In OFCCP v. Bank of America (BOA), the Administrative Review Board (ARB) overturned OFCCP’s allegation that BOA discriminated against Blacks in 2002-2005. As described in another blog in this series, the basis for the ARB reversal of the 2002-2005 case was the lack of evidence through data aggregation. Specifically, the ARB stated that:
“The OFCCP’s evidence of discrimination in 2002-2005 boils down to one standard deviation of 4.0 (or 4.1) for the four-year period, but no standard deviation conclusions year by year.”
Similar to the ruling made in the Lopez v. City of Lawrence case, where aggregation of separate city applicant pools was rejected, the ARB discredited the aggregation of data across several years. Without evidence year by year, the aggregation of four years of data may be arbitrary and inappropriate. As noted by Jacobs, Murphy and Silva (2012)1, there are unintended consequences that may come from analyzing large databases, from which they coined the phrase “Being Big is Worse than Being Bad. This phrase refers to the increased chance of a statistical finding in a large dataset due to statistical power.
The BOA conclusion is noteworthy because it provides further guidance on conducting adverse impact analyses across multiple time periods, and supports the idea of pushing back against the arbitrary, multi-year data aggregation that may be used in analyses performed by the agency. Overall, employers need to be cautious when aggregating data and should consider it only under certain parameters. For additional background and corroboration on data aggregation, we recommend reviewing the Technical Advisory Committee Report on Best Practices in Adverse Impact Analyses.
Unrelated to the data aggregation issue, there have been other new developments in the BOA case. Stay tuned as the 23 year saga continues.
By Amanda Shapiro, Senior Consultant, and Vinaya Sakpal, HR Analyst, DCI Consulting Group
