Amended December 2, 2014 Linda Pippen, on Behalf of Themselves and All Others Similarly Situated v. State

IN THE SUPREME COURT OF IOWA No. 12–0913 Filed July 18, 2014 Amended December 2, 2014 LINDA PIPPEN, et al., On Behalf of Themselves and All Others Similarly Situated, Appellants, vs. THE STATE OF IOWA, et al., and ALL OTHER AGENCIES SIMILARLY SITUATED IN USING THE HIRING AND PROMOTION OF ADMINISTRATIVE SERVICES, Appellees. Appeal from the Iowa District Court for Polk County, Robert J. Blink, Judge. The plaintiffs appeal from an adverse district court judgment after a trial in a class action brought under both the Federal Civil Rights Act and the Iowa Civil Rights Act against the State of Iowa and various executive branch departments, generally alleging that the State unlawfully discriminates against African Americans in employment. AFFIRMED. Thomas A. Newkirk and Leonard E. Bates of Newkirk Law Firm, P.L.C., Des Moines; J. Bryan Wood of Law Office of J. Bryan Wood, Chicago, Illinois; and David H. Goldman and Michael J. Carroll of Babich Goldman, P.C., Des Moines, for appellants. — 2 — Thomas J. Miller, Attorney General, Jeffrey S. Thompson, Deputy Attorney General, and Julia S. Kim and Tyler M. Smith, Assistant Attorneys General, for appellees. Jill R. Gaulding and Lisa C. Stratton, St. Paul, Minnesota, and Mark D. Sherinian of Sherinian & Hasso Law Firm, West Des Moines, for amicus curiae Gender Justice. Kim M. Keenan, Baltimore, Maryland, for amicus curiae National Association for the Advancement of Colored People. Russell E. Lovell II, Des Moines, for amicus curiae Iowa/Nebraska State Conference NAACP. Joshua P. Thompson, Sacramento, California, and Aaron T. Oliver and Jay D. Grimes of Hansen, McClintock & Riley, Des Moines, for amicus curiae Pacific Legal Foundation. — 3 — APPEL, Justice. In this case, we consider an appeal from a district court judgment after a lengthy trial adverse to the plaintiffs in a class action brought under both the Federal Civil Rights Act and the Iowa Civil Rights Act against the State of Iowa and various executive branch departments. The plaintiffs generally allege that the State of Iowa unlawfully discriminates against African Americans in employment. For the reasons expressed below, we affirm the decision of the district court. I. Factual and Procedural Background. There are thirty-seven departments within the executive branch of the State of Iowa. Each exercises its own hiring authority. The State employs a merit hiring system, which establishes “a system of human resource administration based on merit principles and scientific methods to govern the appointment, compensation, promotion, welfare, development, transfer, layoff, removal, and discipline of its civil employees, and other incidents of state employment.” Iowa Code § 8A.411(1) (2007). The Code further directs that “[a]ll appointments and promotions to positions covered by the state merit system shall be made solely on the basis of merit and fitness, to be ascertained by examinations or other appropriate screening methods.” Id. § 8A.411(3). The Iowa Department of Administrative Services (DAS) is responsible for ensuring that hiring decisions are made in accordance with the merit system. See id. § 8A.104(12) (“The director [of DAS] shall . . . [e]xamine and develop best practices for the efficient operation of government and encourage state agencies to adopt and implement these practices.”). DAS is tasked with providing rules for the departments to follow. See id. § 8A.413(1) (DAS adopts rules for the administration of the merit employment system). DAS collects statewide data and — 4 — monitors compliance. In order to comply with the stated goals of the merit system, DAS has a wide range of options, including retaining independent consultants. 1 Upon request, DAS assigns personnel officers as human resource advisors to various departments to assist with employment functions, such as providing materials and training, helping develop screening tools, and assisting with hiring. Applicants to executive branch positions, as well as current employees applying for promotions, submit applications to DAS, either online or by hard copy. DAS maintains electronic data on every applicant and application in their database, the BrassRing. 2 The district court summarized the hiring system as employing three separate decision-making steps: (1) “DAS receives applications for merit-covered job posting, screens those applications for basic eligibility of the job classification, and refers eligible applicants to the hiring department” (emphasis omitted) (referral); (2) “the hiring department screens the referred applicants for the job-title specific requirements, determines which candidates to interview” (interview selection); and (3) “the hiring department interviews the selected candidates and decides which candidate to offer the job” (hire or promotion). Although all departments follow the general practices of the merit system, their practices in the hiring process vary. These varied practices include: using a second résumé screen, requiring candidates to more 1Violation of the human-resources subchapter of Iowa Code chapter 8A or DAS’s regulations is a simple misdemeanor. Iowa Code § 8A.458. Further, “[t]he director may institute and maintain any action or proceeding at law or in equity that the director considers necessary or appropriate to secure compliance with this subchapter and the rules and orders under this subchapter.” Id. § 8A.453(1). 2DAS converted to the BrassRing system between 2004 and 2006. Before this, the State used the AS-400 system. — 5 — fully explain how their experiences qualify them for a specific job function, or requiring a typing test. Each department maintains data relating to each applicant, which is stored in paper hiring files, unlike the DAS data system, which is electronic. Each paper hiring file contains a BrassRing registration number so a correlation between a specific job posting and the applicant’s performance on the screening devices and/or interview records can be correlated. In this case, fourteen 3 African-American plaintiffs brought a lawsuit under both Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C. § 2000e–2000e-17 (2006), and the Iowa Civil Rights Act of 1965, as amended, Iowa Code chapter 216. In their petition, the plaintiffs alleged that the State of Iowa, including the thirty-seven different executive branch departments, engaged in practices that resulted in a failure to maintain a diverse, nondiscriminatory workplace through its merit employment system. The plaintiffs contend that because of the State’s failure to enforce extant statutory and regulatory policies, a disproportionate number of African Americans were denied an equal opportunity for employment. They claim this was the natural unintended consequences of the State’s failure to follow rules designed to ensure equal opportunity in the workplace and was not done intentionally or with malice. Further, the plaintiffs alleged that in May of 2006 they provided the State of Iowa with a document entitled “Initial Evidentiary Report,” alleging systemic racial bias and a pattern of retaliation by top managers and officials of the State of Iowa. The plaintiffs further alleged that the 3The plaintiffs’ lawsuit was filed in October 2007 and subsequently amended three times, adding nine additional plaintiffs, for a total of twenty-three named plaintiffs. — 6 — State hired a consultant to study employment practices in late 2006 or early 2007 who produced a report known as the CPS Report. The plaintiffs alleged that the Initial Evidentiary Report and the CPS Report put the State on notice that the hiring practices of the State imposed barriers to equal employment opportunities for African Americans. On September 28, 2010, on stipulation of the parties, the district court ordered certification of the case as a class action. The class definition and class claim were: CLASS DEFINITION: All African American applicants or employees who sought appointment to or held a meritsystem position with an Executive Branch agency (not including Board of Regents) at any point from July 1, 2003 through [date of Court’s decision regarding liability]. CLASS CLAIM: Disparate Impact or Adverse Impact discrimination with respect to hiring and promotion decisions and/or unequal terms and conditions of employment associated with those decisions under Title VII and the Iowa Civil Rights Act arising from subjective, discretionary decision-making permitted by the State’s abdication of statutory or regulatory responsibilities and obligations and/or failure to follow its own policies. The case came to trial on September 12, 2011. The plaintiffs offered evidence relating to the efforts of the State to document its employment practices, expert testimony by a statistical expert, labor economist Mark Killingsworth, social science testimony from psychology professors Anthony Greenwald and Cheryl Kaiser, testimony from DAS representatives and personnel, and anecdotal testimony from various plaintiffs related to their experience with state government. In support of the plaintiffs’ claims, Killingsworth testified that based on his statistical work employing conventional and probit — 7 — regression analysis 4 statistical procedures, African Americans were treated differently and more disadvantageously than whites with respect to the referral of applications by DAS for interviews, with respect to the selection for interviews by various agencies and departments, and with respect to ultimate hiring. Further, he opined that once hired, African Americans have lower salaries within a given job title or are hired for job titles that pay less than others, and were treated differently in performance evaluations. In making his calculations, Killingsworth only analyzed applicants who had been deemed by DAS to meet the minimum qualifications for the job classification and had been referred to departments. He approached the data from a variety of perspectives, as his analytical models could include or exclude different variables. Regarding the separability of the elements in the hiring process, Killingsworth testified: [I]t’s not that it’s incapable of being separated, but I think there are very serious questions about whether it can reliably be separated, which is a different story. Mechanically, one could certainly separate it. And I know [this] because [the State’s] experts have done [it.] The plaintiffs offered social science evidence through two psychology professors: Anthony Greenwald and Cheryl Kaiser. Greenwald’s field of study is implicit social cognition, a phrase which he introduced in a coauthored article in 1995. According to Greenwald, implicit bias, also known as hidden or unconscious bias, is a person’s 4The district court described the difference between conventional and probit regression analysis as follows: a [conventional] regression analysis seeks to predict or forecast how a dependent variable might change based upon changes in one or more independent variables. The probit analysis differs primarily in that the dependent value in that context may only have one of two values. — 8 — automatic preference for one race over another. He asserted that it was possible that implicit bias affected Iowa decision-makers in this case, although he did not review any of the hiring files, nor any specific employment decisions relating to any class members. He could not rule out other race-neutral causes for the statistical imbalance in the State’s hiring system. In his opinion, even in the best case scenario, bias could still unconsciously invade the State’s hiring process. Kaiser studies stereotyping and prejudice and their effects on decision-making. She testified that she viewed implicit bias as pervasive and believed all people fall within a spectrum with explicit bias on one end and limited implicit bias on the other. She opined that training and accountability, including recordkeeping, are means of reducing implicit bias and, if used more extensively by the State, would have a positive effect on reducing the implicit bias in the State system. Additionally, several class members, referring to documents relating to their applications for hire or promotion, testified at trial and offered examples in which, they claim, the hiring system did not function as intended. These examples included: a qualified African-American applicant who was not referred to a department by DAS, due to DAS incorrectly reporting the applicant was not qualified; in some cases, résumés of African Americans were marked to highlight spelling and grammatical errors; and some hiring files contained African-Americans’ résumés, but not the screening devices used to score or evaluate them. The plaintiffs buttressed the testimony of their witnesses with the CPS Report, a review of the State’s hiring practices commissioned by the State, prepared by human resources consultants in response to expressed concerns about racial discrimination in state employment. The CPS Report found, among other things, that during fiscal years — 9 — 2004–2006, qualified minority applicants were interviewed less frequently than qualified white applicants (13.46% for minorities and 20.24% for whites), figures which the authors noted “may support the perception of discriminatory hiring practices.” Further, the report noted that while African Americans constituted six percent of the total qualified pool, they represented no more than 2.8% of the total hires for fiscal years 2004–2006. By comparison, whites represented eighty-eight percent of the qualified applicants and ninety-one percent of the total hires. The CPS Report noted that the statistical difference in employment appeared to arise from the process between the referral step and the interview step (African Americans were reduced from 5.95% referred to 3.47% interviewed of the total applicant pool for fiscal years 2004–2006 combined) and the process between the interview step and the hire decision (African Americans were reduced to 2.82% while whites increased to 91.52%). According to the CPS Report, “The actual personnel decisions may create a rebuttable inference of adverse impact.” It recommended that DAS “institute a policy of regular and systematic oversight . . . to ensure compliance with [required] policies and procedures.” The CPS Report cautioned, however, that because the State’s application tracking system (the BrassRing) did not track individual people, but rather applications, it was difficult to identify with any certainty the exact makeup of the applicant pool or the actual number of applicants. The State offered evidence related to the decision-making process in state government. It also offered the testimony of economist Robert Miller, who was tasked by the State with analyzing Killingsworth’s findings and examining the employment outcomes in Iowa state government to determine if African Americans were systemically — 10 — disadvantaged. Miller found Killingsworth’s reports to be incomplete and his conclusions not well-founded. He testified that, in his opinion, there was no statistically significant evidence of system wide racial discrimination in the merit employment system in the State of Iowa. Miller also testified that it was possible for the plaintiffs to break down the aggregate analysis into more discrete consideration of employment decisions by department or by other classifications. On April 17, 2012, the district court filed a detailed and thoughtful fifty-six page decision in favor of the State. The district court first noted that with regard to the plaintiffs’ first theory, even assuming that “the components of the decision-making process in this case are not capable of being separated, [the] Plaintiffs have failed to provide legal authority for concluding that ‘abdication of statutory or regulatory responsibilities and obligations and/or failure to follow its own policies’ is a particular employment practice.” Next, in regards to the plaintiffs’ second theory, the court found the plaintiffs had not carried their burden of “demonstrating the inseparability of the employment system components for analytical purposes.” The court concluded “[t]he former theory fail[ed] as a matter of law; the latter as a matter of fact.” Alternatively, looking to the plaintiffs’ statistical and implicit bias evidence, the district court noted that the plaintiffs failed to prove the causation element of their disparate impact claim. The plaintiffs appealed. II. Standard of Review. In this appeal of a trial to the court, the standard of review on all issues is for correction of errors at law and for findings of fact not supported by substantial evidence. Iowa R. App. P. 6.907; Falczynski v. Amoco Oil Co., 533 N.W.2d 226, 230 (Iowa 1995). “Evidence is — 11 — substantial for purposes of sustaining a finding of fact when a reasonable mind would accept it as adequate to reach a conclusion.” Falczynski, 533 N.W.2d at 230 . We view the substantiality of evidence in the light most favorable to upholding the trial court’s judgment. Id.; Fuller v. Iowa Dep’t of Human Servs., 576 N.W.2d 324, 328 (Iowa 1998). Reversal is required when an error of law or fact materially affects other findings or rulings. See Falczynski, 533 N.W.2d at 230 . “[W]hen the trial court following a bench trial has denied recovery because a party failed to sustain its burden of proof on an issue, we will not interfere with the trial court’s judgment unless we find the party has carried its burden as a matter of law.” Id.; see Vincent v. Four M Paper Corp., 589 N.W.2d 55, 62 (Iowa 1999). “We will conclude a party has carried such a burden only when evidence is so overwhelming that only one reasonable inference on each critical fact issue can be drawn.” Falczynski, 533 N.W.2d at 230 . We are not bound by the trial court’s application of legal principles or its conclusions of law. Fuller, 576 N.W.2d at 328 . “When the trial court has applied erroneous rules of law which materially affected its decision, we will reverse.” Falczynski, 533 N.W.2d at 230 . III. Overview of Legal Framework Established by Modern State and Federal Civil Rights Acts. A. Context of State and Federal Legislation. The issues raised in this case cannot be approached without consideration of the larger context in which they arise. The legacy of slavery and Jim Crow may be in the past, but their effects cast a shadow into the present. Specifically, African Americans continue to be underrepresented in many categories of employment. While the days of “Whites Only Need Apply” signage are fortunately long passed, institutional barriers to equality of economic — 12 — opportunity remain intractable. See Susan Sturm, Second Generation Employment Discrimination: A Structural Approach, 101 Colum. L. Rev. 458 , 459–60 (2001) (“Smoking guns—the sign on the door that ‘Irish need not apply’ or the rejection explained by the comment that ‘this is no job for a woman’—are largely things of the past. . . . Cognitive bias, structures of decision making, and patterns of interaction have replaced deliberate racism and sexism as the frontier of much continued inequality.”); see also generally Melissa Hart, Subjective Decisionmaking and Unconscious Discrimination, 56 Ala. L. Rev. 741 (2005); Audrey J. Lee, Note, Unconscious Bias Theory in Employment Discrimination Litigation, 40 Harv. C.R.-C.L. L. Rev. 481 (2005). The remedies afforded under civil rights legislation disparate impact analysis are a critical component in eliminating barriers or headwinds faced by African Americans in the employment marketplace. The purposes of both the Iowa Civil Rights Act and the Federal Civil Rights Act are designed to address these ongoing problems. The United States Supreme Court has declared that the primary purpose of Title VII of the Civil Rights Act of 1964 is “ ‘to assure equality of employment opportunities and to eliminate those discriminatory practices and devices which have fostered racially stratified job environments to the disadvantage of minority citizens.’ ” Int’l Bhd. of Teamsters v. United States, 431 U.S. 324, 348 , 97 S. Ct. 1843, 1861 , 52 L. Ed. 2d 396, 423 (1977) (quoting McDonnell Douglas Corp. v. Green, 411 U.S. 792, 800 , 93 S. Ct. 1817, 1823 , 36 L. Ed. 2d 668, 676 (1973)); see Connecticut v. Teal, 457 U.S. 440 , 448–49, 102 S. Ct. 2525 , 2531–32, 73 L. Ed. 2d 130 , 137–38 (1982) (explaining Title VII’s purposes). Similarly, the Iowa Civil Rights Act was enacted “in an effort to establish parity in — 13 — the workplace and market opportunity for all.” Vivian v. Madison, 601 N.W.2d 872, 873 (Iowa 1999). B. Historical Development of Disparate Impact. There are two distinct theories of liability under civil rights laws for discrimination in employment, namely, cases involving disparate treatment and cases involving disparate impact. See Int’l Bhd. of Teamsters, 431 U.S. at 335 n.15, 97 S. Ct. at 1854 n.15, 52 L. Ed. 2d at 415 n.15. In a disparate treatment case, the plaintiff bears the burden of showing he or she has been harmed by discriminatory animus of the employer. See id. at 357 , 97 S. Ct. at 1866 , 52 L. Ed. 2d at 429 . Proving discriminatory animus is often a difficult task as it involves probing the subjective motivations of the decision-maker. Although cases of blatant racism still exist, most discrimination is more subtle and difficult to demonstrate. In the alternative, however, a civil rights claim may be brought based on disparate impact. In a disparate impact case, what matters is not the subjective motivation of the employer, but the effects of an employment practice. See Teal, 457 U.S. at 447 n.8, 102 S. Ct. at 2531 n.8, 73 L. Ed. 2d at 137 n.8 (“Experts familiar with the subject now generally describe the problem in terms of ‘systems’ and ‘effects’ rather than simply intentional wrongs.” (Internal citations omitted.)); Int’l Bhd. of Teamsters, 431 U.S. at 335 n.15, 97 S. Ct. at 1854 n.15, 52 L. Ed. 2d at 415 n.15 (“Proof of discriminatory motive . . . is not required under a disparate-impact theory.”); Griggs v. Duke Power Co., 401 U.S. 424, 432 , 91 S. Ct. 849, 854 , 28 L. Ed. 2d 158, 165 (1971) (noting “good intent or absence of discriminatory intent does not redeem employment procedures or testing mechanisms that operate as ‘built-in headwinds’ for minority groups and are unrelated to measuring job capability”). — 14 — It is sometimes asserted that disparate impact analysis of civil rights claims is outside the “core” of civil rights statutes and represents a novel legal development. See Stewart J. Schwab & Steven L. Willburn, Reasonable Accommodation of Workplace Disabilities, 44 Wm. & Mary L. Rev. 1197 , 1201 (2003) (characterizing disparate impact cases as “noncore cases of discrimination under Title VII”). But this is at least somewhat misleading. Concern about institutional barriers to equal opportunity in employment predated civil rights statutes and can be seen at the beginning of the modern civil rights movement. For example, President Roosevelt issued Executive Order 8802 in June of 1941, which prohibited discrimination by race by private employers engaged in government contracting and created a Fair Employment Practices Commission which monitored broad trends, pushed and cajoled employers in the war industries, and assessed the bottom line in terms of overall progress. See Susan D. Carle, How Myth-Busting About the Historical Goals of Civil Rights Activism Can Illuminate Future Paths, 7 Stan. J. C.R. & C.L. 167 , 172–73 (2011); see also Exec. Order No. 8802, 6 Fed. Reg. 3109 (June 25, 1941). Disparate impact claims may be complex and complicated, but they are not disfavored. C. Treatment of Disparate Impact Analysis Under Title VII of the Civil Rights Act of 1964 by the United States Supreme Court. 1. Introduction. Although federal law is not controlling on state law questions, we begin substantive discussion of disparate impact analysis with an overview of cases of the United States Supreme Court. The reason for this is simple: in a series of disparate impact cases, the Supreme Court has developed doctrine in both majority and dissenting opinions in considerable detail. Further, one of the claims in this case was brought under federal law. On the federal law claim, of course, the — 15 — decisions of the United States Supreme Court constitute binding authority which we must faithfully apply in our interpretation of federal law. With respect to the state law claim, the reasoning of the United States Supreme Court opinions, and the dissenting opinions, may well be persuasive, although it is certainly not binding upon us. As a result, understanding the range of interpretive options for state courts in interpreting state law can be enhanced by analysis of majority and dissenting opinions of the United States Supreme Court. 2. Griggs: A unanimous court’s broad construction of the Federal Civil Rights Act. The first decision of the United States Supreme Court which considered a case based on disparate impact was Griggs, 401 U.S. at 424 , 91 S. Ct. at 849 , 28 L. Ed. 2d at 158 . In Griggs, the Supreme Court considered a class action alleging that the requirement of a high school education or passing a standardized general intelligence test as a condition of employment violated the Federal Civil Rights Act. Id. at 425– 26, 91 S. Ct. at 851 , 28 L. Ed. 2d at 161 . In Griggs, the plaintiff asserted that neither “standard [was] shown to be significantly related to job performance,” that both standards operated to disqualify African Americans at a substantially higher rate than white applicants, and that the jobs in question had been previously filled by whites only as a result of long-standing practice. Id. at 426 , 91 S. Ct. at 851 , 28 L. Ed. 2d at 161 . A unanimous Supreme Court found for the plaintiffs. As noted by Chief Justice Burger, “[P]ractices, procedures, or tests neutral on their face, and even neutral in terms of intent, cannot be maintained if they operate to ‘freeze’ the status quo of prior discriminatory employment practices.” Id. at 430 , 91 S. Ct. at 853 , 28 L. Ed. 2d at 163 . In much quoted language, Chief Justice Burger noted that the Federal Civil Rights — 16 — Act “proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation.” Id. at 431 , 91 S. Ct. at 853 , 28 L. Ed. 2d at 164 . Further, the Chief Justice noted that “good intent or absence of discriminatory intent does not redeem employment procedures or testing mechanisms that operate as ‘built-in headwinds’ for minority groups and are unrelated to measuring job capability.” Id. at 432 , 91 S. Ct. at 854 , 28 L. Ed. 2d at 165 . Again, the Chief Justice noted that “Congress directed the thrust of [the Federal Civil Rights Act] to the consequences of employment practices, not simply the motivation.” Id. (emphasis omitted). Griggs clearly established that a civil rights claim could be based on disparate impact without proving discriminatory animus or motivation in cases involving objective standardized tests or employment criteria. See id. at 436 , 91 S. Ct. at 856 , 28 L. Ed. 2d at 167 . But what about a claim that the exercise of subjective discretion of supervisory employees has produced illegal discrimination? 3. Watson: The court divided. The first United States Supreme Court case to consider a federal civil rights claim based upon a subjective decision-making process was Watson v. Fort Worth Bank & Trust, 487 U.S. 977 , 108 S. Ct. 2777 , 101 L. Ed. 2d 827 (1988). In that case, an African American employee of a bank alleged that she had been rejected in favor of white applicants for four supervisory positions at the bank. Id. at 982 , 108 S. Ct. at 2782 , 101 L. Ed. 2d at 837 . In Watson, all participating members of the Supreme Court held that a claim could be brought based upon the exercise of subjective discretion, but the court split sharply on the contours and scope of such a disparate impact claim. Compare id. at 991–99, 108 S. Ct. at 2787–91, 101 L. Ed. 2d at 843–48 (plurality opinion), with id. at 1000–11, 108 S. Ct. at 2792–98, — 17 — 101 L. Ed. 2d at 849–56 (Blackmun, J., concurring in part and concurring in judgment). Speaking for four members of the Court, Justice O’Connor laid out the stark alternatives presented by the parties. According to the plaintiffs, if disparate impact analysis were confined to objective tests, an employer would be able to simply substitute subjective criteria having substantially identical effects. Id. at 989 , 108 S. Ct. at 2786 , 101 L. Ed. 2d at 841 (plurality opinion). If so, Griggs would be a dead letter. Id. On the other hand, according to the defendants, recognizing a claim of disparate impact in a subjective selection process would make the claims so impossibly difficult to defend that employers would be forced to adopt numerical quotas in order to avoid liability. Id. at 989 , 108 S. Ct. at 2786 , 101 L. Ed. 2d at 842 . Justice O’Connor seemed to agree with the arguments of both parties. In section IIB of her opinion, which was joined by all members of the Court, she recognized that Griggs “could largely be nullified if disparate impact analysis were applied only to standardized selection practices.” Id. She further wrote that disparate impact analysis is in principle “no less applicable to subjective employment criteria than to objective or standardized tests.” Id. at 990 , 108 S. Ct. at 2786 , 101 L. Ed. 2d at 842 . In addition, Justice O’Connor observed that while an employer’s policy of leaving promotion decisions to unchecked discretion of lower level supervisors should itself raise no inference of discriminatory conduct, it does not follow that “supervisors to whom this discretion is delegated always act without discriminatory intent.” Id. In addition, in an observation of particular interest in this case, Justice O’Connor noted that even without overt discriminatory intent, “the — 18 — problem of subconscious stereotypes and prejudices would remain.” Id. at 990 , 108 S. Ct. at 2787 , 101 L. Ed. 2d at 842 . Justice O’Connor then turned to the employer’s concern in parts IIC and IID of her opinion. Now writing for only four members of the Court, she pivoted to express concerns about the use of “bare statistics” in a subjective decision-making case that an employer could rebut only upon a showing of “business necessity” or “job relatedness.” Id. at 991– 93, 108 S. Ct. at 2787–88, 101 L. Ed. 2d at 843–44 (citations and internal quotation marks omitted). She expressed concern that employers would find it difficult to validate subjective selection criteria and impossible to defend and, as a result, would engage in a surreptitious quota system. Id. at 992–93, 108 S. Ct. at 2787–88, 101 L. Ed. 2d at 843–44. She observed that it would be “completely unrealistic to assume that unlawful discrimination is the sole cause of people failing to gravitate to jobs and employers in accord with the laws of chance.” Id. at 992 , 108 S. Ct. at 2787 , 101 L. Ed. 2d at 843 . She further noted that “[i]t would be equally unrealistic to suppose that employers can eliminate, or discover . . ., the myriad of innocent causes that may lead to statistical imbalances” in the workplace. Id. To avoid impossible defenses and surreptitious quotas, Justice O’Conner went well beyond the question posed in the petition for writ of certiorari 5 to undertake what she called a “fresh and somewhat closer 5The question posed in the Petition for Writ of Certiorari was: “Is the racially adverse impact of an employer’s practice of simply committing employment decisions to the unchecked discretion of a white supervisory corps subject to the test of Griggs v. Duke Power Co., 401 U.S. 424 [ 91 S. Ct. 849 , 28 L. Ed. 2d 158 ] (1971)?” 487 U.S. at 1011 , 108 S. Ct. at 2797 , 101 L. Ed. 2d at 856 (Stevens, J., concurring). Justice Stevens, in a separate opinion concurring with the judgment, considered it unwise for the court to engage in a “fresh” interpretation of prior cases in light of the narrow question presented. See id. at 1011 , 108 S. Ct. at 2797–98, 101 L. Ed. 2d at 856 . — 19 — examination” of the evidentiary standards that apply in disparate impact cases. Id. at 994 , 108 S. Ct. at 2788 , 101 L. Ed. 2d at 844 . She began by emphasizing that a plaintiff must identify a “specific employment practice” that is challenged. Id. at 994 , 108 S. Ct. at 2788 , 101 L. Ed. 2d at 845 . She then turned to causation. Id. at 994–95, 108 S. Ct. at 2789 , 101 L. Ed. 2d at 845 . She emphasized that statistical disparities must be “sufficiently substantial that they raise . . . an inference of causation.” Id. at 995 , 108 S. Ct. at 2789 , 101 L. Ed. 2d at 845 . In a footnote, Justice O’Connor noted that lower courts have sometimes looked to the EEOC’s Uniform Guidelines on Employee Selection Procedures and adopted an enforcement rule that an inference of discrimination could not be drawn unless members of a particular race, sex or ethnic group are selected at a rate less than four-fifths of the group with the highest selection rating. Id. at 995 n.3, 108 S. Ct. at 2789 n.3, 101 L. Ed. 2d at 845 n.3. Justice O’Connor noted that this method “has not provided more than a rule of thumb for the courts.” Id. Justice O’Connor next cautioned that courts should not assume “that plaintiffs’ statistical evidence is reliable.” Id. at 996 , 108 S. Ct. at 2790 , 101 L. Ed. 2d at 846 . According to Justice O’Connor, weaknesses can include small or incomplete data sets, inadequate statistical techniques, and applicant pools “containing individuals lacking minimal qualifications.” Id. at 996–97, 108 S. Ct. at 2790 , 101 L. Ed. 2d at 846 . Justice O’Connor thus stressed that in disparate impact cases, employers have the opportunity to attack the quality of the plaintiff’s statistical evidence and the inferences that may be drawn from it. Id. at 996–97, 108 S. Ct. at 2790 , 101 L. Ed. 2d at 846–47. Justice O’Connor next turned to the nature of the business necessity defense. Id. at 997–98, 108 S. Ct. at 2790–91, 101 L. Ed. 2d at — 20 — 847. Although Griggs stated that the burden of showing business necessity rested with the defendant, Justice O’Connor wrote that the burden of proving discrimination always rests with the plaintiff at all times. Id. at 997 , 108 S. Ct. at 2790 , 101 L. Ed. 2d at 847 . According to Justice O’Connor, the plaintiff now had the burden of showing other tests or selection devices would serve the employer’s legitimate interest. Id. at 998 , 108 S. Ct. at 2790 , 101 L. Ed. 2d at 847 . Justice Blackmun, joined by two colleagues, took exception to Justice O’Connor’s notion that the burden of proof and production in disparate impact cases remained with the plaintiff on the business necessity defense. Id. at 1002–03, 108 S. Ct. at 2793 , 101 L. Ed. 2d at 850 (Blackmun, J., concurring in part and concurring in the judgment). Justice Blackmun maintained that in disparate impact cases, a prima facie case is established by showing a significant statistical disparity. Id. at 1004 , 108 S. Ct. at 2794 , 101 L. Ed. 2d at 851 . Once an employment practice is shown to have discriminatory consequences, according to Justice Blackmun, an employer can escape liability only if it persuades the court that the selection process has “a manifest relationship to the employment in question.” Id. (internal quotation marks omitted). Even if such a relationship is present, according to Justice Blackmun, the plaintiff may show that “other selection processes that have a lesser discriminatory effect could also serve . . . the employer’s [legitimate] business needs.” Id. at 1005–06, 108 S. Ct. at 2795 , 101 L. Ed. 2d at 852 . Justice Blackmun was also concerned about language in Justice O’Connor’s opinion suggesting that “[i]n the context of subjective or discretionary employment decisions, the employer will often find it easier than in the case of standardized tests to produce evidence of a manifest — 21 — relationship to the employment in question.” Id. at 1006 , 108 S. Ct. at 2795 , 101 L. Ed. 2d at 853 (internal quotation marks omitted). Justice Blackmun asserted that “[a]llowing an employer to escape liability simply by articulating vague, inoffensive-sounding subjective criteria would [do a disservice to the federal statute]’s goal of eradicating discrimination in employment.” Id. at 1009 , 108 S. Ct. at 2797 , 101 L. Ed. 2d at 855 . In sum, the Watson opinions clearly stood for the proposition that disparate impact could, at least in some circumstances, apply to subjective employer decision-making. An evenly divided court, however, had different visions of the scope and contour of disparate impact analysis on subjective decision-making. The plurality, joining Justice O’Connor, was prepared to modify the burdens of proof in order to enhance the ability of an employer to defend disparate impact claims arising from subjective decision-making, while Justice Blackmun feared the modification of law proposed by Justice O’Connor would provide an escape hatch for employers from potential liability. 4. Wards Cove: Narrow construction prevails. A year after Watson was decided, the Supreme Court decided Wards Cove Packing Co. v. Atonio, 490 U.S. 642 , 109 S. Ct. 2115 , 104 L. Ed. 2d 733 (1989), superseded by statute on other grounds, 42 U.S.C. § 2000e–2(k), as recognized in Wal-Mart Stores, Inc. v. Dukes, 564 U.S. __ , 131 S. Ct. 2541 , 180 L. Ed. 2d 374 (2011). In that case, Justice Kennedy, who did not participate in Watson, tipped the balance. In Wards Cove, Justice Kennedy joined an opinion by Justice White which essentially converted the approach of the plurality opinion of Justice O’Connor in Watson into a majority opinion in Wards Cove. Wards Cove dealt with employment practices of two companies that operated salmon canneries in remote areas of Alaska during the — 22 — salmon runs in the summer months. Id. at 646, 109 S. Ct. at 2119 , 104 L. Ed. 2d at 744 . Jobs at the canneries fell into two general categories, “cannery jobs” and “noncannery jobs.” Id. at 647 , 109 S. Ct. at 2119 , 104 L. Ed. 2d at 745 . Most cannery jobs were nonskilled positions, while, conversely, most noncannery jobs were classified as skilled positions. Id. The cannery jobs were filed predominantly by nonwhites, while the noncannery jobs were filled predominantly with white workers. Id. at 647 , 109 S. Ct. at 2119–20, 104 L. Ed. 2d at 745 . The canneries operated separate dormitories and separate mess halls for the cannery and noncannery workers. Id. at 647 , 109 S. Ct. at 2120 , 104 L. Ed. 2d at 745 . The district court found in favor of the defendants on all disparate impact claims, and a panel of the Court of Appeals for the Ninth Circuit affirmed, but an en banc hearing of the Ninth Circuit held that “[o]nce the plaintiff . . . has shown disparate impact caused by specific, identifiable employment practices or criteria, the burden shifts to the employer.” Id. at 648 , 109 S. Ct. at 2120 , 104 L. Ed. 2d at 746 (internal citations omitted). On remand to the original panel, the court held that the plaintiffs had made out a prima facie case of disparate impact in hiring for both skilled and unskilled noncannery positions and remanded the case to the district court to determine if the employer could meet its burden of showing business necessity. Id. at 649 , 109 S. Ct. at 2120 , 104 L. Ed. 2d at 746 . Because the case raised issues upon which the Court was evenly divided in Watson, the Supreme Court granted certiorari. Id. at 649–50, 109 S. Ct. at 2121 , 104 L. Ed. 2d at 747 . As previously mentioned, the majority opinion by Justice White in Wards Cove basically converted Justice O’Connor’s plurality opinion in Watson into a majority opinion. Compare Watson, 487 U.S. at 982–1000, — 23 — 108 S. Ct. at 2782–92, 101 L. Ed. 2d at 837–48 (plurality opinion), with Wards Cove, 490 U.S. at 645–61, 109 S. Ct. at 2118–27, 104 L. Ed. 2d at 744–54. The conversion of Justice O’Connor’s views from plurality opinion to precedent drew a sharp rejoinder from the minority. Justice Blackmun, ruing the outcome, declared that “a bare majority of the Court takes three major strides backwards in the battle against race discrimination.” Wards Cove, 490 U.S. at 661 , 109 S. Ct. at 2127 , 104 L. Ed. 2d at 754 (Blackmun, J., dissenting). He questioned “whether the majority still believes that race discrimination—or, more accurately, race discrimination against nonwhites—is a problem in our society, or even remembers that it ever was.” Id. at 662 , 109 S. Ct. at 2127 , 104 L. Ed. 2d at 755 . Justice Stevens’ dissent emphasized the role of federal courts and agencies in promoting the national goal of “eliminating barriers that define economic opportunity not by aptitude and ability but by race, color, national origin, and other traits that are easily identified but utterly irrelevant to one’s qualification for a particular job.” Id. at 662– 63, 109 S. Ct. at 2128 , 104 L. Ed. 2d at 755 (Stevens, J., dissenting). According to Justice Stevens, “The changes the majority makes today, tipping the scales in favor of employers, are not faithful to [established disparate impact] principles.” Id. at 673 , 109 S. Ct. at 2133 , 104 L. Ed. 2d at 762 . Even for an often divided Supreme Court, the holdings in Wards Cove can only be characterized as bitterly contested. It should be emphasized that nothing in the language of Title VII compelled the result in Wards Cove or the position of the dissents. Instead, as one commentator has noted, the battle over proper interpretation of open-ended language of Title VII was over understandings about “whether discrimination is still happening” in the modern workplace, about “how it manifests itself,” and about how society — 24 — should address such concerns. See Sandra F. Sperino, Revitalizing State Employment Discrimination Law, 20 Geo. Mason L. Rev. 545 , 546 (2013) [hereinafter Sperino, Revitalizing]. It seems fair to say that the majority on the Supreme Court saw racial discrimination in employment as primarily a relic of the past that does not require broad remedial measures, while the minority saw racial discrimination more like an intractable and enduring part of the American landscape. 5. The Civil Rights Act of 1991: Congress reacts to narrow construction by the Supreme Court. Because it was decided over two decades ago, it is easy to forget the controversy that Wards Cove engendered. Civil rights advocates were outraged by the decision and other decisions of what seemed to be an increasingly hostile Supreme Court. The decision in Wards Cove was compared to the 1883 Civil Rights Cases and was said to foretell the end of the Second Reconstruction that commenced with Brown v. Board of Education, 347 U.S. 483 , 74 S. Ct. 686 , 98 L. Ed. 873 (1954), and the passage of the Civil Rights Act of 1964. See Robert Belton, The Dismantling of the Griggs Disparate Impact Theory and the Future of Title VII: The Need for a Third Reconstruction, 8 Yale L. & Pol’y Rev. 223, 247–49 (1990). In response, Congress passed legislation overruling Wards Cove and other 1989 Supreme Court rulings in the Civil Rights Act of 1990. S. 2104, 101st Cong. (1990); see 136 Cong. Rec. S991–01 (1990) (statement of Sen. Kennedy) (“In the past year, however, the Supreme Court has issued a series of rulings that mark an abrupt and unfortunate departure from its historic vigilance in protecting civil rights. ... The Civil Rights Act of 1990 is intended to overturn these Court decisions and restore and strengthen these basic laws.”); see also Philip S. Runkel, Note, The Civil Rights Act of 1991: A Continuation of the — 25 — Wards Cove Standard of Business Necessity?, 35 Wm. & Mary L. Rev. 1177 , 1177 n.5, 1186 (1994) 6 [hereinafter Runkel] (“Against this background, Congress attempted to overturn Wards Cove with a new civil rights bill in 1990.”). President Bush, however, vetoed the measure. 136 Cong. Rec. S16,562 (1990) (recording President Bush’s veto). The Senate attempted to override the veto, an effort which failed to meet the two thirds vote required in the Senate by one vote. 136 Cong. Rec. S16,589 (1990) (recording final tally of 66 to 34). Although very large margins in both houses wished to overturn Wards Cove, the effort failed by one vote. One provision of the Civil Rights Act of 1990 is particularly noteworthy for the discussion in this case. The vetoed bill expressly addressed the situation where an employer fails to keep sufficient 6Footnote 5 states: Civil rights advocates wanted to overturn five Supreme Court decisions that worked to restrict employees’ ability to successfully sue employers over workplace discrimination. The most important of these Supreme Court decisions was Wards Cove Packing Co. v. Atonio, 490 U.S. 642 (1989). . . . Other important cases included: Patterson v. McLean Credit Union, [ 491 U.S. 164, 188 , 109 S. Ct. 2363, 2379 , 105 L. Ed. 2d 132, 158 (1989)] (holding that discrimination in the performance of employment contracts is not prohibited explicitly under existing federal law); Lorance v. AT&T Technologies, Inc., [ 490 U.S. 900, 911 , 109 S. Ct. 2261, 2268 , 104 L. Ed. 2d 961, 975 (1989)] (limiting the previous interpretation of federal law regarding the ability of workers to challenge discriminatory seniority systems); Martin v. Wilks, [ 490 U.S. 755, 759 , 109 S. Ct. 2180, 2183 , 104 L. Ed. 2d 835, 842 (1989)] (expanding the ability of workers not affected by discrimination to challenge agreements made between previously discriminatory employers and the discriminated party); and Price Waterhouse v. Hopkins, [ 490 U.S. 228, 258 , 109 S. Ct. 1775, 1795 , 104 L. Ed. 2d 268, 293 (1989)] (holding that employment decisions based on both discriminatory and nondiscriminatory reasons may be valid if the employer proves it would have made the same decision based solely on the non-discriminatory factors). Runkel, 35 Wm. & Mary L. Rev. at 1177 n.5. — 26 — records to allow for disparate impact analysis. The relevant provision stated: (iii) the complaining party shall be required to demonstrate which specific practice or practices are responsible for the disparate impact in all cases unless the court finds after discovery (I) that the respondent has destroyed, concealed or refused to produce existing records that are necessary to make this showing, or (II) that the respondent failed to keep such records . . . S. 2104, 101st Cong. § 4(k)(B)(iii). A group of moderate Republican Senators, however, determined that notwithstanding the failure to override the President’s veto, a compromise could be struck between Congress and the President. See Runkel, 35 Wm. & Mary L. Rev. at 1198. In the end, after an intense period of diplomacy between warring factions, Congress passed the Civil Rights Act of 1991. Pub. L. No. 102-166, 105 Stat. 1071 (1991) (codified at 42 U.S.C. § 2000e to 2000e-16 (Supp. III 1991)). According to the statute, the purpose of the Act was “to codify the concepts of ‘business necessity’ and ‘job related’ enunciated by the Supreme Court in Griggs v. Duke Power Co., . . . and in other Supreme Court decisions prior to Wards Cove Packing Co.” and to “respond to recent decisions of the Supreme Court by expanding the scope of relevant civil rights statutes in order to provide adequate protection to victims of discrimination.” Id. § 3(2), (4). The Act also addressed the Wards Cove ruling regarding the general requirement of identification of particular employment practices. Id. § 105(a) (codified at 42 U.S.C. § 2000e-2(k)(1)). While Congress generally required that a plaintiff identify particular employment practices that cause disparate impact, Congress also provided that the decision-making process could be challenged as a whole under certain circumstances. Specifically, Congress provided: — 27 — With respect to demonstrating that a particular employment practice causes a disparate impact . . . the complaining party shall demonstrate that each particular challenged employment practice causes a disparate impact, except that if the complaining party can demonstrate to the court that the elements of a respondent’s decisionmaking process are not capable of separation for analysis, the decisionmaking process may be analyzed as one employment practice. Id. § 105(a) (codified at 42 U.S.C. § 2000e-2(k)(1)(B)(i)). The language in the Civil Rights Act of 1991 did not include the specific language regarding record keeping that was present in the unsuccessful Civil Rights Act of 1990, but the general language used in the Civil Rights Act of 1991 to establish an exception to the identification of particular employment practices was stated in terms broad enough to cover situations where an employer fails to keep records. 6. Wal-Mart: Sharp divisions again. The last significant United States Supreme Court case regarding disparate impact is Wal-Mart. In this case, the Supreme Court considered a nationwide class action brought by female employees on behalf of some 1.5 million current and former female employees of Wal-Mart stores. Wal-Mart, 564 U.S. at ___ , 131 S. Ct. at 2547 , 180 L. Ed. 2d at 385 . The employees claimed that local Wal-Mart managers exercised their discretion over pay and promotions disproportionately in favor of men, causing an unlawful disparate impact under the Federal Civil Rights Act. Id. The district court certified the class and the Ninth Circuit affirmed. Id. at ___, 131 S. Ct. at 2549 , 180 L. Ed. 2d at 388 . On appeal, a bare majority of the Supreme Court reversed. In an opinion written by Justice Scalia, the majority held that the class should not have been certified under the applicable federal rules. Id. at ___, 131 S. Ct. at 2556–57, 180 L. Ed. 2d at 395–96. The class certification question, however, was intertwined with the merits of the case. Id. at — 28 — ___, 131 S. Ct. at 2552 , 180 L. Ed. 2d at 391 . Justice Scalia stressed that allowing discretion by local managers is the opposite of a uniform pattern or practice that would provide commonality needed for a class action. Id. at ___, 131 S. Ct. at 2554 , 180 L. Ed. 2d at 392 . Justice Scalia noted that “[i]n a company of Wal-Mart’s size and geographic scope, it is quite unbelievable that all managers would exercise their discretion in a common way without some common direction.” Id. at ___, 131 S. Ct. at 2555 , 180 L. Ed. 2d at 393 . Justice Scalia rejected the “social framework analysis” as not offering a sufficient basis for finding commonality across the class. Id. at ___, 131 S. Ct. at 2554–55, 180 L. Ed. 2d at 393 . According to Justice Scalia, “Other than the bare existence of delegated discretion, respondents have identified no ‘specific employment practice’—much less one that ties all their 1.5 million claims together.” Id. at ___, 131 S. Ct. at 2555–56, 180 L. Ed. 2d at 394 . According to Justice Scalia, “Merely showing that Wal-Mart’s policy of discretion has produced an overall sex-based disparity does not suffice.” Id. at ___, 131 S. Ct. at 2556 , 180 L. Ed. 2d at 394 . Justice Ginsburg dissented in part and was joined by Justices Breyer, Sotomayor, and Kagan. Id. at ___, 131 S. Ct. at 2561 , 180 L. Ed. 2d at 400 (Ginsburg, J., concurring in part and dissenting in part). Justice Ginsburg adopted the framework embraced in the Civil Rights Act of 1991, but applied that framework in a fashion different than the majority. According to Justice Ginsburg, the district court had identified “systems for . . . promoting in-store employees that were sufficiently similar across regions and stores to conclude that the manner in which these systems affect the class raises issues that are common to all class members.” Id. at ___, 131 S. Ct. at 2563 , 180 L. Ed. 2d at 402 (internal quotation marks omitted). Justice Ginsburg wrote that “[t]he practice of — 29 — delegating to supervisors large discretion to make personnel decisions, uncontrolled by formal standards, has long been known to have the potential to produce disparate effects.” Id. at ___, 131 S. Ct. at 2564 , 180 L. Ed. 2d at 403 . Citing Watson and Wards Cove, Justice Ginsburg stressed that “[a]ware of ‘the problem of subconscious stereotypes and prejudices,’ we held that the ‘employer’s undisciplined system of subjective decisionmaking’ was an ‘employment practice’ that ‘may be analyzed under the disparate impact approach.’ ” Id. at ___, 131 S. Ct. at 2565 , 180 L. Ed. 2d at 404 (quoting Watson, 487 U.S. at 990–91, 108 S. Ct. at 2777 , 101 L. Ed. 2d at 842–43). Justice Ginsburg noted that the plaintiffs had offered statistical evidence that showed, after controlling for factors including “job performance, length of time with the company, and the store where an employee worked,” there was a sufficient statistical basis to give rise to an inference of discrimination. Id. at ___, 131 S. Ct. at 2564 , 180 L. Ed. 2d at 403 . D. Treatment of Disparate Impact Analysis Under the Iowa Civil Rights Act. We have had only a few occasions to consider cases under the Iowa Civil Rights Act when disparate impact claims were presented. See, e.g., Hy-Vee Food Stores, Inc. v. Iowa Civil Rights Comm’n, 453 N.W.2d 512 , 517–19 (1990); Wilson-Sinclair Co. v. Griggs, 211 N.W.2d 133 , 140–41 (Iowa 1973). In those cases, the parties did not argue that state law should be interpreted differently than federal law. Nonetheless, it is generally true that “Iowa courts have traditionally looked to federal law for guidance in interpreting” the Iowa Civil Rights Act. Pecenka v. Fareway Stores, Inc., 672 N.W.2d 800, 803 (Iowa 2003). We are, however, “not bound by federal law, despite consistent utilization of the federal analytical framework.” Id. (citing Hulme v. Barrett, 449 N.W.2d 629, 631 (Iowa 1989)). — 30 — We have not, however, explicitly adopted under state law either the teaching of Wards Cove or Wal-Mart. It is true that in Hy-Vee, we cited Wards Cove in describing the differences between discriminatory treatment and discriminatory impact cases. See Hy-Vee, 453 N.W.2d at 518–19. We did not, however, adopt the holdings in Wards Cove lock, stock, and barrel, and in no case has a party asked us to consider the merits of the minority opinion in Wards Cove or some other approach under the Iowa Civil Rights Act. In Hy-Vee, there was no question regarding the presence of a particular discriminatory practice, namely, the sexual segregation of stocker and checker positions. See id. at 520. Similarly, with respect to Wal-Mart, we have had no occasion to consider whether the majority or minority opinion in this 5–4 decision has the most persuasive power. We thus have a number of interpretive options under the Iowa Civil Rights Act. Do we follow the majority or the minority in Wards Cove or Wal-Mart? Or, do we follow a third path? While Congress passed the Civil Rights Act of 1991 in response to Wards Cove, no similar amendment has been made to the Iowa Civil Rights Act. The fact that Congress enacted a legislative change in response to a binding majority opinion of the United States Supreme Court does not have persuasive force in the interpretation of the Iowa Civil Rights Act. We have not adopted the principles of Wards Cove in the construction of the Iowa Civil Rights Act and are not bound to do so. Congressional reaction to a specific case decided by the United States Supreme Court does not shed light on the meaning of state law when there has been no comparable narrow state court precedent to stimulate a legislative override. E. Academic Literature on Disparate Impact in the Setting of Subjective Decision-Making. There is a body of literature grappling — 31 — with disparate impact theory in the context of subjective decisionmaking. In a ground breaking article in 1993, David Benjamin Oppenheimer suggested that negligence theory might be a basis for disparate impact theory. See David Benjamin Oppenheimer, Negligent Discrimination, 141 U. Pa. L. Rev. 899 , 899 (1993) (examining “psychological and sociological data on racism [to] demonstrate why discrimination is more closely analogous to negligent conduct than it is to intentional conduct”). This article has spawned significant offspring in the literature. See generally, e.g., Elizabeth Tippett, Robbing a Barren Vault: The Implications of Dukes v. Wal-Mart for Cases Challenging Subjective Employment Practices, 29 Hofstra Lab. & Emp. L.J. 433 (2012). There is reason to believe that at least some members of the United States Supreme Court might be interested in negligence theory in the context of subjective decision-making. At oral argument in the Wal-Mart matter, Justice Kennedy and Justice Roberts asked questions about whether the plaintiff was advancing a “notice theory,” namely, that an employer aware of the discriminatory impact of its subjective practices may be liable under the Federal Civil Rights Act. See Deborah M. Weiss, A Grudging Defense of Wal-Mart v. Dukes, 24 Yale J.L. & Feminism 119 , 123 (2012). The plaintiffs’ lawyer apparently walked away from the theory. See id. at 123, 167–68 (advocating a notice liability approach when an employer is aware of problems and does not fashion an appropriate remedy). IV. Discussion of Specific Employment Practice and Incapable of Separation for Analysis Issues. A. Positions of the Parties. — 32 — 1. The plaintiffs. The plaintiffs generally claim that the district court erred in finding that the plaintiffs failed to show that the State’s job selection process was not capable of separation for analysis. According to the plaintiffs, the primary issue on appeal is “whether Defendants’ common hiring and promotion system permitted [the Plaintiffs] to perform statistical analysis of [selection methods or protection practices] or elements of decision-making.” In support of their argument, the plaintiffs claim that the defendants failed to properly record the use or lack of use of any specific employment practices applied by any of the thirty-seven executive branch departments, thus making a statistical analysis of any separate element impossible. The plaintiffs challenge both the lack of aggregate data maintained by DAS and the underlying documentation in department hiring files. The plaintiffs point out that DAS is responsible for the oversight of merit and affirmative action in employment. Yet, according to the plaintiffs, the information maintained by DAS did not contain data sufficient to allow analysis of specific employment practices. The plaintiffs note that the district court correctly found that “DAS retains no data, computerized or otherwise, that allows one to see how a certain person was screened and/or scored as compared to another applicant by a department.” Thus, the information maintained by DAS was not capable of separation for analysis because you could not compare the treatment of one applicant to another in any objective way, a necessary foundation in aggregate statistical analysis. The plaintiffs then turn to the underlying, hard copy records maintained by the departments. The plaintiffs note that under applicable administrative regulations, agencies are required to — 33 — keep records as required by the director . . . [which] shall, at a minimum, include tracking of the composition of applicant groups, their movement through steps in the hiring processes, and the impact of personnel actions on various group members when records are not otherwise available in centralized information systems. Iowa Admin. Code r. 11—68.2(3). Yet, the plaintiffs assert that the record shows that the underlying department records are inadequate for three reasons. First, many of the underlying employment files simply have missing documents. Second, many of the underlying files reveal that the agencies did not create documents in the first place showing why employees were chosen or not chosen after having been screened by DAS and being minimally qualified for the position. Finally, the plaintiffs assert that the agencies did not apply common standards when scoring systems were used to evaluate applicants. In support of their claim that the departmental records were inadequate to allow separation for analysis by a specific employment practice, the plaintiffs cite two types of evidence in the record. First, the plaintiffs point to admissions in the record by state employees or agents. When consultant CPS conducted a review of state hiring practices for the State in 2007, it noted that files it reviewed “were not complete and did not indicate why some applicants were interviewed and others were not.” CPS declared that its studies confirmed that the selection of persons for interviews from DAS lists of minimally qualified applicants created “a rebuttable inference of adverse impact” but noted that inconsistencies in the State’s records within the same department prevented a more detailed analysis. When DAS attempted to do an employment audit pursuant to Executive Order No. 4, DAS officials indicated that they found more lack of documentation than CPS. See Exec. Order No. 4 — 34 — (2007), available at http://www.statelibraryofiowa.org/services/ collections/law-library/govexecorders/copy_of_execordculver. Further, while Executive Order No. 4, among other things, required agencies to assess the impact of screening methods on employee groups in the selection process, see id., a DAS official, when asked if DAS was incapable of complying with Executive Order No. 4, responded, “Right. We needed to do more.” Ultimately, DAS abandoned its effort to conduct an audit in compliance with Executive Order No. 4. Second, the plaintiffs analyzed the 667 hiring files produced by the State in discovery. The plaintiffs assert that an analysis of the files reveals that fifty percent did not include résumé review screening devices, over one in ten did not include interview questions, nearly one in five did not include interview notes, and over twenty-five percent did not include an interview scoring matrix. The plaintiffs recognize that the aggregate data can be divided into smaller parts. The plaintiffs concede that it is possible to sort the data by Equal Employment Opportunity (EEO) job category, by year, and by step in the hiring process. But the plaintiffs maintain that such division of the data into smaller units does not anchor the statistical analysis in specific employment practices, but simply lessens the size of the sample for statistical analysis. In other words, stacking documents by year does not help focus on an employment practice nor does stacking documents by EEO job category or step in the hiring process. The main effect of such slicing and dicing is to lessen the size of the sample, thereby reducing the power of aggregate statistical analysis without achieving any increase in focus. In short, the plaintiffs claim they did the best they could with the available data and that the aggregate analysis of disparate impact was — 35 — “as specific as the choices the employer permitted.” Based on the aggregate data, they point out that the racial disparity in the hiring of applicants deemed qualified for the job by DAS was statistically significant and that the likelihood of the result occurring in a raceneutral environment was as much as two billion to one, depending on the data set used. Further, the plaintiffs assert that the record showed that African Americans would have a forty percent better chance of being hired or promoted if they were white. The plaintiffs point to Port Authority Police Asian Jade Society v. Port Authority, in support of their argument. 681 F. Supp. 2d 456 (S.D.N.Y. 2010). In that case, the district court held that because an employer failed to maintain records, “the role of each step cannot be determined, [and] the steps cannot be examined separately to discover whether a particular step causes a disparate impact.” Id. at 464 . The plaintiffs further connect the disparate impact shown by their statistics with the lack of accountability in the State’s personnel system. They point out that their experts testified that accountability is an important aspect of integrated employment standards working to prevent biased or invalid decision-making. Finally, in addition to lack of record keeping, the plaintiffs note that the subjective manner in which the State makes its personnel decisions prevents separation for analysis of more specific personnel practices. See Watson, 487 U.S. at 989–90, 108 S. Ct. at 2786 , 101 L. Ed. 2d at 841–42. The plaintiffs cite Watson for the proposition that where an employment system combines objective and subjective features, it should be considered subjective in nature because of the ripple effect of subjective practices. See id. According to the plaintiffs, — 36 — under Watson, subjective features can be analyzed as one practice under disparate impact analysis. In support of its assertion that decision-making processes that combine objective and subjective decision-making should be considered as one employment practice under federal law, the plaintiffs cite Stender v. Lucky Stores, Inc., 803 F. Supp. 259 , 335–36 (N.D. Cal. 1992) and McClain v. Lufkin Industries, Inc., 187 F.R.D. 267, 275 (E.D. Tex. 1999). In Stender, the district court considered a class action brought by African American and female employees working in the approximately 150 to 185 retail stores within Lucky’s Northern California Food Division. 803 F. Supp. at 266, 267 . The Stender court declared that the plaintiff need not identify a particular employment practice “[w]here the system of promotion is pervaded by a lack of uniform criteria, criteria that are subjective as well as variable, discretionary placements and promotions, the failure to follow set procedures and the absence of written policies or justifications for promotional decisions.” Id. at 335 . In support of its opinion, the Stender court cited Allen v. Seidman, 881 F.2d 375 (7th Cir. 1989). Stender, 803 F. Supp. at 335 . In Seidman, the court considered a Title VII challenge brought by black bank examiners employed by the FDIC. 881 F.2d at 378 . The plaintiffs challenged a program evaluation test, which only thirty-nine percent of the African American candidates passed compared to eighty-four percent of the white candidates. Id. No regression analysis was performed. Id. at 380 . Judge Posner wrote that the statistics alone, without any further proof, established a prima facie case. Id. He noted that where “there has been a full trial, the issue of prima facie case drops out, and the question becomes whether the judge is persuaded that the test or other challenged — 37 — practice is discriminatory because it has a disparate impact unjustified by the defendant’s legitimate business needs.” Id. at 379 . In Lufkin, the district court considered the issue of class certification in a case where African Americans sued an employer on a disparate impact theory. Lufkin, 187 F.R.D. at 272 . Candidly characterizing the law as “complex and convoluted,” the district court canvassed the law on disparate impact claims. Id. at 271 , 272–75. With respect to the identification of employment practices, the district recognized that under the Civil Rights Act of 1991, a plaintiff was required to demonstrate that “ ‘each particular challenged employment practice causes a disparate impact’ ” except where an employer’s decision-making process is “ ‘not capable of separation for analysis.’ ” Id. at 272 (quoting 42 U.S.C. § 2000e-2(k)(1)(B)(i)). The Lufkin court noted that under Lufkin’s employment process, “[a] broad array of . . . employment practices rest on . . . subjective decision making” both in central administration and within each division of the company. Id. at 273. As in this case, applicants are channeled through a centralized human resources department where candidates who meet the objective minimum criteria for jobs are forwarded to management employees for approval unguided by any objective standards. See id. The district court further noted that “[t]he pervasive subjective decision-making process interacts with other facially neutral employment conditions to the disadvantage of African-Americans” through a ripple effect. Id. at 274. Concluding “Lufkin’s subjective employment practices [were] inextricably intertwined,” the district court held that elements of the respondent’s decision-making process were not capable of separation of analysis for purposes of class certification. Id. at 275. — 38 — 2. The State. The State begins its discussion by asserting that whether the State’s decision-making process is capable for separation for analysis is a question of fact. The State asserts that the plaintiffs’ claim that the decision-making process was not capable of separation fails because the plaintiffs never attempted to make such an analysis and because they received voluminous amounts of data and hiring files. The State claims that the evidence demonstrates that the plaintiffs never tried to identify and analyze any particular employment practice or decision-making process. For example, the State suggests that the plaintiffs did not attempt to analyze hard documents in the departments because they were not in a convenient digital format. In any event, the State maintains that it provided substantial information in digital form in the BrassRing system, and the Human Resource Information System (HRIS) data system, which were maintained by DAS. In short, the State claims that the plaintiffs were not forced to engage in system-wide analysis because the decision-making process was not capable of separation for analysis, but instead simply chose to engage in a systemwide challenge. The State emphasized that the plaintiffs had the affirmative burden of showing that the process was not capable of separation for analysis. The State emphasizes the factual nature of the inquiry. The State further claims that at trial the plaintiffs did not offer testimony that the decision-making process was not capable of separation for analysis. The State asserts that the plaintiffs never tried to analyze separate practices or processes, but simply preferred to proceed on a system-wide basis. The State highlights the staggering amount of information that was presented to the plaintiffs in the BrassRing files and in the HRIS data — 39 — system. The State notes that Killingsworth engaged in analysis of the data by EEO category, by year, and by step in the hiring process. B. Analysis of Specific Employment Practice and Incapable of Separation for Analysis Under Title VII, as Amended by the Civil Rights Act of 1991. Under the Civil Rights Act of 1991, a plaintiff in a disparate impact case must identify a “particular employment practice” being challenged or, in the alternative, demonstrate why an employer’s decision-making process is “not capable of separation for analysis.” 42 U.S.C. § 2000e-2(k)(1)(B)(i). The district court held that the plaintiffs failed to show that the State’s hiring practice was not capable of separation for analysis. On appeal, the plaintiffs do not claim they identified a particular employment practice. Instead, the plaintiffs contend that the record demonstrates that because of the poor record keeping of the State, and because of the use of subjective criteria in the various departments, it was impossible to engage in a more focused analysis of the hiring practices of the State beyond what it presented in the case. Although the plaintiffs’ evidence shows notice to the State of potential disparate impact arising out of its employment processes, the plaintiffs, like the plaintiffs in Wal-Mart, did not pursue a notice/negligence-type theory. A few preliminary matters should be discussed. First, the mere fact that the class involves a number of departments and different positions over a period of years by one employer does not necessarily demonstrate that the State’s decision-making process is capable of separation for analysis. To the extent relevant, the class in this case is much more compact than in Wal-Mart, where 1.5 million employees were located in 3400 stores in all fifty states. See id. at ___, 131 S. Ct. at 2557 , 180 L. Ed. 2d at 395 . Many lower federal courts have — 40 — distinguished Wal-Mart on the ground that the classes being challenged were more compact. 7 Nor does the fact that the State flooded the plaintiffs with computerized data and documents decide the case. There is no question that the State databases provided to the plaintiffs contain thousands and thousands of bits of data. Conclusory statements by witnesses and lawyers regarding the nature of the information presented yield little value. And, the mere fact that a trial lasted seventeen days does not mean there must be substantial evidence supporting key findings of fact. We must put the conclusory rhetoric aside and consider, first, what does it mean for an employer’s decision-making to be incapable of separation for analysis? Then, once we understand the meaning of the statutory phrase, we must examine the record to determine if the plaintiffs have met their burden in this case. On the issue of what is meant by a decision-making process that is incapable of separation for analysis, the parties provide us with little guidance. No one disputes that the plaintiffs bear the burden of proof on the issue. But what does it mean to be “incapable” of “separation” for “analysis”? We begin with a review of the three key statutory words: incapable, separation, and analysis. None are statutorily defined. The word incapable generally refers to something that cannot be done. See 7See,e.g., Meyer v. U.S. Tennis Ass’n, 297 F.R.D. 75, 84 (S.D.N.Y. 2013) (distinguishing nationwide class in Wal-Mart from narrow class of hundreds of U.S. Open umpires); McReynolds v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 672 F.3d 482, 488 (7th Cir. 2012) (noting claim was limited to about 700 brokers, a far cry from WalMart’s class of 1.5 million); Ellis v. Costco Wholesale Corp., 285 F.R.D. 492, 509 (N.D. Cal. 2012) (noting that the proposed class was 700, which was much smaller than the 1.5 million employees who sought to be certified in Wal-Mart). — 41 — Merriam-Webster’s Collegiate Dictionary 585 (10th ed. 2002). Separation has several different but related meanings, including “a point, line, or means of division,” or “an intervening space.” Id. at 1064 (“separate” used as a verb means to set or keep apart, to remove from a mixture or to isolate). In the context of disparate impact, we believe the term analysis must mean statistical analysis. While an understanding of these three statutory terms is helpful, we still need to probe the statutory context. What kind of separation is sufficient? Separated or isolated from what? In context, it seems clear that what must be separated out for analysis from the employers decision-making process is particular employment practices, as the separation of particular employment practices is what the statute ordinarily requires. The plaintiffs must show they cannot spin out separate employment practices from the larger whole that are capable of statistical analysis. Given these statutory terms and their common sense definitions, it seems that a decision-making process may be incapable of separation for analysis under at least three circumstances. First, the substantive features of the decision-making process itself may be such that the decision-making process is incapable of separation for analysis into specific employment practices. That is the teaching of Stender. See 803 F. Supp. at 335 (finding employee’s “subjective and ambiguous decisionmaking processes” incapable of separation for analysis). For instance, a wholly subjective process, even if decentralized, would be incapable of separation because of a lack of objective criteria. See Ronald D. Rotunda, The Civil Rights Act of 1991: A Brief Introductory Analysis of the Congressional Response to Judicial Interpretation, 68 Notre Dame L. Rev. — 42 — 923, 934 (1993) (noting that “hiring processes are often complicated, with ill-defined or ill-followed guidelines”). Second, even well-defined employment practices may be so intertwined as not capable of meaningful analysis separately. The classic example is Dothard v. Rawlinson, 433 U.S. 321 , 97 S. Ct. 2720 , 53 L. Ed. 2d 786 (1977). In Dothard, height and weight requirements for correctional counselor positions in the Alabama state penitentiary system, if considered separately, had relatively mild adverse impacts on women, but when considered in combination, the adverse impact significantly increased. 433 U.S. at 329–30, 97 S. Ct. at 2727, 53 L. Ed. 2d at 797 . If the various employment practices cannot be isolated and considered independently, they are not capable of separate analysis. Third, the failure of the employer to keep adequate records can make an employment decision incapable of separation for analysis. See Port Auth. Police Asian Jade Soc., 681 F. Supp. 2d at 464 . It is true that the Civil Rights Act of 1991 did not contain the more specific language of the proposed Civil Rights Act of 1990 (which, as noted above, provided that a lack of records could excuse the particularity requirement), but the adoption of the more general language of the Civil Rights Act of 1991 is certainly broad enough to encompass such an approach and plainly does not preclude it. If, for example, the various departments of an employer do not maintain records of interview criteria, including the manner in which the interview is scored, and the scores awarded by the employer based on the interview, it is difficult to see how a plaintiff could engage in separate analysis of disparate impact. A contrary result would be an incentive for employers to refuse to keep adequate records of their employment processes. Thus, an employer who declines to keep employment records from which particular employment practices are — 43 — capable of separation for analysis may face a lawsuit based on systemas-a-whole-disparate impact. Although the point is not always clear, the plaintiffs on appeal in this case do not make an argument based solely upon one of the above scenarios. Instead, the plaintiffs offered a hybrid argument, based upon a combination of the above factors. The plaintiffs argue that a combination of (1) ill-defined subjective practices, (2) intertwined elements of decision-making, and (3) lack of adequate record keeping by the employer prevented them from identifying specific employment practices for purposes of disparate impact analysis. Based on the above discussion, some of the analysis of the district court on the separation issue appears off the mark. For example, the district court declared that the analysis of employment practice or process is focused on the “job specific” level. But this is not necessarily true. A plaintiff is not required to focus on a job specific level if it can be shown that any potential job specific employment practices are not capable of separation for analysis. Indeed, such a contention is antithetical to system-as-a-whole attacks that are permitted under the statutory exception in 42 U.S.C. § 2000e-2(k)(1)(B)(i). Thus, the mere fact that the data could be broken down by EEO-4 categories, for instance, does not mean the plaintiff cannot proceed on a system-as-awhole theory. Similarly, there is language in the district court ruling suggesting that the existence of separate departments within an employer prevented the plaintiffs from proceeding on a decision-making-as-a-whole theory. The mere fact that there are departments within an employer, however, does not in itself mean that a plaintiff cannot show that the decisionmaking process is not capable of separation for analysis. If the plaintiff — 44 — can demonstrate, for instance, that the departments do not operate under separate and identifiable employment practices, or that the records are so deficient the alleged department practices cannot be separately analyzed, then the plaintiff may be able to proceed on a decision-making process as a whole theory. In short, the fact that the plaintiffs were provided with lots of data that can mechanically be sliced and diced in numerous ways proves nothing; massive data can always be divided into countless different piles. But the key question is not whether the massive data can be divided up into piles, the question is whether the plaintiffs demonstrated any resulting piles that might be formed do not reveal particular employment practices that are capable of separation for statistical analysis. Consider the following hypothetical. Suppose a class of African Americans challenged a state merit system of thirty-seven departments which hired thousands of persons over a ten-year period over many EEO categories of jobs. Suppose further that the State admitted that the hiring decisions were made at the unfettered discretion of individual managers in each department. Assuming no other facts, the plaintiffs would be entitled to bring their class action under Title VII because there would be no identifiable particular employment practices that were capable of separate (statistical) analysis. Take the above hypothetical and add the fact that the State provided the plaintiff with a number of databases with hundreds of thousands of data points. Suppose further that these documents included numerous job résumés, many miscellaneous descriptions of the job positions, dates the interviews were conducted, and the names of — 45 — managers who made the decisions. Would this barrage of data mean the plaintiff was barred from proceeding on a system-as-a-whole basis? The answer might be no. Even if the data contains thousands or even millions of bits of information, the plaintiff may show the data does not provide a basis for a plaintiff to identify particular employment practices in an employer’s decision-making process that are subject to separate (statistical) analysis. In short, the amount of information produced is irrelevant. It is the quality of the information that is key. Let us now change the hypothetical somewhat. Assume further that the data dump in fact contained more substantive information on the decision-making process. Suppose the data was a collection of imperfect individual employment files, many of which contained specific job related criteria, or matrices for scoring interviewees and the scoring resulting from such interviews, and other significant material on the hiring process. Now, under this modified hypothetical, could the plaintiff identify specific employment practices that could be subject to separate (statistical) analysis? Nonetheless, the above discussion does not mean the plaintiffs must prevail. Under Wards Cove and Wal-Mart, the plaintiffs still must show there were not specific employment practices within the universe of the state merit employment system with sufficient aggregate numbers that they could be separated out for meaningful statistical analysis. See Wal-Mart, 564 U.S. at ___ , 131 S. Ct. at 2555–56, 180 L. Ed. 2d at 394 ; Wards Cove, 490 U.S. at 656 , 109 S. Ct. at 2124 , 104 L. Ed. 2d at 751 . In order to prevail, the plaintiffs would have to show that the vast universe of job selection could not be divided into smaller, better defined subsets of specific employment practices with sufficient decision points to be capable of statistical analysis. — 46 — The district court seems to have found that the plaintiffs may have inadvertently done just that. The plaintiffs’ expert Mark Killingsworth testified he could statistically analyze the disparate impact at what the parties called step two of the analysis, namely, at the stage where DAS cleared minimally qualified applicants and passed them onto the individual departments for selection for an interview. But, we doubt step two is an employment practice “capable of separation for analysis.” All employment processes have chronological or procedural steps, but these are not the equivalent of an employment practice with sufficient definition that is subject to statistical analysis. Further, Killingsworth testified step two could not really be separated from step three, the final decision-making step in the process. The district court did not make a specific finding regarding whether step two could be separated or isolated from step three. Nonetheless, as the district court pointed out, the plaintiffs have a further problem. While the parties utilized central databases maintained by DAS for what the district court accurately called “slicing and dicing” of the statistical data, the databases may have been inadequate to engage in analysis of specific employment policies for disparate impact by department or other nonsystem-wide approaches. Even so, the departments themselves maintained hard file copies of employment records that may have included more information than was available on the database. The question arises whether the plaintiffs adequately proved this information could not have been utilized to separate out employment practices by the various departments or agencies that would be capable of separation for analysis. While the plaintiffs claim the hiring data in the hard files was incomplete, the question remains — 47 — whether there was sufficient information in the hiring files to construct a meaningful database to analyze specific employment practices. The district court made findings related to the underlying documents. It declared that “the hiring files themselves permit a focused view of the different screening devices and practices in the referral, interview, or hiring of applicants for any given job between the departments.” Further, the district court stated that “one can focus on any number of discrete employment decisions made as individual, separate, discrete employment practices” and provided examples, including a “second résumé screen” and a “spelling and grammar screen.” Read in a fashion to support the district court’s verdict, these findings demonstrate the district court found that employment practices could be extracted from the underlying documentary files and statistically analyzed in a meaningful way. There is, however, no requirement the defendants prove that employment practices are capable of separation for meaningful statistical analysis. The precise legal issue is whether the plaintiffs met their burden in showing that the particular employment practices could not be separated and analyzed from the documentary files maintained by the State. See 42 U.S.C. § 2000e2(k)(l)(B)(i). In short, at least under the theory of the case as litigated by the parties, the plaintiffs have the burden of proving the negative. We conclude the district court correctly resolved the issue adversely to the plaintiffs and that such a finding is supported by substantial evidence in the record. Killingsworth did not review the underlying documentary files and offered no testimony indicating specific employment practices could not be extracted from the underlying files for statistical analysis notwithstanding the flaws in some of the files. The — 48 — State’s expert, Miller, suggested the underlying documents were capable of separation for analysis. Dr. Greenwald characterized the hiring files as “a gold mine that hasn’t been analyzed.” While it is true the underlying files were often incomplete and flawed, that does not necessarily mean employment practices could not be identified and statistically analyzed in a meaningful way. The bottom line, on the record before us, is that while the plaintiffs demonstrated the recordkeeping was sometimes incomplete, the district court on the record before it could conclude that the plaintiffs failed to show the negative, namely, that employment practices could not be extracted from the underlying documents and analyzed in a statistically significant manner. On this issue, the district court got it right. As a result, under applicable federal law, the State was entitled to summary judgment on the record developed in the district court on the plaintiffs’ claim under Title VII of the Civil Rights Act of 1964. V. Discussion of Specific Employment Practice and Incapable of Separation Analysis Under the Iowa Civil Rights Act. We now turn to the question of whether the defendant was entitled to summary judgment under the Iowa Civil Rights Act. See Iowa Code § 216.6 . Although it is often said that state civil rights acts were patterned after the Federal Civil Rights Act, in fact more than twenty state civil rights acts predated the Federal Act. See Arthur E. Bonfield, State Civil Rights Statutes: Some Proposals, 49 Iowa L. Rev. 1067 , 1107 & n.140 (1964) (listing states). In an important article advocating the passage of the Iowa Civil Rights Act, Bonfield relied extensively on state models in proposing legislative action in Iowa. See id. at 1082 (discussing states’ antidiscrimination laws). Thus, though the Iowa Civil Rights Act was enacted in the year following the enactment of the Federal — 49 — Civil Rights Act, the Iowa Civil Rights Act draws on substantial state as well as federal legislative precedent. See id. at 1095–1123 (reviewing states’ antidiscrimination laws and proposing statutes for Iowa). The substantive provisions of the Iowa Civil Rights Act and Title VII of the Civil Rights Act of 1964 are often similar though not identical. With respect to discrimination in employment, the Iowa Civil Rights Act provides that “[i]t shall be an unfair or discriminatory practice for any . . . [p]erson to refuse to hire” or “otherwise discriminate in employment against any applicant for employment or any employee because of . . . race.” Iowa Code § 216.6 (1)(a). The parallel provision under the Federal Civil Rights Act provides that “[i]t shall be an unlawful employment practice for an employer . . . to fail or refuse to hire . . . or otherwise discriminate against any individual because of . . . race.” 42 U.S.C. § 2000e-2(a)(1). There are, however, differences between the state and federal acts. For instance, the Iowa legislature has declared that the Iowa Civil Rights Act “shall be construed broadly to effectuate its purposes.” Iowa Code § 216.18 (1). There is no similar language in the Federal Civil Rights Act and, indeed, the case can be made that the recent cases of the United States Supreme Court, particularly Wards Cove and Wal-Mart, tend to construe the federal counterpart narrowly. Other state courts have interpreted similar legislative directions to mean that the remedies afforded by the state civil rights statues require the “widest constitutional application.” Fair Emp’t Practices Comm’n v. Rush-Presbyterian-St. Luke’s Med. Ctr., 354 N.E.2d 596, 600 (Ill. App. Ct. 1976); see also Frieler v. Carlson Mktg. Grp., Inc., 751 N.W.2d 558 , 571–73 (Minn. 2008) (reviewing various courts’ interpretations of the term “supervisor” and concluding for purposes of sexual harassment claim under state law, the court — 50 — would adopt a broader view because state law required “liberal construction of its terms”); Genaro v. Cent. Transp., Inc., 703 N.E.2d 782, 785 (Ohio 1999) (citing language that the state chapter “shall be construed liberally for the accomplishment of its purposes” in departing from federal precedent (internal quotation marks omitted)). An Iowa court faced with competing legal interpretations of the Iowa Civil Rights Act must keep in mind the legislative direction of broadly interpreting the Act when choosing among plausible legal alternatives. Any state court decision that adopts a narrow construction of Title VII by the United States Supreme Court without confronting the requirement in Iowa law that the Iowa Civil Rights Act be interpreted broadly misses an essential difference in state and federal civil rights laws. Even where language in a state civil rights statute is parallel to the Federal Civil Rights Act, a state court is under no obligation to follow federal precedent. As noted by the Vermont Supreme Court, federal civil rights decisions may be persuasive, but they are not the only sources of persuasive authority on the interpretation of state civil rights statutes. Lavalley v. E.B. & A.C. Whiting Co., 692 A.2d 367, 369 (Vt. 1997). Federal court decisions under the Federal Civil Rights Act are not binding on state courts, which are free to consider other persuasive authority and come to independent conclusions. See, e.g., Brown v. F.L. Roberts & Co., 896 N.E.2d 1279, 1285 (Mass. 2008) (noting the court “frequently” does not follow the reasoning of federal precedent in interpreting the state civil rights statute (internal quotation marks omitted)); Elezovic v. Ford Motor Co., 697 N.W.2d 851, 859 (Mich. 2005) (finding supervisor liable for sexual harassment under Michigan civil rights statute, noting that “we are not compelled to follow . . . federal interpretations” (internal quotation marks omitted)); Frieler, 751 N.W.2d — 51 — at 571–73 (considering varying interpretations of the term supervisor for claims of sexual harassment); Grimwood v. Univ. of Puget Sound, Inc., 753 P.2d 517, 520 (Wash. 1988) (noting that “[w]hile these federal cases are a source of guidance, we bear in mind that they are not binding and that we are free to adopt those theories and rationale which best further the purposes and mandates of our state statute”); Goodyear Tire & Rubber Co. v. Dep’t of Indus., 273 N.W.2d 786, 791 (Wis. Ct. App. 1978) (noting Wisconsin courts “must construe Wisconsin statutes as it believed the Wisconsin legislature intended, regardless of how Congress may have intended comparable statutes”); cf. State v. Baldon, 829 N.W.2d 785 , 811–16 (Iowa 2013) (Appel, J., concurring specially) (noting, inter alia, that state constitutional provisions need not be interpreted uniformly with federal caselaw under parallel federal constitutional provisions). See generally Alex B. Long, “If the Train Should Jump the Track . . .”: Divergent Interpretations of State and Federal Employment Discrimination Statutes, 40 Ga. L. Rev. 469 , 482–83 (2006) (finding parallel between independent state constitutional interpretation and independent state court interpretation of state employment discrimination statutes). Recognition of the independent character of state civil rights statutes is particularly important when Congress passes legislation designed to overcome decisions of the United States Supreme Court narrowly interpreting civil rights statutes. For instance, when the United States Supreme Court held in General Electric Co. v. Gilbert that discrimination based on pregnancy was not sex discrimination, Congress overrode the decision. 429 U.S. 125 , 138–39, 97 S. Ct. 401 , 409–10, 50 L. Ed. 2d 343, 356 (1976), superseded by statute, Pregnancy Discrimination Act of 1978, Pub. L. No. 95-555, 92 Stat. 2076, as — 52 — recognized in Shaw v. Delta Air Lines, Inc., 463 U.S. 85 , 103 S. Ct. 2890 , 77 L. Ed. 2d 490 (1983). After the United States Supreme Court decided Wards Cove, Congress enacted legislation in response to the decision. See 42 U.S.C. § 2000e-2(k). Congress recently overrode the restrictive United States Supreme Court cases of Sutton v. United Air Lines, Inc., 527 U.S. 471 , 119 S. Ct. 2139 , 144 L. Ed. 2d 450 (1999), and Toyota Motor Manufacturing, Kentucky, Inc. v. Williams, 534 U.S. 184 , 122 S. Ct. 681 , 151 L. Ed. 2d 615 (2002), through the ADA Amendments Act of 2008, Pub. L. No. 110-325, § 2 (b)(2)–(5), 122 Stat. 3553 . Similarly, Congress acted in response to Ledbetter v. Goodyear Tire & Rubber Co. by enacting curative legislation. 550 U.S. 618 , 127 S. Ct. 2162 , 167 L. Ed. 2d 982 (2007), superseded by statute, Lilly Ledbetter Fair Pay Act of 2009, Pub. L. No. 111-2, 123 Stat. 5. The frequent narrow construction of civil rights laws by the United States Supreme Court, followed by congressional intervention, has been cited as a ground for decreased deference to United States Supreme Court decisions by state courts. See Sperino, Revitalizing, 20 Geo. Mason L. Rev. at 564–68 (“To the extent that the development of the federal [employment discrimination] frameworks depends on references to statutory languages and its historical development over time, reading the state statutes in accordance with these federal frameworks is highly suspect.”); Sandra F. Sperino, Diminishing Deference: Learning Lessons from Recent Congressional Rejection of the Supreme Court’s Interpretation of Discrimination Statutes, 33 Rutgers L. Rec. 40 , 42–43 (2009) (arguing that “repeated Congressional rejection of [the Supreme Court’s narrow interpretations of civil rights statutes] suggest[s] that state regimes should not be so beholden to what may likely be faulty interpretation on the part of the Supreme Court”). — 53 — The failure of the Iowa legislature to enact similar curative legislation, however, is of no particular moment when there has been no similar narrow judicial construction of the Iowa Civil Rights Act by this court. Federal cases are not binding on questions of state law and thus there is no need to override them through state legislative action. As noted by the Vermont Supreme Court, a state legislature is not required to “react to every federal decision interpreting Title VII or risk that its inaction will be interpreted as an endorsement of the federal decision.” Lavalley, 692 A.2d at 370 . The above principles are consistent with our caselaw. For example, in Hubbard v. State, we noted that [a]lthough decisions and interpretations of federal courts may be illustrative and instructive to state courts in construing statutes patterned after those enacted by Congress and entitled to great weight in determining construction to be given the same phrase in subsequently enacted state statutes, they are neither conclusive nor compulsory, especially when it appears earlier statutes substantially similar have also been enacted in other states. 163 N.W.2d 904, 909 (Iowa 1969) (emphasis added). Here, the Iowa Civil Rights Act was enacted only one year after the Federal Act and thus there was little preexisting caselaw that would be entitled to “great weight” under Hubbard. See id. Further, at the time the Iowa Civil Rights Act was passed, there were also state statutory counterparts, which may provide helpful precedents. See id. (“Where the language has been borrowed from the statutes of a sister state we would go for light to the construing decisions . . . of that state.”). Consistent with Hubbard, we look to federal caselaw, and the caselaw of other states under their state civil rights statutes, for persuasive guidance. For example, in Pecenka, we noted that we may look to federal interpretations in construing the Iowa Civil Rights Act but — 54 — are not bound by them. 672 N.W.2d at 803 . And, in holding that a supervisor may be personally liable for harassment under section 216.6(1) of the Iowa Civil Rights Act, we favorably cited a New York case construing state law. See Vivian, 601 N.W.2d at 877–78. The bottom line is that the Iowa Civil Rights Act is a source of law independent of the Federal Civil Rights Act. In construing the Act, we may look to federal and state court precedent, none of which are binding, but which may persuade us in the interpretation of the Iowa statute. In making choices under the Iowa Civil Rights Act, we must be mindful of the legislative direction that the Act be broadly interpreted to effectuate its purposes. See Iowa Code § 216.18 (1). The plaintiffs in their brief, however, do not explicitly invite us to interpret the Iowa Civil Rights Act in a fashion different than Title VII of the Federal Civil Rights Act. The plaintiffs declare that “generally speaking,” the same burden-shifting approach is applied under the Iowa Civil Rights Act as is applied under Title VII of the Federal Civil Rights Act. But the plaintiffs go even further. They seem to take the view that the criteria established in the Civil Rights Act of 1991 also apply under the Iowa Civil Rights Act even though there was no comparable statutory amendment. Plaintiffs simply state that under “the law,” a plaintiff must identify a specific employment practice or show that the decision-making process is not capable of separate analysis. Thus, the plaintiffs do not appear to make the substantive argument that Iowa law should embark on a different path than reflected in Wards Cove and the subsequent amendments to Title VII adopted by Congress or from Wal-Mart. We thus must confront a question of preservation. A narrow private law approach would suggest that we narrowly decide only the questions advanced by the parties. See Melvin Aron Eisenberg, — 55 — Participation, Responsiveness, and the Consultative Process: An Essay for Lon Fuller, 92 Harv. L. Rev. 410 , 413 (1978) (advocating strong responsiveness to the parties arguments “insofar as the parties contemplate that the court will settle their dispute on the basis of the issues as the parties see them”). On the other hand, in dealing with public law questions, the court has a responsibility for the development of law generally and cannot allow the advocacy of private parties to dictate legal development. See generally Abram Chayes, The Role of the Judge in Public Law Litigation, 89 Harv. L. Rev. 1281 (1976). The preservation question was explored at some depth in the context of common law development in Feld v. Borkowski, 790 N.W.2d 72 , 82–86 (Iowa 2010) (Appel, J., concurring in part and dissenting in part). In the constitutional context, we have stated when a party raises both federal and state constitutional claims, but does not establish a different substantive standard between the state and federal constitutions, we assume the federal standard applies, but reserve the right to apply that standard in a fashion different from federal courts. See State v. Edouard, 854 N.W.2d 421 , 452–54 (Iowa 2014) (Appel, J., concurring specially). Using this approach, we have reserved for another day some very important constitutional issues under the Iowa Constitution that, instead of being decided earlier, remain very much alive today. See, e.g., King v. State, 818 N.W.2d 1 , 47 n.52 (Iowa 2012) (Appel, J., dissenting) (reserving question of whether article IX, division 1, section 12 of the Iowa Constitution provides enforceable rights to a public education); State v. Lowe, 812 N.W.2d 554 , 593 n.23 (Iowa 2012) (Appel, J., concurring in part and dissenting in part) (reserving the question of whether Iowa should abandon the multifactor Schneckloth v. Bustamonte test in the search and seizure context in favor of a requiring — 56 — knowing and voluntary consent); State v. Iowa Dist. Ct., 801 N.W.2d 513 , 518 n.2 (Iowa 2011) (reserving question of whether participation in sex offender treatment program requiring offender to admit past crimes violated due process under the state constitution); State v. Effler, 769 N.W.2d 880, 890 , 895–97 (Iowa 2009) (Appel, J., concurring specially) (reserving the important question of whether we should reject the majority view expressed in Davis v. United States, 512 U.S. 452 , 114 S. Ct. 2350 , 129 L. Ed. 2d 362 (1994), in favor of the dissenting position). We could adopt a similar approach on an issue of statutory construction of a parallel statute. In this case, the plaintiffs structured the litigation and advanced arguments solely based upon federal law standards. Had the plaintiffs advanced an argument under state law departing from the federal precedent, for example, that a particular employment practice is not a requirement under the Iowa Civil Rights Act 8—a different factual record might have been developed at trial. Specifically, the State did not attempt to present a defense based upon business necessity, and the State’s response to the plaintiffs’ damage claim was quite limited. If, for example, the State knew the plaintiff was relying upon a different theory of law, it might have affected the factual development at trial. Under these circumstances, we decline to address arguments that were not advanced by the plaintiffs at trial. 8“Also troubling is the Court’s apparent redefinition of the employees’ burden of proof in a disparate-impact case.” Wards Cove, 490 U.S. at 672 , 109 S. Ct. at 2133 , 104 L. Ed. 2d at 761 (Stevens, J., dissenting). “No prima facie case will be made, it declares, unless the employees isolat[e] and identif[y] the specific employment practices that are allegedly responsible for any observed statistical disparities.” Id. (internal quotation marks omitted). “This additional proof requirement is unwarranted.” Id. — 57 — Nonetheless, even when the parties have not argued for different substantive principles, we still may apply the principles advocated by the parties in a fashion different than the federal courts. See Edouard, 854 N.W.2d at 452–54 (Appel, J., concurring specially). Yet, we find no basis to do so in this case. Under the framework presented to the district court, which for the purposes of this case we adopt, the plaintiffs failed to show that the underlying documents did not provide sufficient information to allow employment practices to be separated for meaningful statistical analysis. As a result, given the posture of this case, we affirm the decision of the district court under the Iowa Civil Rights Act. VI. Conclusion. For the above reasons, the district court judgment is affirmed. AFFIRMED. All justices concur except Waterman, Mansfield, and Zager, JJ., who concur specially. — 58 — #12–0913, Pippen v. State WATERMAN, Justice (concurring specially). I respectfully concur in the result only. I am unable to join the majority opinion’s affirmance of the district court’s thorough, wellreasoned decision on the merits in favor of defendants. See Pippen v. State, No. 05771 LACL107038, 2012 WL 1388902 (Iowa Dist. Ct. April 17, 2012), available at http://www.iowaappeals.com/wpcontent/uploads/2012/05/Pippen-Ruling.pdf. The majority’s discussion of the law includes dicta unnecessary to its holding. For example, the majority gratuitously undermines our court’s long-standing practice of relying on federal decisions to interpret equivalent provisions of the Iowa Civil Rights Act. Yet, the majority’s analysis of the dispositive issue is cursory. Given the importance of this case, I offer a more complete analysis of the grounds to affirm the judgment and respond to the dicta to help guide cases to follow. Plaintiffs are a certified class of over 5000 members defined as “[a]ll African American applicants and employees who sought appointment to or held a merit-system position with an Executive Branch agency (not including Board of Regents) at any point from July 1, 2003, through April 17, 2012.” They brought disparate impact racial discrimination claims under Title VII of the Civil Rights Act of 1964 and the Iowa Civil Rights Act of 1965 (ICRA). These civil rights laws were enacted to eliminate discriminatory practices, provide remedies for discrimination, and allow equal opportunities in employment. Defendants are the State of Iowa and its thirty-seven executive branch departments. There are over 700 types of employment positions within the executive branch. During the relevant period, nearly 500,000 applications were submitted by 100,000 applicants for 20,000 openings. — 59 — Approximately 2000 supervisors within the executive branch have authority in the hiring process. Plaintiffs sought to prove at trial that the State’s overall hiring system during the relevant time period discriminated on the basis of race. Disparate impact law generally requires plaintiffs to prove a particular employment practice caused a disparate impact. The majority fails to discuss the purpose of this proof requirement, which is to enable the district court to fashion a remedy correcting the discriminatory practice without affecting other practices that are not discriminatory. Congress enacted a narrow exception to this general requirement. Specifically, plaintiffs do not have to prove a particular employment practice had a discriminatory impact if they prove the State’s hiring process was incapable of “separation for analysis.” See 42 U.S.C. § 2000e-2(k)(1)(B)(i) (2006). In that event, the law relieves them of the burden of attributing the discrimination to a particular employment practice. Here, the plaintiffs sought to proceed under that exception. Following a month-long bench trial, the district court found the State’s employment decision-making process was capable of separation for analysis. For the reasons explained below, I conclude the “capable of separation” finding is supported by substantial evidence and is dispositive. I therefore agree the district court judgment must be affirmed. I. Additional Background Facts and Proceedings. As required under our standard of review in our appellate role, we review the trial evidence in the light most favorable to the judgment. Falczynski v. Amoco Oil Co., 533 N.W.2d 226, 230 (Iowa 1995). I will elaborate on the expert testimony discussed by the majority opinion. — 60 — As the majority notes, plaintiffs presented testimony from three expert witnesses. Anthony Greenwald and Cheryl Kaiser testified regarding the social science of implicit-bias stereotyping and how implicit biases affect decision-making. Kaiser explained the difference between explicit and implicit bias: explicit bias is “conscious, deliberate, controlled animosity,” whereas implicit bias is due to unconscious negative associations—stereotypes—that people have developed over time. 9 Implicit-bias theory helps explain how statistical disparities can result without intentional discrimination: individuals act on implicit biases without recognizing they are doing so. Greenwald testified unconscious bias leads to discrimination particularly in subjective decision-making and that most test groups demonstrate a seventy percent unconscious preference for whites over blacks. 10 And, as is customary in a disparate impact case, the plaintiffs presented testimony 9Research into the process of socialization and development of social norms [has] led to an understanding that the development of stereotypes—and consequent biases and prejudices—is not a function of an aberrational mind, but instead an outcome of normal cognitive processes associated with simplifying and storing information of overwhelming quantity and complexity that people encounter daily. Melissa Hart, Subjective Decisionmaking and Unconscious Discrimination, 56 Ala. L. Rev. 741 , 746 (2005) (internal quotation marks omitted). Implicit-bias research and its application to legal theories has been thoroughly reviewed in legal scholarship. See Jerry Kang, Trojan Horses of Race, 118 Harv. L. Rev. 1489 , 1515 & n.122 (2005) (providing summary of employment-related implicit-bias studies, including experiment where résumés with “white names” received fifty percent more callbacks than résumés with “black names”); see also Judge Mark W. Bennett, Unraveling the Gordian Knot of Implicit Bias in Jury Selection: The Problems of Judge-Dominated Voir Dire, the Failed Promise of Batson, and Proposed Solutions, 4 Harv. L. & Pol’y Rev. 149, 151–158 (2010) (discussing implicit-bias research as it relates to jury selection); L. Song Richardson, Arrest Efficiency and the Fourth Amendment, 95 Minn. L. Rev. 2035 (2011) (discussing implications of implicit bias for police–citizen interactions and Fourth Amendment jurisprudence, and proposing “debiasing strategies” for police departments). 10Notably,Greenwald is an inventor of the Implicit Association Test, a widely used method of measuring implicit bias. See Project Implicit, About Us, https://www.projectimplicit.net/about.html (last visited July 10, 2014). — 61 — from a statistical expert, labor economist Mark Killingsworth.11 Killingsworth chose to limit his analysis to applications referred by DAS to the departments (thus effectively excluding applicants who did not satisfy minimum qualifications from his review). 12 Killingsworth’s system-wide review for the years 2004 through 2008 showed that African-American applicants were statistically less likely than white applicants to be interviewed and hired in every year. For example, in 2005, nine percent of African-American applicants were interviewed, as compared to twenty percent of white applicants. That same year, 1.8% of African-American applicants were hired, as compared to 4.2% of white applicants. Overall, his analysis showed that a minimally qualified white person had a forty percent greater chance of being hired than a minimally qualified African American. Splitting his analysis by department, Killingsworth testified there was a statistically significant disparity between the percentage of African-American applicants hired and the percentage of white applicants hired in many of the departments, though not all. Killingsworth further testified that, once hired, white and African-American employees were treated differently in performance 11Notably, the CPS study mentioned by the majority did not perform regression analyses that excluded other possible variables that could account for the differences it reported. After pointing out the disparity between referrals and interviews, CPS acknowledged, “There could be . . . very legitimate reasons why the percentage of African Americans is reduced so dramatically between referral and interview” and acknowledged that “the team was unable to determine a definitive reason(s) for these outcomes.” 12Both Killingsworth and the State’s expert, Robert Miller, used applications—as opposed to applicants—as their unit of analysis. As the CPS study noted, it is difficult to identify the exact makeup of the applicant pool or the actual number of applicants because the State’s tracking system did not track individual people, but rather applications. Both the State’s and plaintiffs’ experts acknowledged African-American applicants applied on average for 1.6 more jobs than white applicants. In short, both parties agreed that African-American applicants, on average, followed a strategy of casting a wider State job search net than white applicants. — 62 — evaluations. Killingsworth did not connect his findings to any particular State practice. 13 Robert Miller, another labor economist and the State’s statistical expert, also performed regression analyses. With regard to wages and promotion, he found no statistically significant differences between races after taking into account experience and the pay grade of the job for which the individual applied. He criticized Killingsworth’s regressions as inadequate because they failed to adjust for these other factors. With regard to hiring, Miller subdivided his work into the three steps that the State followed to get to its actual hire. Thus, unlike Killingsworth, who considered only steps two and three and combined them, Miller separately considered steps one, two, and three. At step three, i.e., what occurs after the applicant is granted an in-person interview, Miller found no statistically significant difference between whites and blacks in their success in being hired. At step one, Miller found no statistically significant difference between white and black applicants in getting a referral. On a per application basis, though, African Americans were less successful than whites in getting a referral. However, the data also showed that AfricanAmerican applicants, on average, filed more applications than whites (approximately 5.9 versus 4.3 per applicant over a multiyear period—a thirty-eight percent difference). At step two—referral to interview—Miller’s findings were more nuanced. On the whole, he found that African Americans were 13Killingsworth acknowledged that he looked only at “[t]he total result,” not any particular employment practice other than “hiring” as a whole. He did not offer any opinions that the disparities he observed were the result of subjective as opposed to objective hiring practices. — 63 — statistically less likely to receive a department job interview after their application was referred by DAS. Probing more deeply, he found this was only true for about one-third of State departments. It was not true for the remaining two-thirds of departments. As Miller explained, [t]he overall conclusion with respect to step two is that a common factor could not or does not appear to be operating across all the departments in the same way, because our results show that there are clear departmental differences. For the specific jobs that the thirteen named plaintiffs had applied for, Miller found that African Americans were not statistically less likely to get interviews than whites. The plaintiffs reviewed paper hiring files produced by the State for 667 separate positions that were filled. Based on the examples in the record, these files contained items such as applicant résumés, score sheets for résumé reviews, score sheets for interviews, and letters asking for authority (and justifying the decision) to hire the successful applicant. For instance, for the position of workforce advisor in the unemployment insurance service center, Iowa Workforce Development used a résumé review worksheet that awarded a maximum of fifty-nine points. There were a maximum of twenty points potentially available for education, ten points potentially available for unemployment insurance claims experience, fourteen points potentially available for possessing various computer skills, five points available for being a veteran (or ten for being a disabled veteran), and five points available for “ability to follow resume and cover letter submission directions.” These files were not provided to or reviewed by Killingsworth. As Killingsworth put it, “I don’t have any access or haven’t had any hiring files.” — 64 — Miller testified that the data were “absolutely” capable of separation for analysis, and in fact, he separated them to the extent noted. II. Analysis. On appeal, plaintiffs do not argue that the State’s failure to follow its own equal-employment-opportunity policies constituted a discrete employment practice. Instead, plaintiffs challenge the district court’s determination that the State’s overall hiring process was capable of separation for analysis. Plaintiffs argue the hiring process could not be analyzed in terms of separate practices. Plaintiffs also assert on appeal that the district court erred in determining they failed to prove causation. Because the district court correctly decided the dispositive separation issue, we need not reach the causation issue. I will provide an overview of the governing law to place the separation issue into context. A. Disparate Impact Law. Title VII of the 1964 Civil Rights Act’s “central statutory purposes [are] eradicating discrimination throughout the economy and making persons whole for injuries suffered through past discrimination.” Albemarle Paper Co. v. Moody, 422 U.S. 405, 421 , 95 S. Ct. 2362, 2373 , 45 L. Ed. 2d 280, 299 (1975). To that end, Title VII seeks “to eliminate those discriminatory practices and devices which have fostered racially stratified job environments to the disadvantage of minority citizens.” McDonnell Douglas Corp. v. Green, 411 U.S. 792, 800 , 93 S. Ct. 1817, 1823 , 36 L. Ed. 2d 668, 676 (1973). “The ICRA was modeled after Title VII” and enacted to serve the same purposes: it “was passed in 1965 in an effort to establish parity in the workplace and market opportunity for all.” Vivian v. Madison, 601 N.W.2d 872, 873 (Iowa 1999). — 65 — Title VII and the ICRA each provide two principal ways to prove employment discrimination: disparate impact and disparate treatment. Hy-Vee Food Stores, Inc. v. Iowa Civil Rights Comm’n, 453 N.W.2d 512, 516 (Iowa 1990). The fundamental difference between the two theories is that “the disparate treatment theory focuses on the employer’s motivation; the disparate impact theory focuses on the consequences of the employer’s conduct.” Id. Disparate treatment requires a plaintiff to prove intentional discrimination. Id. Disparate impact, the theory plaintiffs presented at trial, presents an avenue for addressing inequalities in the absence of intentional discrimination. In a disparate impact case, what matters is not the subjective motivation of the employer, but the effects of an employment practice. Griggs v. Duke Power Co., 401 U.S. 424, 432 , 91 S. Ct. 849, 854 , 28 L. Ed. 2d 158, 165 (1971). Disparate impact prohibits employer practices “that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another.” Int’l Bhd. of Teamsters v. United States, 431 U.S. 324 , 335 n.15, 97 S. Ct. 1843 , 1854 n.15, 52 L. Ed. 2d 396 , 415 n.15 (1977). Meant to remove barriers to employment or promotion that are unrelated to job performance, the focus of disparate impact theory is on fairness in operation—not fairness in form. Griggs, 401 U.S. at 431 , 91 S. Ct. at 853 , 28 L. Ed. 2d at 164 (commenting that “equality of opportunity merely in the sense of the fabled offer of milk to the stork and the fox” is insufficient under Title VII). A three-step burden-shifting framework applies to disparate impact claims. In the first stage, to establish a prima facie case, a plaintiff must show that the employer “ ‘uses a particular employment practice that causes a disparate impact’ on one of the prohibited bases.” Lewis v. City — 66 — of Chicago, 560 U.S. 205, 212 , 130 S. Ct. 2191, 2197 , 176 L. Ed. 2d 967, 974 (2010) (emphasis omitted) (quoting 42 U.S.C. § 2000e-2(k)(1)(A)(i)). “Identifying a specific practice is not a trivial burden . . . .” Meacham v. Knolls Atomic Power Lab., 554 U.S. 84, 101 , 128 S. Ct. 2395, 2406 , 171 L. Ed. 2d 283, 296 (2008). Second, if the plaintiff makes this prima facie showing, the burden shifts to the employer to prove that the challenged employment practice reflects a business necessity. See 42 U.S.C. § 2000e-2(k)(1)(A)(i). Third, the plaintiff may rebut the employer’s business-necessity evidence by demonstrating there are “other reasonable alternatives that would have less adverse impact,” Hy-Vee, 453 N.W.2d at 518 , and the defendant “refuses to adopt such alternative employment practice,” 42 U.S.C. § 2000e-2(k)(1)(A)(ii). Only the first stage is at issue in this appeal because the district court held that plaintiffs failed to prove their prima facie case. The identification of a particular employment practice in the first stage helps the court ascertain and remedy the cause of racial disparities. Proof focused on a particular employment practice enables the relevant comparison between “qualified persons in the labor market and the persons holding at-issue jobs.” Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 650 , 109 S. Ct. 2115, 2121 , 104 L. Ed. 2d 733, 747 (1989), superseded by statute on other grounds, 42 U.S.C. § 2000e2(k), as recognized in Wal-Mart Stores, Inc. v. Dukes, 564 U.S. ___ , 131 S. Ct. 2541 , 180 L. Ed. 2d 374 (2011). The United States Supreme Court cautioned that using overbroad statistics to prove a disparate impact claim “would result in employers being potentially liable for ‘the myriad of innocent causes that may lead to statistical imbalances in the composition of their work forces.’ ” Id. at 657, 109 S. Ct. at 2125, 104 L. Ed. 2d at 751–52 (quoting Watson v. — 67 — Fort Worth Bank & Trust, 487 U.S. 977, 992 , 108 S. Ct. 2777, 2787 , 101 L. Ed. 2d 827, 843 (1988)). “Title VII guarantees [individuals] the opportunity to compete equally with white workers on the basis of jobrelated criteria,” but does not guarantee equal outcomes. Connecticut v. Teal, 457 U.S. 440, 451 , 102 S. Ct. 2525 , 2532–33, 73 L. Ed. 2d 130, 139 (1982). As the Wards Cove Court explained, if plaintiffs are allowed to use aggregated statistics alone to prove disparate impact, it is difficult for the Court to determine if the racial composition of hires is at odds with the relevant qualified labor market. 490 U.S. at 650–52, 109 S. Ct. at 2121–22, 104 L. Ed. 2d at 747–48. Wards Cove highlighted that racial disparities revealed in aggregated statistics at times could be justified by a closer examination of the qualified labor pool. Similarly, disparities shown by statistics aggregated at the departmental level may be explained by the specific employment practices of a given department. See Wal-Mart, 564 U.S. at ___ , 131 S. Ct. at 2555 , 180 L. Ed. 2d at 394 (“[I]nformation about disparities at the regional and national level does not establish the existence of disparities at individual stores, let alone raise the inference that a company-wide policy of discrimination is implemented by discretionary decisions at the store and district level.” (Internal quotation marks omitted.)). The particularity requirement allows the court to fashion an appropriate remedy: if a particular employer practice is identified as causing discriminatory impact, the court can order the employer to correct it. Title VII “arm[s] the courts with full equitable powers” in order to address violations. Albemarle, 422 U.S. at 418 , 95 S. Ct. at 2372 , 45 L. Ed. 2d at 297 . A court must be able to determine the cause of discrimination to effectively exercise these equitable powers. As the — 68 — United States Supreme Court has long recognized, the usefulness of statistics “depends on all of the surrounding facts and circumstances.” Int’l Bhd. of Teamsters, 431 U.S. at 340 , 97 S. Ct. at 1856–57, 52 L. Ed. 2d at 418 . It is vital to determine the particular employment practice causing the disparate impact in order to fix the problem. For example, the record shows that African Americans actually comprise a higher percentage of the State executive branch workforce than they do in the Iowa workforce as a whole. Yet, it would be wrong to conclude from that overall number that the State is not discriminating on the basis of race. One has to focus on actual employment practices. While Congress generally required that a plaintiff identify particular employment practices that cause disparate impact, Congress also provided that the decision-making process could be challenged as a whole under certain circumstances. Specifically, Congress provided: With respect to demonstrating that a particular employment practice causes a disparate impact . . . the complaining party shall demonstrate that each particular challenged employment practice causes a disparate impact, except that if the complaining party can demonstrate to the court that the elements of a respondent’s decisionmaking process are not capable of separation for analysis, the decisionmaking process may be analyzed as one employment practice. 42 U.S.C. § 2000e-2(k)(1)(B)(i). This law codifies an exception to the requirement that a plaintiff identify a particular employment practice, thereby ensuring that employers cannot avoid liability for disparate impact simply by making it difficult for a plaintiff to separately analyze the decision-making elements. See Griffin v. Carlin, 755 F.2d 1516, 1525 (11th Cir. 1985). This statutory exception balances the concern that employers could evade liability against the need for particularity by requiring the plaintiff to first demonstrate the employer’s process is incapable of separation. As the majority acknowledges, the burden was — 69 — on plaintiffs to prove the State’s decision-making process is incapable of separation for analysis before proceeding to attack the process as a whole. See Grant v. Metro. Gov’t of Nashville, 446 F. App’x 737, 740 (6th Cir. 2011). The ICRA does not have a counterpart to the 1991 congressional amendment. 14 Plaintiffs do not argue here—and did not argue below— that their burden to establish a prima facie case under the ICRA was lower than their burden under Title VII. I will therefore focus on federal law and on the question of whether the State’s employment practices were “not capable of separation for purposes of analysis.” But, first, I must respond to dicta in the majority opinion that misleads by omission and thereby unfairly disparages, sub silentio, our long-standing practice, followed in numerous decisions of this court, of relying on federal decisions to interpret equivalent provisions in the ICRA. The majority, in a discussion unrelated to its holding, distances itself from federal decisions. The majority, relying on Hubbard v. State, 163 N.W.2d 904, 909 (Iowa 1969), suggests federal interpretations of Title VII are only useful if those interpretations preceded the enactment of the Iowa statute. I disagree. Hubbard was decided a half century ago and interpreted the Iowa Tort Claims Act (ITCA), not the ICRA. See id. at 905 . Since 14An act of Congress amending Title VII does not amend the ICRA. The ICRA requires the plaintiff to prove a specific employment practice caused the disparate impact. Hy-Vee, 453 N.W.2d at 518 . The Iowa legislature has not amended the ICRA to add a provision in response to our decision in Hy-Vee or the congressional amendment to Title VII over twenty-three years ago. I would not read such an exception into the ICRA in the guise of judicial interpretation. See Ackelson v. Manley Toy Direct, L.L.C., 832 N.W.2d 678, 689 (Iowa 2013) (declining to reinterpret the ICRA to allow punitive damages in light of precedent disallowing punitive damages and legislative acquiescence). Whether to create such an exception in the ICRA is a policy choice to be made by the legislature. — 70 — Hubbard, our court has repeatedly relied on subsequent federal interpretations of the Federal Tort Claims Act to construe the ITCA. See Walker v. State, 801 N.W.2d 548, 569 (Iowa 2011) (Mansfield, J., dissenting) (collecting Iowa opinions that rely on federal cases decided after the ITCA’s enactment). Indeed, our court has cited Hubbard as support for the proposition that “[i]nterpretations of the federal act are instructive”—without limiting that observation to cases decided before the enactment of the ITCA. Annear v. State, 419 N.W.2d 377, 379 (Iowa 1988); see also Meier v. Sulhoff, 360 N.W.2d 722, 728 (Iowa 1985) (McCormick, J., dissenting) (citing Hubbard and stating “[b]ecause [the Iowa Occupational Safety and Health Act] is based on the federal model, the federal court interpretations constitute persuasive authority for giving a similar interpretation to our statute”); Adam v. Mount Pleasant Bank & Trust Co., 340 N.W.2d 251, 252 (Iowa 1983) (“Because our statute is based on the federal Tort Claims Act, we assume our legislature intended it to have the same meaning as the federal statute. Hubbard, 163 N.W.2d at 911 . Federal decisions interpreting the federal act are therefore entitled to great weight. Id. at 909 .”). In any event, our court, before today, has never relied on Hubbard to interpret the ICRA. The Iowa bench and bar has long understood federal authorities provide guidance to interpret the ICRA. This lends predictability to an important area of the law, particularly when the legislature has long acquiesced in our interpretations of the ICRA based on federal interpretations of the counterpart federal statutory language. See Ackelson v. Manley Toy Direct, L.L.C., 832 N.W.2d 678 , 688–89 (Iowa 2013) (discussing legislative acquiescence). Remarkably, the majority fails to acknowledge that our court decided many cases by relying on federal interpretations of equivalent statutory language in the civil rights — 71 — acts. See, e.g., Casey’s Gen. Stores, Inc. v. Blackford, 661 N.W.2d 515, 519 (Iowa 2003) (“[W]e have looked to the corresponding federal statutes to help establish the framework to analyze claims and otherwise apply [ICRA].”); Schlitzer v. Univ. of Iowa Hosps. & Clinics, 641 N.W.2d 525, 529 (Iowa 2002) (“The common goals of the Federal ADA and our civil rights act have encouraged us to look to the federal statutory and regulatory standards in applying our statute.”); Vivian, 601 N.W.2d at 873 (“The ICRA was modeled after Title VII of the United States Civil Rights Act. Iowa courts therefore traditionally turn to federal law for guidance in evaluating the ICRA.”); Vincent v. Four M Paper Corp., 589 N.W.2d 55 , 59– 60 (1999) (“[W]e have recognized the common purposes of the [F]ederal [ADA] . . . and the ICRA as well as the similarity in terminology of the statutes. Moreover, we have looked to the ADA and federal regulations implementing that [A]ct in developing standards under the ICRA for disability discrimination claims.” (Citations omitted.)); Hulme v. Barrett, 449 N.W.2d 629, 631 (Iowa 1989) (“Our court has ruled that civil rights cases brought under chapter [216] will be guided by federal law and federal cases.” (Internal quotation marks omitted.)); Probasco v. Iowa Civil Rights Comm’n, 420 N.W.2d 432, 435 (Iowa 1988) (“On several occasions, our courts have looked to the federal system for guidance in construing our similar civil rights legislation. We employ this approach again today because, as demonstrated below, the civil rights legislation and implementing rules involved in this case mirror those adopted on the federal level.” (Citations omitted.)); Iowa State Fairgrounds Sec. v. Iowa Civil Rights Comm’n, 322 N.W.2d 293, 296 (Iowa 1982) (“The parties assume we will find federal cases persuasive in selecting the analytical framework for deciding discrimination cases under the Iowa civil rights statute. This assumption is warranted by our prior decisions.”); Wilson72 Sinclair Co. v. Griggs, 211 N.W.2d 133, 139 (Iowa 1973) (noting “the similarity of language” of Title VII and the ICRA and relying on “numerous relevant federal decisions”); Iron Workers Local No. 67 v. Hart, 191 N.W.2d 758, 765 (Iowa 1971) (recognizing the ICRA contains “[a]nalogous language” to Title VII and “is another manifestation of a massive national drive to right wrongs prevailing in our social and economic structures for more than a century”). 15 None of these cases interpreting the ICRA limited consideration of federal authorities to those decided before enactment of the Iowa statute. In the majority’s view, if it does not like how federal decisions were decided, it can freely disregard them. The cost of this new approach is the stability and predictability of our law. See State v. Short, 851 N.W.2d 474, 475 (Iowa 2014) (Waterman, J., dissenting). After today, it is at best unclear what weight litigants and district court judges or the court of appeals should give federal cases when divining how our court will construe equivalent provisions in the ICRA. This is unfortunate. A more restrained majority would have deferred its pronouncements until a case in which they made a difference to the outcome. B. Does Substantial Evidence Support the District Court’s Finding that the State’s Decision-Making Process Was Capable of Separation for Analysis? The district court specifically found the State’s decision-making process was capable of separation for analysis. I agree with the majority that this is a question of fact. See, e.g., McClain v. Lufkin Indus., Inc., 519 F.3d 264, 278 (5th Cir. 2008). We are bound 15The same majority has ignored an even longer line of cases adhering to our court’s long-standing practice of relying on federal decisions under the Fourth Amendment when interpreting the nearly identical search-and-seizure provision in the Iowa Constitution. See State v. Short, 851 N.W.2d 474, 511 (2014) (Waterman, J., dissenting). — 73 — by the district court’s factual finding if it is supported by substantial evidence. Schlitzer, 641 N.W.2d at 529 . Plaintiffs argue on appeal there was no substantial evidence supporting this finding and that their evidence proved as a matter of law that the State’s decision-making process was incapable of separation for analysis. I disagree. Plaintiffs assert that the State’s record-keeping practices precluded separate analysis of the decision-making process because (1) subjective assessments pervaded the decision-making process; (2) subjective assessments have a “ripple effect,” whereby the discriminatory impact accumulates as applicants move through the hiring process; and (3) the State did not retain some records regarding applicant evaluations. The district court rejected all three theories based on the factual record developed at trial. To put the district court’s dispositive factual finding into its legal context, I construe the operative statutory language. The phrase “each particular challenged employment practice” in 42 U.S.C. § 2000e-2(k)(1)(B)(i) indicates that a particular employment practice is distinct from a decision-making process. A particular employment practice is considered an element within the larger decisionmaking process. Congress’s choice to use the singular form, combined with the words “particular” and “each” demonstrates that the challenged practice must be individually identified. “This syntax would be strange if a plaintiff could bundle a number of discrete steps of a multi-phase hiring process together, based on a common characteristic.” Davis v. Cintas Corp., 717 F.3d 476, 496, 497 (6th Cir. 2013) (holding plaintiff “did not identify a ‘particular employment practice’ within the meaning of Title VII by pointing to all of the subjective elements in the [employer’s hiring system]”). — 74 — I first consider plaintiffs’ contentions regarding subjective decisionmaking. How subjective decision-making plays into hiring and promotion depends on the type of job and the process used to fill it. Thus, subjective conduct may serve as the “particular employment practice” underlying a disparate impact claim if plaintiffs can prove that the conduct operates uniformly throughout an employer’s decisionmaking process to cause a disparate impact. For example, in Davis, the Sixth Circuit rejected plaintiff’s argument that the defendant’s decisionmaking process was incapable of separation when “not all of the system’s subjective elements are the same.” Id. at 497 . The court noted “[e]ach different interview . . . has a specific interview guide, and different managers conduct interviews at different stages of the process.” Id. Likewise, in Bennett v. Nucor Corp., the Eighth Circuit concluded “this is not a case where the components of the employer’s selection process were incapable of separation.” 656 F.3d 802, 817 (8th Cir. 2011). The court pointed to the fact that the employer’s “five departments used a variety of measures to evaluate candidates for promotion, including objective criteria like experience, training, disciplinary history, and test scores, and subjective criteria such as interview performance and the opinion of the candidate’s current supervisor.” Id. at 817–18; see also Grant, 446 F. App’x at 740 (“The problem, however, is that Plaintiffs make no effort to isolate any of these [decision-making] practices or to examine their individual effects on the promotions process.”). Chin v. Port Authority is a good example of a subjective process that was incapable of separation for purposes of analysis. 685 F.3d 135 , 154–55 (2d Cir. 2012). In that case, the plaintiffs had passed exams and had been placed on eligibility lists for promotions to sergeant but had not received promotions. Id. at 142–43. The plaintiffs proved that the — 75 — decisions as to whom would be promoted from the eligible list were based on essentially subjective recommendations by commanding officers and subjective final decisions by the superintendent. Id. at 154–55. The process was entirely discretionary, and the final decision rested with one person—the superintendent. See id. Wal-Mart, filed shortly before this case went to trial, is instructive. There, the Supreme Court decertified a nationwide class of 1.5 million current and former employees of Wal-Mart alleging gender discrimination. Wal-Mart, 564 U.S. at ___ , ___, 131 S. Ct. at 2547, 2561 , 180 L. Ed. 2d at 385, 400 . Pay and promotion decisions at Wal-Mart were generally committed to local managers’ broad, subjective discretion. Id. at ___, 131 S. Ct. at 2547 , 180 L. Ed. 2d at 385 . However, plaintiffs alleged that a strong and uniform corporate culture led that discretion to be exercised, even subconsciously, against the hiring and advancement of women. Id. at ___, 131 S. Ct. at 2548 , 180 L. Ed. 2d at 386 . Yet, the Court held “[r]espondents [did] not identif[y] a common mode of exercising discretion that pervades the entire company.” Id. at ___, 131 S. Ct. at 2554–55, 180 L. Ed. 2d at 393 . The district court observed Wal-Mart “highlights both the need to identify a particular employment practice, the pertinence of discretionary decisionmaking in the employment process, and the interconnection with statistical proof.” The district court here found that “[t]he State’s system has both objective and subjective components” which are “not so confused . . . as to prevent Plaintiffs from honing in on one particular employment practice.” This finding is supported by the record. For example, the DAS screen that occurred at step one, the résumé score sheets that were part of step two in some departments, the second résumé screens and spelling and grammar screens that were part of step two in some — 76 — departments, and the interview score sheets that were part of step three in some departments were objective components that could have been separately analyzed. This was not a purely subjective process. Furthermore, while there undoubtedly was subjectivity and—as the plaintiffs credibly demonstrated—implicit bias in multiple State hiring decisions during the relevant time period, this case is a far cry from Chin, in which the decision-making process ultimately came down to a single individual’s discretion. By contrast here, the State’s hiring decisions were in the hands of numerous department managers. The State of Iowa Executive Branch employs a far more diverse range of job categories than any of the defendants in the cases in which subjective decision-making has served as a basis for attacking the decision-making process as a whole. Those working for the State include corrections officers, parole officers, registered nurses, food service workers, power plant engineers, state troopers, and DOT road-maintenance crew members, to name a few. As noted by the district court, “The State system can be dissected into numerous decision-making stages among numerous independent agencies of the executive branch,” and there is “inconsistency in results among the numerous agencies.” Miller showed that, at step two, African Americans had a lower statistical likelihood of getting an interview in some departments but not in others. These outcomes could be due to either objective or subjective considerations, but either way they do not suggest the existence of a common practice (even a subjective one) that would justify class-wide relief. See Wal-Mart, 564 U.S. at ___ , 131 S. Ct. at 2555–56, 180 L. Ed. 2d at 393 (noting that discretion when exercised by different managers in different ways is not in itself an employment practice). — 77 — No witness affirmatively testified the process was incapable of separation for analysis. To the contrary, the State’s expert, Miller, “emphatically” testified that the State’s hiring system was capable of separation for analysis. Plaintiffs on appeal do not challenge the admissibility of Miller’s testimony. His testimony alone constitutes substantial evidence supporting the district court’s finding. Furthermore, Miller actually did separate the process into the three steps for purposes of his analysis. Additionally, plaintiffs’ own experts testified that the State’s decision-making process could be separated for analysis. The district court noted that Greenwald conceded the State’s employment process could be separated to individually analyze each step of the process and commented “that one ‘could determine whether or not there was bias at each one of the independent stages of the hiring process.’ ” Kaiser discussed how written résumés and in-person interviews can trigger implicit racial biases differently. The State’s statistical expert, Killingsworth, utilized a regression analysis to evaluate the State’s hiring process using data from different stages of the process and different departments. The district court summarized: Dr. Killingsworth was capable of separating data for the referral stage, the interview stage, and the hiring stage for African Americans as compared to whites over a period of years. His work permits a fact finder to analyze the departments of the executive branch in each of those years at each of those stages. This charting of State data allows a fact finder to compare the various departments and draw important conclusions as to how the individual departments compare to each other at the various stages. While he elected not to begin his analysis at the application stage, the data available would permit this. And it could be used to track applicant flow from that first stage to the hiring of one applicant for the specific job opening in any given department—including the progress of each applicant through the various stages and examining the particular screening-devices used. — 78 — Further, Killingsworth “could limit his models by new applicants or incumbent State employees, initial pay, and performance evaluations.” However, he did not correlate any of his findings to a particular screening device. Significantly, Miller performed separate statistical analyses by department on the three steps in the hiring process. In fact, as I discuss below, the NAACP relies on that analysis in pursuing an alternative argument for reversal. Furthermore, Killingsworth, as quoted above, agreed with Miller that the three steps could be separated for purposes of analysis. Additionally, in four pages of findings on separability, the district court repeatedly referred to expert testimony that indicated the stages of the process could be separated. This does not foreclose the possibility of further separation, as the district court found. For example, the record revealed other employment practices with potentially discriminatory effects such as résumé screens could have been isolated and separately analyzed. But, in any event, the record supports the district court’s finding that the plaintiffs did not meet their burden. Plaintiffs did not even attempt to challenge an actual employment practice and simply argued that the “total result” (to use Killingsworth’s phrase) was discriminatory. In response to plaintiffs’ “ripple effect” argument, the district court found “the fact that one errant practice compounds a problem at a later stage of the process does not prevent investigation of either the earlier or later separate stage or practice.” I agree. Though the use of a discriminatory criterion at one stage may impact applicants throughout a decision-making process, such a “ripple effect” does not preclude separate analysis. In some cases, subjective and objective criteria may be so intertwined as to prevent separation. See McClain v. Lufkin Indus., — 79 — Inc., 187 F.R.D. 267, 275 (E.D. Tex. 1999) (finding an employer’s actions inseparable for analysis when “[t]he disparate impacts begin on the day one is hired and are potentially magnified each time one’s career . . . intersects a subjective decision-making process”). But, plaintiffs here have not proven this is such a case. The very point of regression analysis is to allow isolation of particular elements and determine whether there is a “ripple effect.” Finally, substantial evidence supports the district court’s finding that the plaintiffs failed to show the condition of the paper files precluded separate analysis of specific employment practices within the State’s hiring process. Killingsworth never looked at the hiring files. Significantly, as the majority notes, the district court found “the hiring files themselves permit a focused view of the different screening-devices and practices in referral, interview or hiring of applicants for any given job between the departments.” Plaintiffs make much of the missing documents from the files, but eighty-six percent of the files included interview questions, eighty-one percent had interview notes, and seventythree percent provided an interview scoring matrix. Half of the files had reference checks. Eighty-four percent of the files also contained an individual’s application, résumé, and cover letter. As the majority recognizes, Greenwald commented, “The hiring files of the State are a gold mine that hasn’t been analyzed.” There is no evidence that the plaintiffs took even one of the objective standards the State employed and tried to determine whether it had a disparate impact using the available records. Plaintiffs argue that the sample sizes would get smaller and statistical reliability would decline as one tries to analyze the effect of a practice that was only employed in certain areas at certain times. This — 80 — may be correct, but does not excuse the failure to try. The statute does not permit courts to aggregate a collection of different hiring practices across different times and departments just to increase the size of the sample. The only case plaintiffs cite in support of their argument that a lack of records can prevent separate analysis is the district court opinion in Chin. Notably, the Second Circuit on appeal relied on a different rationale from the district court—i.e., that the process was basically entirely subjective and the final decision rested with one person. Chin, 685 F.3d at 154–55. Stepping back and looking at Chin with the benefit of both opinions, the lack of records and the subjectivity of the process appear to be two sides of the same coin: No one documented why someone received a promotion because there was nothing to document. Port Auth. Police Asian Jade Soc. of N.Y. & N.J. Inc. v. Port Auth., 681 F. Supp. 2d 456 , 460–61, 464–65 (S.D.N.Y. 2010) (finding the decisionmaking process could not be separated “both because records do not exist for every step and because the causal role of each step is called into doubt by the records that do exist”). Here, by contrast, it is undeniable that the records were incomplete, but equally irrefutable that no one who tried to analyze the records was unable to do so. As the district court found, [t]he State’s data—its recordkeeping—while not perfect, was sufficient for both Dr. Killingsworth and Dr. Miller to conduct their analyses. The presence in the record of their models and opinions dispels the argument that the State’s recordkeeping is such that it precludes anything but a “systemic employment practice.” I conclude substantial evidence supports the district court’s findings regarding the State’s record keeping. — 81 — I reiterate the importance of separately analyzing the different processes used by the various departments. In some departments, African-American applicants fared better than white applicants at certain stages; in others they fared worse, even much worse. This suggests that the different screening processes used by the departments may have had different impacts on applicant success. As the district court noted, these “[v]arying outcomes between the departments and stages of the process invite[] localized scrutiny.” For example, based on the record in this case, I have concerns about the various résumé screening devices used by State departments at the step-two level. It is certainly possible that inappropriate screening devices may have been used in some of the departments in which Miller found a statistically significant disparity between blacks and whites at step two. But, it is just a possibility and not an aspect of the case that the plaintiffs chose to pursue. Here, the district court observed that “the hiring files themselves permit a focused view of the different screening-devices and practices in referral, interview or hiring of applicants for any given job between the departments.” For example, as the court pointed out, DAS has a systemwide applicant screening manual, and an analysis could have been performed based on the manner in which DAS instructs managers on the use of the manual. The district court went on to comment that “one can focus on any number of discrete employment decisions made as individual, separable, identifiable particular employment practices” and then gave two more examples: One example of the separability of the process is the “second résumé screen” that had been utilized by some departments. It was a particular employment practice that — 82 — was evaluated, determined to be inappropriate, and curtailed at the suggestion of DAS. Similar refinement of the hiring process by focusing on the inappropriate use of “spelling and grammar screening” is another example of DAS having addressed a particular employment practice. The record reflects not only the ability to focus on these particular employment practices but when and which separate agencies responded to the suggested changes by DAS. The foregoing has shown why I am confident the court reached the right conclusion. The district court methodically went through the record, focused appropriately on the testimony of statistical experts for each side, and identified various employment practices that could have been separately analyzed, including the three steps in the employment practice (separately analyzed by Miller) and more specific practices within those steps. For all these reasons, I conclude substantial evidence supports the district court’s finding that the plaintiffs failed to show the State’s employment practices are not capable of separation for purposes of analysis. I close with three observations. First, it is significant that the NAACP, in a well-argued amicus brief, relies on some of the same data that plaintiffs dismiss as inadequate. Thus, the NAACP asserts that Miller’s findings show there was an adverse impact at step two in eight departments that employed approximately fifty-eight percent of the State workforce. On that basis, it asks us to reverse the district court. In my view, the NAACP’s brief raises serious questions as to whether the State committed unlawful discrimination. The problem with this argument is that it is not the case the plaintiffs elected to pursue. For instance, we do not know what practices were followed in those eight departments during step two. This seems like a relatively straightforward inquiry that could have been pursued in discovery. We — 83 — also do not know which representative plaintiffs—if any—applied for jobs with those departments. And, the remedies sought by the plaintiffs would apply not only to those departments but to the State as a whole. Instead of narrowing their focus, plaintiffs brought a class action alleging a common pattern of discrimination by the entire state executive branch of government. Having brought such a large case, it was then up to the plaintiffs to undertake the considerable work required to prove it. Under the prevailing law, this included analysis of specific hiring practices and their impact. Plaintiffs did not meet their burden. Second, I do not downplay what this case has shown. Even according to Miller, it appears African Americans on the whole were disadvantaged in getting job interviews from some agencies, including some large departments like the department of human services and the department of transportation. This conclusion, from a defense expert, is disturbing although inconclusive. The district court, in my view correctly, questioned why “given all this data held by the State, it did not on a regular basis review it, as did these experts, with an eye toward measuring impact.” Third, it bears emphasis that the defeat of this class action does not bar a person who believes he or she was a victim of discrimination from bringing an individual lawsuit on his or her own against the State for new acts of discrimination. What is clear here is that plaintiffs failed to prove, because they ultimately did not try to prove, that the State of Iowa engaged in specific employment practices that had discriminatory effects against African-American job applicants and that would allow for class-wide relief. For the foregoing reasons, I agree the district court’s judgment must be affirmed. Mansfield and Zager, JJ., join this special concurrence.