We’ve written before on the questionable statistics used by the U.S. Equal Employment Opportunity Commission (EEOC) in other cases, and a recent court of appeals case involving background checks suggests that the EEOC is continuing to use such methods despite scathing criticism from courts.

On February 20, 2015, the U.S. Court of Appeals for the Fourth Circuit affirmed a Maryland federal district court’s entry of summary judgment against the EEOC with respect to its lawsuit alleging that an employer’s background check program violated Title VII of the Civil Rights Act of 1964. EEOC v. Freeman, Case No. 13-2365 (4th Cir. Feb. 20, 2015). We blogged the district court’s decision here. In both a unanimous majority opinion and a separate concurring opinion authored by U.S. Circuit Judge G. Steven Agee, the judges excoriated the EEOC and its expert for the statistical analysis submitted in support of the agency’s claim that the employer’s criminal and credit checks had a discriminatory disparate impact on black and male applicants.

As we noted in our prior blog, the district court’s 40-page opinion systematically demolished virtually every facet of the EEOC’s statistical case. Agreeing with the district court, the Fourth Circuit found that the reports provided by the EEOC’s expert, industrial/organizational psychologist Kevin R. Murphy, contained “an alarming number of errors and analytical fallacies.” Specifically, the court stated that Murphy failed to include in his database “hundreds, if not thousands,” of applicants who had been identified by the employer. The court also pointed to the “‘mind-boggling’ number of errors and unexplained discrepancies in Murphy’s database” that the district court found, such as omission of certain claimants from the database, the incorrect coding of individuals as passing or failing the criminal background check, the incorrect coding of claimants’ race or gender, double-counting of certain claimants, and failure to code certain claimants with respect to their credit check results. The Fourth Circuit agreed with the district court’s finding that Murphy’s attempt to fix errors in his analysis was inadequate and in fact created more errors. Based upon the foregoing, the Fourth Circuit concluded that Murphy’s errors rendered his analysis “outside the range where experts might reasonably differ,” and, thus, the district court did not abuse its discretion in excluding Murphy’s report as unreliable under Federal Rule of Evidence 702.

Judge Agee, while concurring in the court’s opinion, wrote separately “to address [his] concern with the EEOC’s disappointing litigation conduct.” While echoing the majority’s concerns about Murphy’s “fatally flawed” analysis, Judge Agee pointed out that other courts had found similar issues with Murphy’s work – including the Sixth Circuit in another background-check case, EEOC v. Kaplan Higher Education Corp., 749 F.3d 749 (6th Cir. 2014) – and he criticized the EEOC for continuing to defend Murphy’s testimony despite what the judge called Murphy’s “record of slipshod work, faulty analysis, and statistical sleight of hand.” Judge Agee concluded that by continuing to proffer a substandard expert analysis, the EEOC failed to discharge its duty to employers to cease enforcement attempts after discovering that an action lacked merit, and that it had failed to be vigilant to ensure that it did not abuse the power conferred on it by Congress.

The bottom line: Employers facing the EEOC in disparate impact litigation can expect to face junk statistics, but courts are increasingly skeptical and critical of such efforts.