Approaches for Dealing with Small Sample Sizes in Employment Discrimination Litigation

The focus of all statistical analyses of alleged employment discrimination is to compare the treatment of similarly situated individuals. Economists and statisticians as experts offering quantitative support for or against allegations of employment discrimination are now commonplace in the courtroom. Although there is a rich and growing literature dealing with the tools and techniques used by these experts in large, class action litigation, relatively little of this literature addresses the problems associated with the approaches that may be used when the sample size being analyzed is small. The purpose of this paper is to outline the problems of developing statistical analysis involving "small sample size" problems and to present techniques for dealing with this problem. The paper provides several hypothetical examples of employment discrimination cases involving small sample sizes, illustrates alternative statistical techniques that can be used, and discusses the important consideration of the differences between statistical significance and practical significance