Breaking and associated brands will be offline for scheduled maintenance Friday Feb. 26 9 PM US EST to Saturday Feb. 27 6 AM EST. We apologize for the inconvenience.


Thank you for sharing!

Your article was successfully shared with the contacts you provided.
These days, counsel for the employer in discrimination cases often find that they must address statistical evidence at the class certification stage. In seeking to bring a class action, plaintiffs frequently rely on expert analyses purporting to demonstrate statistically significant differences in employment outcomes between workers based on race, gender, or some other protected characteristic. Plaintiffs argue that the disputed disparities present common questions of law and fact, making class certification appropriate under Federal Rule of Civil Procedure 23(a). The employer’s challenge to the probative value of this statistical evidence is too often rejected as an inappropriate and premature “battle of the experts.” Based in part upon the plaintiffs’ figures, the court may then certify the purported class of employees. Statistical disparities can be misleading, however, and they should not so easily withstand Rule 23(a) analysis. In cases alleging a pattern of subjective decision-making, for example, plaintiffs’ statistics frequently provide little or no probative evidence that there is a typical or common way in which the challenged policies or practices were applied. In cases where plaintiffs seek a multi-facility class, their statistics often fail to establish the commonality and typicality necessary to include workers from far-flung locations. So companies facing Title VII pattern or practice discrimination claims should be prepared to challenge the other side’s statistics at the class certification stage. There are ways to reveal the deficiencies that may undercut the plaintiffs’ statistical case — and thus to present a stronger challenge to it. THE PLAINTIFFS’ STATISTICS Title VII of the Civil Rights Act of 1964 prohibits employers from engaging in discrimination on the basis of race, national origin, gender, or religion. The act prohibits both intentional discrimination (disparate treatment) and facially neutral practices that disproportionately harm women or minorities (disparate impact). Like other class actions, employment discrimination class actions must satisfy the requirements of Rule 23(a), including commonality and typicality. See, e.g., General Telephone Company of the Southwest v. Falcon, 457 U.S. 147 (1982). Commonality exists where plaintiffs assert questions of law or fact common to the entire class. Typicality exists where the claims of the lead plaintiffs are typical of those of the entire class. Statistical experts are usually presented at the pre-certification phase to address these two elements and to demonstrate that the challenged policies or practices occur on a classwide basis. The statistical analyses submitted by plaintiffs at this stage are often the same type of reports that experts for both sides would later submit at the liability stage. Using the example of a gender-based promotions case, these analyses include, but are not limited to, the following: • “ Glass ceiling” analysis: Calculations that measure the percentage of women at each pay grade (or other organizational level), purporting to show a substantial drop-off in the female representation above a certain pay grade. • Promotion pools: A comparison of the actual and expected number of female promotions (based upon the presence of women in the pool of eligible workers), purporting to show statistically significant disparities in promotion rates. • Time to promotion: An analysis of the average years of service before employees receive a promotion, again purporting to show statistically significant differences between men and women. • Logistic regression: An analysis purporting to show a statistically significant difference in the probability that a man or a woman will be promoted even after controlling for factors believed to influence the likelihood of winning promotion. Certainly these kinds of analyses can identify differences, and clearly they can be provocative. But they often stop short of meeting the Rule 23(a) requirements and, therefore, fail to support allegations of classwide discrimination. In class-based litigation, the crux of the plaintiffs’ allegations is the purported existence of widespread discriminatory practices that are commonly applied. To demonstrate commonality and typicality, plaintiffs must establish statistically significant disparities consistently adverse to the protected group across job categories, pay groups, departments, etc. See, e.g., Reid v. Lockheed Martin Aeronautics Co., 205 F.R.D. 655 (N.D. Ga. 2001) (absence of consistent pattern of adverse treatment precludes a finding of commonality and typicality). In cases alleging a multi-facility class, the consistent occurrence of disparate adverse outcomes at all facilities seems critical to a proper finding of commonality and typicality. This is particularly true where the named plaintiffs seek to represent workers at facilities where they themselves never worked — and thus where they lack personal knowledge that the same practices occurred in the same manner. See, e.g., Stastny v. Southern Bell Telephone & Telegraph Co., 628 F.2d 267 (4th Cir. 1980); Webb v. Westinghouse Electric Corp., 78 F.R.D. 645 (E.D. Pa. 1978). Yet plaintiffs frequently advance analyses that rely on aggregated findings of statistical disparities across a company. It is important for the employer to probe into the components of those aggregated results, which may reveal statistically insignificant outcomes at some facilities or, indeed, outcomes that actually favor the protected group. Outcomes at the facilities where the named plaintiffs work may vary widely from outcomes at other facilities that plaintiffs seek to include in the litigation. Such variations will not be apparent in the aggregate analyses. Unless further investigated by the employer’s counsel, the plaintiffs’ report of aggregated disparities may present the court with a false reality. RETHINKING THE NUMBERS What specifically can employers’ counsel do when faced with this situation? First, the defense expert can test for statistical differences between the various facilities included in the plaintiffs’ analyses. If there is statistically significant variation from, say, department to department (especially if there is also evidence in the record indicating that the challenged employment decisions were made by the department heads independently from one another), then there is statistical evidence that grouping employees into one class across departments is erroneous. Second, statistical tests can be used to determine whether the factors that affect individuals’ promotion probabilities are significantly different from one department to another. For example, one manager may place a high value on performance ratings, while another disregards such ratings but relies heavily on seniority. While these differences may be hidden in an analysis of overall outcomes, multiple regression analysis, for example, may demonstrate that, in fact, the decision-making processes are not the same from department to department. Outside the context of a multiple regression analysis, other statistics can be used to show that the employer’s decision-making processes and the outcomes of those processes vary significantly from unit to unit and job level to job level. For example: • The representation of the protected class among employees eligible for promotion may be substantially different from department to department, so that grouping departments together into one analysis creates an applicant pool that does not represent accurately any individual department. • The employer’s databases of employee work history may demonstrate that some managers typically hire from the outside, whereas others typically promote from within. • The minimum qualifications necessary for promotion may differ from department to department. While the employer’s policy documents should spell out general guidelines for promotion, the work history databases may show departmental variation in practices even within those guidelines. • Departments may face vastly different external labor markets, resulting in differences in the protected workers’ availability and thus in the employer’s ability to successfully fill positions from outside the company. REVIEWING THE PLAINTIFFS The employer may also defend against class certification through tactical analysis of employee databases to demonstrate that the named plaintiffs don’t share common traits of the class they purport to represent. Such an analysis might: • Identify the named plaintiffs and gather data on their work histories. • Identify the jobs, departments, divisions, etc., in which they were employed during the relevant time period. • Identify the total number of jobs, departments, divisions, etc., across the entire class during the relevant period. • Show that the named plaintiffs occupied only a small percentage of those jobs, departments, divisions, etc., available to the class that they purport to represent. • Demonstrate that the named plaintiffs have no basis in either personal knowledge or statistical fact to impute the policies and practices applied to their positions to other jobs that they never held. Of course, the choice of statistical methodologies used is contingent upon the specific claims made by the plaintiffs, the quality of the employer’s databases, and the decision-making processes at issue. It is critical to defending against class claims that the employer maintain accurate and consistent databases on its employees. To the extent that the records are inaccurate or incomplete, defense counsel will be limited in their ability to provide the court with a meaningful response to the plaintiffs’ expert. Defense counsel can and should challenge the probative value of plaintiffs’ statistical data with regard to Rule 23(a) requirements. The Supreme Court’s decision in Falcon permits such challenges by imposing an affirmative obligation on the courts to assess whether plaintiffs satisfy the prerequisites for class certification. Anything less than a rigorous analysis of plaintiffs’ proof, including their statistical evidence, sanctions the blackmail value of class certification in forcing defendants to settle. Leslie M. Turner is a partner in the labor and employment practice group at Akin Gump Strauss Hauer & Feld in Washington, D.C. She can be reached at [email protected]. Paul F. White is vice president and director of the D.C. office of the ERS Group, a labor economics research and consulting firm. He can be reached at [email protected].

This content has been archived. It is available through our partners, LexisNexis® and Bloomberg Law.

To view this content, please continue to their sites.

Not a Lexis Advance® Subscriber?
Subscribe Now

Not a Bloomberg Law Subscriber?
Subscribe Now

Why am I seeing this?

LexisNexis® and Bloomberg Law are third party online distributors of the broad collection of current and archived versions of ALM's legal news publications. LexisNexis® and Bloomberg Law customers are able to access and use ALM's content, including content from the National Law Journal, The American Lawyer, Legaltech News, The New York Law Journal, and Corporate Counsel, as well as other sources of legal information.

For questions call 1-877-256-2472 or contact us at [email protected]


ALM Legal Publication Newsletters

Sign Up Today and Never Miss Another Story.

As part of your digital membership, you can sign up for an unlimited number of a wide range of complimentary newsletters. Visit your My Account page to make your selections. Get the timely legal news and critical analysis you cannot afford to miss. Tailored just for you. In your inbox. Every day.

Copyright © 2021 ALM Media Properties, LLC. All Rights Reserved.