OPINION AND ORDER

JOAN HUMPHREY LEFKOW, District Judge.

Twelve construction workers filed this putative class action against Walsh Construction Company ("Walsh"), alleging that Walsh discriminates against black employees in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq. ("Title VII"), and the Civil Rights Act of 1866, 42 U.S.C. § 1981, as amended. Before the court are (1) Walsh's motion to bar the expert report and testimony of plaintiffs' expert, Stan V. Smith; and (2) Walsh's motion to strike the Equal Employment Opportunity Commission ("EEOC") Investigative Memorandum and EEOC Determinations that plaintiffs have submitted in support of their motion for class certification. For the following reasons, Walsh's motion to strike Smith's report and testimony [#135] will be granted in part and denied in part, and Walsh's motion to strike the EEOC Investigative Memorandum and EEOC Determinations [#138] will be denied.1

ANALYSIS2

I. Motion to Strike Expert Report and Testimony of Stan V. Smith

Walsh has moved to exclude plaintiffs' expert's testimony and report as inadmissible under Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L. Ed. 2d 469 (1993). Because Smith's report is critical for class certification, the court must rule on its admissibility before ruling on plaintiffs' motion for class certification. See Messner v. Northshore Univ. HealthSystem, 669 F.3d 802, 812 (7th Cir. 2012) (citing Am. Honda Motor Co. v. Allen, 600 F.3d 813, 815-16 (7th Cir. 2010)).

To admit expert testimony, the court must find that the expert is proposing to testify to (1) valid scientific, technical, or other specialized knowledge, and (2) his testimony will assist the trier of fact to understand or determine a fact in issue. See, e.g., Durkin v. Equifax Check Servs., Inc., 406 F.3d 410, 420 (7th Cir. 2005) (quoting Ammons v. Aramark Unif. Servs., Inc., 368 F.3d 809, 816 (7th Cir. 2004)).3 "The first prong tests the reliability of the testimony; the second prong tests its relevance." Frey v. Chi. Conservation Ctr., 119 F.Supp.2d 794, 797 (N.D. Ill. 2000). The objective "is to make certain that an expert, whether basing testimony upon professional studies or personal experience, employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field." Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 152, 119 S.Ct. 1167, 143 L. Ed. 2d 238 (1999). The proponent of the testimony bears the burden of proving that the proffered testimony meets these requirements. Frey, 119 F. Supp. 2d at 797. "Determination on admissibility should not supplant the adversarial process; `shaky' expert testimony may be admissible, assailable by its opponents through cross-examination." Gayton v. McCoy, 593 F.3d 610, 616 (7th Cir. 2010).

A. Evidence of Discriminatory Hiring Practices

Smith concludes that Walsh's time card data support the conclusion that Walsh's hiring and discharge practices discriminate against black journeymen and foremen. (Smith Rep't at 26.) Smith bases his conclusion on the fact that the percentage of black journeymen and foremen at Walsh decreased significantly between 2000 and 2009, from 25 percent to 11 percent. In addition, the absolute number of black journeymen and foremen decreased significantly, from slightly over 160 black employees in 2000 to approximately 50 in 2009. According to Smith, these changes support an inference of discrimination because the black population in Cook County was a constant 26 percent during this same time period. (Smith Rep't at 20-26.)

Walsh argues, and the court agrees, that Smith's analysis does not consider the relevant labor market. "The identification of a relevant labor market-the key issue in a class-based Title VII case-means not only identifying qualified potential applicants for the job at issue but also identifying interested potential applicants." EEOC v. Chicago Miniature Lamp Works, 947 F.2d 292, 302 (7th Cir. 1991) (citing Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 650, 109 S.Ct. 2115, 104 L. Ed. 2d 733 (1989));4 accord Bennett v. Roberts, 295 F.3d 687, 697 (7th Cir. 2002); see also Janes v. Chicago Bd. of Ed., No. 00 C 6128, 2002 WL 31557619, at *5 (N.D. Ill. Nov. 18, 2002). Smith's analysis fails to consider potential applicants' commuting distance or the location of Walsh's large construction projects, both of which would affect the size of the pool of interested potential applicants.5 See Bennett, 295 F.3d at 697 (expert report not reliable because, among other flaws, it did not account for commuting patterns within the Chicago metropolitan area); Chicago Miniature Lamp Works, 947 F.2d at 302 (use of Chicago as relevant labor market, with no weighting of distance, meant that no consideration was given to the fact that blacks living on the South and West Sides of Chicago would be less willing to apply for jobs because of commuting time involved); Ramona L. Paetzold & Steven L. Willborn, The Statistics of Discrimination § 4.4 (2011-2012 ed.) ("[C]ourts seek to identify areas from which the employer would reasonably be most likely to obtain applicants. Factors such as commuting distance and availability of public transportation should be taken into account. . . ."); cf. Mister v. Ill. Cent. Gulf R.R. Co., 832 F.2d 1427, 1429-31 (7th Cir. 1987) (expert examined the number of black applicants for positions at a railroad company, and therefore it was clear that the data included only those individuals who were interested in the positions at issue). Indeed, many of the construction projects at issue here were located outside Cook County. Another problem is that the black population in Cook County includes children, retired persons, and other individuals who are not considered to be part of the labor market. In addition, Smith failed to consider physical ability to do construction work, for which age would likely be a proxy, or the percentage of blacks in Cook County with prior construction experience. See Mister, 832 F.2d at 1431. Indeed, according to the 2000 census data, blacks account for 15.6 percent of construction laborers in Cook County. (Topel Rep't at 14.) This is substantially lower than their population share, which Smith asserts is 26 percent. As of 2000, the percentage of blacks employed at Walsh was greater than the percentage of black laborers in Cook County. (Id.)

Smith offers no compelling justification for his failure to consider these relevant factors in defining the relevant labor market. He simply asserts that commuting distance is insignificant because workers can take public transportation or drive on freeways to get to construction sites throughout Chicago. (Smith Rebuttal Rep't at 13.) The court need not accept this assertion because Smith has cited no supporting data. See Bennett, 295 F.3d at 697 (expert's theory could be rejected where he did not attempt to verify that it was reasonable to assume that Naperville school would hire applicants from within the state of Illinois); Chicago Miniature Lamp Works, 947 F.2d at 302 (criticizing district court judge for "uncritically accept[ing] the assertion of the EEOC's expert that one and one-half hours was a reasonable commuting time for a low-paying entry-level job"). Because Smith failed to take several key factors into account in determining the relevant labor market, his analysis of Walsh's hiring data will not assist a trier of fact. Section VII.A of Smith's expert report will not be admitted.

B. Evidence of Decline in Work Hours

Smith also concludes that a decline in hours worked by black journeymen and foremen supports the conclusion that Walsh systematically discriminated against blacks. (Smith Rep't at 27.) Smith analyzed the total number of weighted (regular plus overtime) hours for black and non-black journeymen and foremen and concludes that black hours represented 17 percent of Walsh employees' hours in 2000 but only 11 percent in 2009. (Id. at 26.) He concludes that this trend supports the inference of discrimination because the population of blacks in Cook County stayed the same during the relevant time period. (Id.)

As discussed above, the black population in Cook County is not the relevant comparison population. Smith's failure to take relevant factors such as geography, commuting distance, ability, and interest into account critically undermines the usefulness of his analysis. Section VII.B of Smith's expert report will not be admitted.

C. Evidence of Shorter Tenure

Smith analyzed Walsh's data regarding employee tenure and concludes that the data are consistent with plaintiffs' allegations that Walsh's employee discharge practices were systematically discriminatory against blacks. (Smith Rep't at 33.) On average, black employees had a tenure of 1.6 years and non-black employees had a tenure of 2.2 years. (Id. at 30.) In addition, Smith's analysis shows that the ratio of the number of black employees to the number of non-black employees falls as tenure increases. (Id. at 34.) For the first year of tenure, the ratio of blacks to non-blacks is .294, and by the sixth year the ratio is .139. (Id.) Smith concludes that the decline in the number of black employees is consistent with plaintiffs' allegations that Walsh treated blacks less favorably "in the terms and conditions of employment." (Id.)

Walsh argues, citing Topel's report, that Smith's analysis is not reliable because it fails to consider whether employees were terminated for cause. Walsh's data includes a "do not rehire" flag when Walsh terminated an employee for cause. (Topel Rep't at 19.) Topel analyzed data with the "do not rehire" flag and concluded that there is no significant difference in the rate at which black and non-black workers lose their rehire eligibility. (Id. at 19.) Smith does not dispute that the rate of terminations for cause is similar for black and non-black employees. (See Smith Rebuttal Rep't at 16.) Walsh is correct that Smith's analysis is neither reliable nor useful to the extent that plaintiffs rely on it in support of their claim that Walsh's discharge practices discriminated against blacks. Moreover, Smith's analysis looks at data relating to all black employees at Walsh. Plaintiffs' proposed tenure class, however, only includes journeymen. Therefore the analysis would not be useful to the trier of fact.

Smith's additional conclusion that Walsh likely discriminated against blacks in the "terms and conditions of employment" is too vague to be useful to a trier of fact. As will be discussed infra, the plaintiff in an employment discrimination case "must begin by identifying the specific employment practice that is challenged." Wal-Mart Stores, Inc. v. Dukes, ___ U.S. ___, ___, 131 S.Ct. 2541, 2556 (2011) (quoting Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 994, 108 S.Ct. 2777, 101 L. Ed. 2d 827 (1988)). Smith's broad reference to the "terms and conditions of employment" does not allow the trier of fact to determine, with any specificity, which of Walsh's employment practices plaintiffs challenge. In addition, as noted above, Smith's analysis fails to distinguish between black journeymen and other employees; therefore it does not provide useful support for plaintiffs' class claims. Sections VII.C and VII.D of Smith's report will not be admitted.

D. Evidence of Disparities in Work Hours, Compensation, and Promotions

Smith concludes that (1) on average, black journeymen worked 25 percent fewer weighted hours than non-blacks during the relevant time period, (2) black journeymen received fewer weighted hours relative to non-black journeymen as a job progressed (declining from 18 percent to 14 percent), (3) a smaller percentage of black employees were foremen (3 percent of blacks compared to 8 percent of non-blacks), and (4) fewer black journeymen were promoted to foreman (2.1 percent of blacks as compared to 4.7 percent of non-blacks working during the same period). (Smith Rep't at 33-43.) For each analysis, Smith compared employees who were in the same union, held the same laborer or foreman position, and worked during the same year. (Smith Rebuttal Rep't at 7.) Walsh argues that Smith's analysis is not reliable because he failed to conduct a multiple regression analysis to account for non-discriminatory factors that might cause the disparity between black and non-black employees.6

Usually, "the failure to include variables will affect the analysis' probativeness, not its admissibility." Adams v. Ameritech Servs., 231 F.3d 414, 423 (7th Cir. 2000) (quoting Bazemore v. Friday, 478 U.S. 385, 400, 106 S.Ct. 3000, 92 L. Ed. 2d 315 (1986)).7 A statistical analysis does not need to take all possible variables into account, but rather must consider the major variables that might explain the disparity. Bazemore, 478 U.S. at 400; People Who Care v. Rockford Bd. of Educ., Sch. Dist. No. 205, 111 F.3d 528, 537-38 (7th Cir. 1997); Mozee v. Am. Commercial Marine Serv. Co., 940 F.2d 1036, 1045 (7th Cir. 1991). Expert evidence is only excluded where the regression is "so incomplete as to be inadmissable as irrelevant," Bazemore, 478 U.S. at 400 n.10, or where it fails "even to make the most elementary comparisons." People Who Care, 111 F.3d at 537-38. Accordingly, the court must evaluate "what [the expert] did, rather than hypothetical tests that he or another expert might have done." Adams, 231 F.3d at 425.

Walsh argues that Smith's work hours, compensation, and promotion analysis should have considered the location of the various job projects, the crews to which laborers were assigned, the distance from the workers' residence to the construction site, the union stewards' first right to overtime, and the voluntary nature of overtime. (Walsh Reply at 5-12.) Topel asserts that Smith should also have considered employees' marital status, age, seniority, and commuting distance to the work site in analyzing the number of hours worked. (Topel Rep't at 21.)8

The factors listed by Walsh and Topel, though relevant, are not so "major" that Smith's failure to include them renders his analysis inadmissible. See Bazemore, 478 U.S. at 400. The employment practices at issue are not complex. Smith's data set already controls for employees' positions and union membership. As alleged by plaintiffs, all of the relevant employment decisions — assignment of work hours, overtime, and promotions — are left to the discretion of Walsh's managers. There are no established criteria or qualifications. Compare with Sheehan v. Daily Racing Form, Inc., 104 F.3d 940, 942 (7th Cir. 1997) (expert failed to consider any potential explanatory variables other than age in age discrimination case, in particular familiarity with computers, and as "[e]veryone knows that younger people are on average more comfortable with computers than older people are"); Carpenter v. Boeing, 456 F.3d 1183, 1198-99 (10th Cir. 2006) (where overtime qualifications were established in collective bargaining agreement, court properly excluded report that failed to compare qualified men to qualified women). If, as plaintiffs allege, Walsh's managers exercise their discretion to offer overtime to non-black employees instead of black employees, then the voluntary nature of overtime will not be a significant factor in the analysis. Moreover, neither Walsh nor Topel explains how the available data would even allow Smith to control for the voluntary nature of overtime.

Likewise, the court is not convinced that workers' commuting distance to the job site, age, or marital status are salient factors for Smith's analysis of the number of hours worked. Topel performed a regression analysis that took these factors into account, see Topel Rep't at 22, but it is difficult to compare Topel's analysis to Smith's because Topel, unlike Smith, analyzed white employees and Hispanic employees as though they were two separate groups.9 It is significant, however, that there is no evidence that work or overtime assignments at Walsh are related to age or seniority. In addition, Topel's emphasis on the importance of commuting distance is undermined by the assertion, made elsewhere in his report, that Walsh's construction sites are likely to employ workers from the surrounding community. (See Topel Rep't at 16 (noting that "the Dan Ryan renovation ran directly through the African-American neighborhoods of the south side of Chicago, so the available supply of Black laborers was huge").) Commuting distance would likely not be a significant factor if most workers on a particular job site live nearby. Mindful of the fact that the failure to include variables generally goes to probativeness, and not admissibility, see Adams, 231 F.3d at 423, the court will deny Walsh's motion to strike Smith's work hours, compensation, and promotion analysis.

E. Hedonic Damages

Finally, Walsh objects to Smith's calculation of hedonic damages, which Smith describes as the damages attributable to the class members' "loss of enjoyment of life" as a result of Walsh's discriminatory employment practices. His method is based on a "willingness to pay" model for calculating the value of a life, which looks at consumer purchases, wage risk premiums, and regulatory cost-benefit analysis to determine a value that society places on an individual human life. (Smith Rep't, App'x E at 1.) Smith concludes that the statistical value of a human life is $4.2 million and estimates that black Walsh employees would sustain a 10 percent loss of enjoyment of life as a result of racial discrimination. (Id. at 2.) In support, Smith cites an article that describes a conceptual approach for applying estimates of the loss of the pleasure in life (referred to as hedonic damages) in personal injury cases. See Edward P. Berla, Michael L. Brookshshire & Stan V. Smith, Hedonic Damages and Personal Injury: A Conceptual Approach, J. OF FORENSIC ECONOMICS, Vol. 3, No. 1, pp. 1-8 (1990). As Smith admitted in his deposition, however, he is aware of no studies that apply hedonic damages in the context of a hostile work environment or otherwise discriminatory environment. (Smith Dep. at 173-74.) His estimate of a 10 percent loss in enjoyment of life is an assumption that is not subject to any scientific testing, and the estimate may vary depending upon the fact-finder's determination of individualized damages. (Id.; Smith Rebuttal Rep't at 18-19.) Plaintiffs have cited no case or peer-reviewed article where hedonic damages were used to determine the "loss of enjoyment of life" that results from employment discrimination. Smith's calculation of hedonic damages will be excluded.

F. Supplemental Statistical Analysis in the Smith Declaration

In response to Walsh's motion to strike, plaintiffs submitted a thirty-page declaration from Smith that includes substantial additional statistical analysis. Walsh filed a motion to strike Smith's declaration because the additional analysis violates the parties' Joint Scheduling Plan, which required the submission of rebuttal expert reports by January 21, 2011. See Dkt. #120. The court denied Walsh's motion to strike, based in part on the representations of plaintiffs' counsel that any new analysis was merely adding "more meat on the bone" of Smith's original expert report.

After reviewing Smith's declaration in more detail, the court agrees with plaintiffs that the same assumptions underlie Smith's original expert report and supplemental declaration. Therefore the new analysis contained in Smith's expert report does not affect the court's rulings on Walsh's motion to strike.

First, Smith conducted additional analysis regarding Walsh's alleged "wrongful dismissal" of black employees. He again analyzed the ratio of the tenure of black journeymen as compared to non-black journeymen. (Smith Decl. at 23-24.) Smith concludes that "the difference in ratios statistically is far greater than two standard deviations . . . [i]t is in fact 11 standard deviations. . . . This means that the probability that the adverse impact against Blacks arose from random chance is far less than one in ten thousand, showing a very strong statistically significant adverse impact." (Id. at 23.) Therefore, according to Smith, Walsh "wrongfully dismissed" more black journeymen than non-black journeymen. Smith's supplemental analysis suffers from the same flaws as his expert report. He failed to consider whether black journeymen were terminated for cause. Given that the termination-for-cause data clearly shows that blacks were not terminated at a higher rate than non-blacks, Smith's assertion that blacks suffered disproportionately from "wrongful dismissal" is not helpful to the trier of fact because is unclear what the phrase "wrongful dismissal" refers to. The data shows that black journeymen and foremen had shorter tenure at Walsh than non-blacks. They do not show that blacks were "dismissed" or "terminated" at a higher rate. Smith's supplemental tenure analysis does not overcome the substantial flaws in Smith's original tenure analysis.

Second, Smith performed new statistical analysis regarding Walsh's alleged failure to promote black journeymen to the foreman position. This analysis looks at substantially the same factors that were relevant to Smith's original analysis of Walsh's promotion data. It does not change the court's conclusion that Smith's analysis of the promotion data is admissible.

Third, Smith performed additional statistical analysis of the change in percentage of black Walsh employees over time. In doing so, Smith again assumed that the relevant labor market is the black population in Cook County. As explained supra, this assumption is incorrect. The supplemental analysis is unreliable for the same reasons as Smith's original analysis.

II. Motion to Strike EEOC Investigative Memorandum and EEOC Determinations

Walsh also moves to strike the EEOC Investigative Memorandum and EEOC Determinations on the ground that they contain inadmissible hearsay. In general, administrative findings regarding claims of discrimination are admissible under Federal Rule of Evidence 803(8) so long as "neither the source of information nor other circumstances indicate a lack of trustworthiness." Fed. R. Evid. 803(8)(B); Young v. James Green Mgmt., Inc., 327 F.3d 616, 624 (7th Cir. 2003) (citing Chandler v. Roudebush, 425 U.S. 840, 863 n.39, 96 S.Ct. 1949, 48 L. Ed. 2d 416 (1976)). District courts are also permitted to determine, on a case-by-case basis, what EEOC investigatory materials should be admitted at trial. Tulloss v. Near N. Montessori Sch., Inc., 776 F.2d 150, 152-55 (7th Cir. 1985). Given that this case is only at the class certification stage, it is premature to exclude the EEOC Investigative Memorandum and EEOC Determinations. The court is not determining the merits of plaintiffs' claims and there is no risk of jury confusion or unfair prejudice. Moreover, the facts contained in the EEOC Investigative Memorandum are largely consistent with the evidence that plaintiffs have submitted in support of class certification. Walsh has cited nothing that indicates a lack of trustworthiness. Walsh's motion to strike the EEOC Investigative Memorandum and EEOC Determinations will be denied without prejudice.

CONCLUSION AND ORDER

Walsh's motion to strike Smith's report and testimony [#135] is granted in part and denied in part. Walsh's motion to strike the EEOC Investigative Memorandum and EEOC Determinations [#138] is denied.