Toward a Coherent Test for Disparate Impact Discrimination

Statistics are generally plaintiffs' primary evidence in establishing a prima facie case of disparate impact discrimination. Thus, the use, or misuse, of statistics dictates case outcomes. Lacking a coherent test for disparate impact, courts choose between the two prevailing tests, statistical significance and the four-fifths rule, in deciding cases, and these tests frequently produce opposite results. Litigants thus face considerable uncertainty and the risk that a judge's preferred outcome will dictate which test is applied.

This Article recognizes that the two tests perform complementary functions that both play a useful role in determining whether liability should be imposed. statistical significance establishest hat the challenged practice likely caused the disparity, and the four-fifths rule establishes that the disparity is large enough to matter. Rather than choose between the two tests, courts should use a uniform and coherent standard that combines both them. Determining the parameters of this standard involves difficult policy decisions about the purposes of the doctrine as well as who, and to what extent, should bear the risk of error.