAST's subtopics constitutes a suitable job analysis, and argues that the knowledge identified in the subtopics is relevant to teachers' jobs because: (1) NES developed the subtopics by reviewing teaching and education materials, and consulting with education experts; and (2) the BRC, the CAC, and the survey of teachers and college faculty members confirmed that the subtopics were important for teachers.
The Board also argues that NES went through similar procedures when it "revalidated" the LAST in 1997, and that this revalidation process provides additional evidence that NES completed a suitable job analysis. However, the Board failed to offer any specific evidence regarding the revalidation process in 1997; the evidence presented on revalidation is very similar to the evidence presented for the initial development process. The Court finds this evidence unpersuasive.
The Court agrees that NES made some effort to ensure that the LAST's subtopics identify knowledge of the liberal arts and sciences, which is important to teaching. There are, however, several flaws in the way that NES developed and reviewed the subtopics that prevent Court from finding that the company conducted a suitable job analysis.
First, NES never created a list of the tasks teachers perform, nor determined whether the subtopics identify knowledge needed to perform those tasks. See 29 C.F.R. § 1607.14(C)(2); Gulino III, 2003 WL 24764041, at *18-19 ¶¶ 95-101 (finding that creator of the Core Battery exam did complete a suitable job analysis, in part by performing a task analysis that identified eighty-three tasks important to the jobs of all teachers and six core job functions common to all teachers). At trial, NES representatives testified that the company had intended to analyze the tasks needed for teaching prior to drafting the subtopics, but the representatives did not know whether such task identification was ever actually carried out; there is no documentary evidence of task identification. Id. at *23 ¶¶ 121-126; Trial Tr. at 114-15; 161, 168-69; 701-02 (although a task analysis was supposed to be performed, two SED officials testified that they were not sure if it actually happened). There is no evidence that NES ever identified important tasks, let alone assessed the tasks' relative importance; NES thus cannot explain the relationship between the tasks and the subtopics. That failure makes it impossible for the Court to determine whether the subtopics identify knowledge that is pertinent to job tasks that all teachers must perform. See Vulcan II, 637 F. Supp. 2d at 111 (finding that defendant's failure to provide evidence linking knowledge, skills, and abilities to job tasks "undermines the court's confidence that 'the pertinent abilities have been selected for measurement'" (quoting Guardians, 630 F.2d at 96)).
A task analysis would have been particularly useful in this case because the City's public school teachers teach a range of subjects and many grades, from kindergarten to advanced high school science. There are similarities and differences in what these teachers do on a daily basis. The LAST, however, is intended to test for knowledge about the liberal arts and sciences that all teachers need in order to be competent in their jobs. A task analysis could have helped identify job tasks common to all teachers, and determine what knowledge about the liberal arts and sciences is necessary to succeed at those tasks.
Second, there is insufficient evidence in the record regarding the materials NES relied on to draft the subtopics. At trial, NES representatives testified that the company collected materials from schools and colleges throughout the state, interviewed deans and administrators of liberal arts programs at colleges and universities in New York, and consulted with education experts. The representatives, however, could not recall any details about the materials or interviews, and could not describe how the information collected was used or how it supported the choice of subtopics. See, e.g., Trial Tr. 1847:23-1848:4; Trial Tr. 416:3-6; 416:21-417:7; 799:4-7; 1868:9-1869:6; Trial Tr. 394-396; 1585-26; 1861-62. Absent such evidence, the Court has no basis to find that "pertinent abilities" have been selected for measurement. Compare M.O.C.H.A. Soc'y I, 2009 WL 604898, at *14 (approving job analysis where test developer had "painstakingly documented" her procedures in developing the exam).
The Board submitted little documentary evidence regarding this process, and NES did not maintain any of the materials it collected. There are no notes or summaries of information gathered from the faculty, teachers, and content experts, and there are no documents showing how their advice was incorporated into the subtopics. The Board cites repeatedly to a report by NES regarding the development of the LAST. SED Ex. 494. This report, however, provides only a broad summary of the steps NES took when drafting the subtopics; it offers no details on the materials or interviews. Id. at pp. 15-19. There is no expert testimony describing the information NES collected or explaining how that information supported the choice of subtopics. Without more specific evidence, the Court cannot determine whether NES relied on relevant information and expertise that identified knowledge about the liberal arts and sciences that is important to common teaching tasks. See Vulcan II, 637 F. Supp. 2d at 112 (finding that report summarizing the steps of defendant's job analysis did not satisfy defendant's burden of proving that it had conducted a suitable analysis); Cuesta v. N.Y. State Office of Court Admin., 657 F. Supp. 1084, 1097 (S.D.N.Y. 1987) (Leval, J.) (finding that defendant had conducted a suitable job analysis by performing an "exhaustive analysis" of the job in question, creating an "extensive task list," identifying the "skills, knowledge, and abilities" required for these tasks, and compiling a report that allowed "one who has had no contact with the job [to] form a clear picture of what incumbents do in the job").
It also appears that NES actually drafted the subtopics prior to collecting materials and conducting interviews. At trial, Plaintiffs introduced a draft of the framework and subtopics, dated March 12, 1991, Pls. Ex. 105, which is quite similar to the final version of the framework and subtopics, dated July 26, 1991, Pls. Ex. 126. NES did not begin collecting materials or conducting interviews until some time after March 18, 1991. Pls. Ex. 107. Although the content experts apparently reviewed the March 12 draft of the subtopics, there is no evidence that they were consulted while NES was developing the draft. This evidence suggests that NES developed the subtopics largely without the assistance of relevant materials or experts, and to the extent such materials and experts were consulted, they had little impact on the final draft. This fact, combined with NES's failure to conduct a task identification or to maintain materials and interview notes, leaves the Court with no specific evidence documenting how NES developed the LAST's subtopics, or whether NES verified that the subtopics were linked to knowledge about the liberal arts and sciences that all teachers must have to be competent in their jobs.
Although NES did make some effort to conduct a job analysis, including the BRC and CAC reviews and the teacher/faculty survey, these efforts are insufficient to cure the weaknesses identified above. The BRC and CAC were not involved in developing the subtopics, they merely reviewed the subtopics after NES had already drafted them, and made only insubstantial changes to the draft subtopics they reviewed. This does not constitute the rigorous job analysis required by Guardians.
Accordingly, the Court finds that the Board has failed to establish that NES conducted a suitable job analysis when developing the LAST.
2. NES Did Not Use Reasonable Competence in Constructing the LAST
The second step of the validation test examines whether the employer "used reasonable competence in constructing the exam." Guardians, 630 F.2d at 95. An exam was likely not prepared with reasonable competence if "(1) the examination was not created by professional test preparers, or (2) no sample study was performed to ensure that the questions were comprehensible and unambiguous." M.O.C.H.A. Soc'y, II, 689 F.3d at 280; see also Guardians, 630 F.3d at 96-97. Pilot testing should use a sample of potential test-takers that is as representative as possible of the population for which the test is intended. Cuesta, 657 F. Supp. at 1097-98; see also Am. Pysch. Ass'n Standards, 3.8. Although some effort was made to ensure that the LAST was competently constructed, gaps in the evidence regarding test construction and flaws in the pilot testing compel the Court to find that NES did not use reasonable competence in constructing the LAST.
These are standards for the development of professional and educational tests, developed by the American Psychological Association ("APA"). All parties to this case agree that the APA Standards represent reliable expert opinion on the validation process.
Although the LAST was developed by a professional test company, NES retained little specific documentation of the construction process. An NES representative testified at trial that NES employees drafted the exam questions; these employees reviewed relevant materials and consulted with experts when drafting questions; the questions were reviewed and revised by outside content experts; and the CAC and the BRC reviewed the final draft of the questions. Trial Tr. 1155-57, 1811-15. This testimony, however, provides only a broad and vague overview of the exam construction process. There is no evidence in the record regarding the materials NES employees relied on when drafting the questions; the names or experience of the experts consulted; or any advice, comments, or revisions provided by these experts. Compare Cuesta, 657 F. Supp. at 1097-98 (finding the exam at issue was competently constructed in part because it was based on the results of a "strong job analysis"); M.O.C.H.A. Soc'y I, 2009 WL 604898, at *14-15 (finding competent construction where "painstakingly...accumulated data was thoroughly organized, analyzed, rated" and "reviewed by a panel of experts"). Accordingly, based on the evidence presented, the Court cannot find that the LAST was competently constructed.
As with the job analysis, the BRC's and the CAC's reviews of the questions provide some assurance that the questions are relevant and accurate, but do not offset the lack of evidence regarding the rest of the drafting process.
As for testing sample questions, NES did pilot test questions by having students at education colleges take shortened versions of the LAST (with thirty-six questions). However, the sample population for the pilot tests—college students preparing to become teachers—was not representative of the full population of test-takers. Pls. Exs. 153, 242. Working teachers, many of whom had been teaching for several years, were required to take the exam, but were not included in the pilot testing. The Court finds that this flaw in the pilot testing is significant. At trial, the Board's and SED's expert testified that the LAST covered information that teachers would learn in college liberal arts and science classes. Trial Tr. 3225-27. It follows, therefore, that the information was likely to be more accessible to a student still in college than to a teacher who has been out of school and teaching in a specific field for a number of years. Pls. Ex. 242.
Indeed, Plaintiffs' expert analyzed pass rates for the LAST and found that college seniors passed the exam at higher rates than test-takers with a Bachelor's or Master's degree, or a Doctorate. Pls. Ex. 241, pp. 71-72.
Accordingly, the Court finds that the Board has failed to establish that the LAST was competently constructed.
3. The Content of the LAST Is Not Directly Related to Teaching
The third Guardians requirement is that the content of the exam must be directly related to the content of the job. This requirement "reflects 'the central requirement of Title VII' that [an exam] be job-related." Vulcan II, 637 F. Supp. 2d at 116 (quoting Guardians, 630 F.2d at 97-98). As a threshold matter, the Court notes that content relatedness is intertwined with the job analysis—that is, an exam's developers should identify the content of the job to be tested. See, e.g., M.O.C.H.A. Soc'y II, 689 F.3d at 281 (approving district court's finding of content relatedness where "underlying job analysis was suitable"). Indeed, if an employer has competently completed the first two steps of the validation process by performing a suitable job analysis and used reasonable competence in constructing the exam, a court may infer that the exam is job related. See Vulcan Soc'y of the N.Y. City Fire Dep't v. Civil Serv. Comm'n of the City of N.Y., 490 F.2d 387, 396 (2d Cir. 1973) ("Vulcan I"); see also Cuesta, 657 F. Supp. at 1098 (finding that "the professional translation of job analysis results into test items provides great assurance that the abilities tested for are actually required for the job"). Where, as here, there are flaws in the job analysis and exam construction process, the employer must present "convincing evidence" that the exam is nonetheless job related. Id.
As discussed above, the Board has failed to prove that NES performed an adequate job analysis or competently constructed the LAST, and has provided no evidence regarding what tasks teachers perform on the job, what knowledge of the liberal arts and sciences teachers need to perform these tasks, or how the LAST's subtopics relate to this knowledge. Given these evidentiary gaps, the Board must present "convincing evidence" that the LAST is nonetheless related to teachers' jobs. The Board has failed to present such evidence.
The Board argues that the LAST is job related because experts, including the New York state task force, have determined that all teachers should know about the liberal arts and sciences, and the LAST tests for knowledge related to the liberal arts and sciences. The Court does not question the experts' judgment that the LAST tests for knowledge regarding the liberal arts and sciences. According to the experts, however, all teachers must have a level of knowledge about the liberal arts and sciences that ensures that they are competent to teach. See SED Ex. 35 Task Force Report at 617-18; SED Ex. 800 Report of Defs.'s Expert at 16-21. The experts further explain that the LAST is intended to measure the minimum level of knowledge necessary to ensure such competence. See SED Ex. 800 at 13-15. The fact that the LAST is related generally to the liberal arts and sciences does not prove that the exam is job related; indeed, the liberal arts and sciences is an extremely broad field that encompasses far more than the basic knowledge all teachers need in order to be competent. Rather, to be job related, the LAST must test for the minimum level of knowledge about the liberal arts and sciences that is necessary to ensure that all teachers are competent to teach.
There is no evidence in the record establishing the minimum level of knowledge about the liberal arts and sciences needed by all teachers. The task force did not make any recommendations regarding specific knowledge that should be tested for on the LAST. See SED Ex. 35 at 617. NES did not conduct a job analysis to determine what knowledge is important for all teachers in carrying out their daily tasks. The company also did not maintain the materials it consulted when drafting the LAST, which might have helped establish a core liberal arts and sciences curriculum that all teachers use when teaching. Finally, there was no testimony or expert opinion offered at trial regarding what teachers need to know about the liberal arts and sciences, or how the LAST tests for that knowledge. Without such evidence, the Court cannot find that the LAST is directly related to knowledge about the liberal arts and sciences that all teachers must possess in order to be competent.
Consequently, because of the flaws in the job analysis and exam construction process, and the Board's failure to present "convincing evidence" of job relatedness, the Court finds that the Board has failed to establish that the LAST is directly related to teachers' jobs.
4. The Content of the LAST Is Not a Representative Sample of Teaching
The fourth Guardians requirement is that the content of the exam must be "a representative sample of the content of the job." Guardians, 630 F.2d at 98; see also 29 C.F.R. § 1607.14(C)(4). This does not mean that "all the knowledge, skills, or abilities required for the job [must] be tested for, each in its proper proportion." Guardians, 630 F.2d at 98. Rather, this requirement is meant to ensure that the exam measures important aspects of the job, and does not overemphasize minor aspects. Id.; see also M.O.C.H.A. Soc'y I, 2009 WL 604898, at *16. To meet this requirement, a defendant must provide evidence of the important knowledge, skills, and abilities required for the job, and demonstrate the extent to which they are tested for on the exam. Vulcan II, 637 F. Supp. 2d at 119.
The Court has already found that the Board has failed to establish what minimum knowledge about the liberal arts and sciences teachers need in order to be competent. The Board thus cannot demonstrate the extent to which the exam tests for this knowledge. Compare M.O.C.H.A. Soc'y I, 2009 WL 604898, at *16 (finding exam to be content representative based on testimony and evidence related test developer's identification of tasks during the job analysis). Accordingly, the Court finds that the Board has failed to establish that the LAST is representative of the content of the job.
The Board argues that the teacher/faculty survey establishes that all of the LAST's subtopics are important to the job of a teacher, and that the LAST tests for all of the subtopics. The Court agrees that the teacher/faculty survey provides some assurance that the subtopics are important for teaching. The Court has already found, however, that the survey respondents' limited role in the development of the LAST weighs against relying too heavily on the results of the survey to prove job relatedness. The Court will not find, on the basis of the teacher/faculty survey alone, that the content of the LAST is representative of the content of the job. See Gulino III, 2003 WL 25764041, at *19 ¶¶ 97-98 (describing how the company that created the Core Battery exam conducted a teacher survey to verify and provide additional evidence that the tasks identified in their original analysis were important for all teachers).
5. The Scoring Requirements on the LAST Do Not Usefully Select Individuals Who Would Be Better Teachers
In the fifth and final step of the Guardians test, a court must determine whether the exam is scored in a way that usefully selects those applicants who can better perform the job. Guardians, 630 F.2d at 105. A cutoff score (a minimum passing score) must be set "so as to be reasonable and consistent with normal expectations of acceptable proficiency within the work force." Id. (quoting 29 C.F.R. § 1607.5(H)). An employer must have "some independent basis for choosing the cutoff [score]." Id. For example, an employer might establish a cutoff score on the basis of "a professional estimate of the requisite ability levels" or "by analyzing the test results to locate a logical 'break-point' in the distribution of scores." Id. To establish that a cutoff score is valid, an employer must present evidence that the score measures the minimum qualifications necessary to succeed at the job. See Vulcan II, 637 F. Supp. 2d at 125. Title VII is violated "when a cutoff score unrelated to job performance produces disparate racial results." Guardians, 630 F.2d at 105.
SED and the Commissioner did make some effort to ensure that the LAST cutoff score was valid. Members of the BRC and the CAC reviewed the first version of the LAST administered and recommended cutoff scores; the Commissioner chose the BRC's recommendation of 38. Gulino III, 2003 WL 25764041 at *28 ¶ 149. The Commissioner later raised the score to 43 on the recommendation of a separate committee of teachers. Id. at *28 ¶ 150. These steps establish that the Commissioner did have some independent basis for choosing the cutoff scores.
There are, however, several flaws in the way that the cutoff scores were calculated that undermine their validity. First, there were three hundred and fifty exam questions in the question bank, all of which could be used in versions of the LAST administered to candidates, but the BRC and CAC reviewed only eighty of these questions. Id. at *28 ¶ 149. There was no independent review or estimation of passing rates for the other two hundred and seventy questions in the question bank. Id.; Trial Tr. 2865-67.
Second, it is not clear that the committee members used appropriate criteria to evaluate the cutoff score. Members of the committees were asked to imagine a hypothetical "minimally- competent teacher" and to predict the likelihood that this teacher would answer each question on the exam correctly. SED Ex. 468, pp. 19-21; Pls. Ex. 215. There is no evidence that the committees were given any guidance as to the definition of "minimally-competent," or the definition of types of teachers might meet that standard. SED Ex. 468; Pls. Ex. 215; Trial Tr. 1338-39 (testimony from Plaintiffs' expert indicating that, in his experience, individuals determining cutoff scores have difficulty understanding what constitutes minimum competence). In particular, SED did not explain to the committees that the cutoff score should measure the minimum level of knowledge teachers need in order to be competent. Gulino III, 2003 WL 25764041, at *28 ¶ 148. There is no evidence regarding the criteria the committee members actually used to evaluate minimal competence: no notes were taken of the committee meetings, and members were not asked to explain their reasoning. At trial, a member of the BRC and score review committee testified that, in her understanding, the purpose of the cutoff score was not to distinguish between minimally competent and incompetent teachers. Trial Trial Tr. 2915-19. Experts, however, agree that this is "precisely the purpose of a cutoff score in a licensing test." Gulino III, *28 ¶ 152; Trial Tr. 3215-16; 1362. Even considering the review conducted by the committees, then, the Court cannot be certain that cutoff scores represent the level of knowledge that all teachers need to be minimally competent.
Finally, when the Commissioner decided to raise the cutoff score, he was presented with statistics showing that a higher score would disproportionately harm minority test-takers; at a score of 44, only 29% of African-Americans and 25% of Latinos would pass the exam, as compared to 78% of whites. Pls. Exs. 217, 218. At the same time, there was no evidence that higher scores on the LAST correlated with better teacher or student performance in the classroom. Trial Tr. 3337-40. At trial, the Commissioner testified that he had asked SED to conduct a study evaluating the relationship between LAST scores and student performance, but that this had not happened. Trial Tr. 3339. The Deputy Commissioner testified that such a study would be "a good idea," but that one had not taken place at the time of the trial. Trial Tr. 600. The Commissioner, therefore, raised the cutoff score knowing that his decision would have a disparate impact on minority test-takers, and without any evidence that the higher score was necessary to select competent teachers.
The Board urges the Court to rely on a 2002 study published by two university researchers. SED Ex. 173. The study has no bearing on the matters under consideration in this Opinion. It is merely descriptive, as is recognized by its authors when they say that the study aims to "describe the sorting of teachers across schools" (i.e., the high proportion of poor, minority and low-performing students who are taught by "less qualified" teachers), and "does not test hypotheses for why this sorting occurs." The study does not test whether incrementally higher scores on the LAST correlate with better student performance. For example, the study does not show—or even attempt to show—that students with a teacher who receives a score of 44 on the LAST perform better than students with a teacher who receives a score of 40. Moreover, the study was published in 2002, and thus could not have been the basis for the Commissioner's decision to raise the passing score in 1998.
--------
Given the problems discussed above, the Court finds that the Board has failed to establish that the LAST cutoff scores accurately measured the minimum knowledge about the liberal arts and sciences that teachers need to be competent and avoid harming students.
* * * * *
In conclusion, the Court finds that the Board has failed to establish that the LAST was properly validated. Accordingly, the LAST is not job related, and the Board violated Title VII by requiring Plaintiffs to pass the exam in order to receive a teaching license.
VI. WHETHER THE BOARD MISUSED THE LAST AND THE CORE BATTERY EXAM
Plaintiffs also argue that the Board misused the Core Battery and the LAST by requiring experienced licensed teachers to take the exams, and by reducing their salaries, benefits, and seniority for failure to pass. The Court has already found that the LAST was not properly validated for use as a licensing exam for any teacher, experienced or otherwise. This holding precludes separate consideration of Plaintiffs' misuse argument as to the LAST. The Court further finds that the Board did not misuse the Core Battery exam.
Plaintiffs argue that, to the extent that the Core Battery exam was properly validated, it were validated only for the purpose of assessing whether inexperienced teacher candidates had the minimum knowledge necessary to begin teaching in the classroom. Plaintiffs assert that the exam was not intended to assess the qualifications of experienced teachers, or to make decisions regarding the salaries, benefits, and seniority of such teachers. Under this view, the Board violated Title VII by "misusing" the Core Battery exam to license experienced teachers, and by reducing their salaries, benefits, and seniority if they failed to pass the Core Battery exam within five years.
Under Title VII, the exam must be properly put to the use for which the employer intends it. See APA Standards 6.3, 11.5; Walston v. Cnty. Sch. Bd. of Nansemond Cnty., Virginia, 492 F.2d 919, 927 (4th Cir. 1974). Here, the Board and SED intended to use the Core Battery exam as a licensing exam for all public school teachers who had not received a permanent license prior to 1985. Accordingly, the exam was required to have been validated for that purpose, and be "significantly correlated with important elements" of teaching that teachers need to know prior to receiving a permanent license. Albemarle Paper, 422 U.S. at 431 (quoting 29 C.F.R. § 1607.4(c)). In arguing that the Board misused the Core Battery exam by requiring experienced teachers such as Plaintiffs to take the exam and reducing their salaries and benefits if they failed to pass, Plaintiffs essentially contend that the Core Battery exam was not properly validated for use on experienced teachers.
Judge Motley found that the Core Battery exam was properly validated for use as a licensing exam for all teachers who were not already permanently licensed, including teachers with temporary or conditional licenses. Gulino III, 2003 WL 25764041, at *19 ¶ 101, *30 ¶ 161. The company that created and validated the Core Battery exam, Educational Testing Services ("ETS"), explicitly stated in its guidelines for the exam's use that it was intended for all teachers who were not permanently licensed, including experienced teachers with temporary or conditional licenses, to pass the exam in order to receive a permanent license. Pls. Ex. 79 at 9. Accordingly, in compliance with Title VII, the Core Battery exam was properly put to the uses for which it was intended when experienced teachers were required to pass the exam. See APA Standards 6.3, 11.5; Walston, 492 F.2d at 927.
Because it was appropriate for the Board to require these teachers to pass the Core exam, the Board also did not violate Title VII when it reduced the salary, benefits, and seniority of teachers who failed to pass the exam and thus became substitutes. Under state and City regulations, the Board was required to revoke the conditional or temporary licenses of teachers who could not obtain permanent licenses, and to demote those teachers to substitute status. Substitute teachers are paid less than licensed teachers, and have no seniority. It was appropriate for the Board to lower teachers' pay and benefits to be consistent with their new positions as substitutes.
The one decision that Plaintiffs cite to support their argument regarding the Core Battery exam, Walston, is inapplicable. In Walston, a school district required fully licensed teachers to pass an exam similar to the Core Battery exam, and fired or demoted those teachers who failed to pass. Walston, 492 F.2d at 921-22. The Court held that the test had not been properly validated for use on fully licensed teachers, and thus that it was unlawful for the school district to fire or demote such teachers because they failed the exam. Id. at 926-27. Here, the Core Battery exam was properly validated for use on temporarily or conditionally licensed teachers, and thus the Board did not violate Title VII by requiring such teachers to pass the exam, and revoking their licenses and reducing their salaries and benefits if they failed to do so.
VII. CONCLUSION
For the reasons set forth above, the Court grants in part and denies in part Defendant's motion to decertify the Plaintiff class. The class survives only as to the declaratory judgment of the Board's liability under Title VII (contained in this Opinion), and injunctive relief benefitting the class as a whole. With respect to the Board's liability, the Court finds that: (1) the Board may be subject to Title VII liability for its use of the LAST; (2) the Board violated Title VII by requiring Plaintiffs to pass the exam in order to receive a permanent teaching license because the LAST was not properly validated; and (3) the Board did not violate Title VII by reducing Plaintiffs' salaries, benefits, and seniority for failing to pass the Core Battery exam.
The Court will hold a status conference in this case on January 10, 2013, at 2:00 p.m. in Courtroom 18B. Plaintiffs are directed to submit a letter to the Court by December 13, 2012 identifying what steps need to be taken in the remedial phase of this action. Defendant is directed to reply to Plaintiffs' submission by December 20, 2012.
SO ORDERED. DATED: New York, New York
December 4, 2012
_____________
KIMBA M. WOOD
United States District Judge
Gulino IV, 460 F.3d at 387.