Did the Results of Promotion Exams Have a Disparate Impact on Minorities? Using Statistical Evidence in Ricci v. DeStefano

摘要

This paper shows how to use data from the Ricci v. DeStefano case in statistics courses. The Ricci v. DeStefano case was about disparate impact of firefighters' promotion exams in New Haven, Connecticut. A statistical analysis of the test scores of both Lieutenant and Captain exams indicates that there is significant difference between the average test scores of minority and majority applicants. Analysis of the passing rates and the rates of potential promotion to the Captain position, however, does not show significant difference. This apparent contradictory result shows students that in real situations, different ways of analyzing data can lead to completely different conclusions. During the trial, the court used the government's “four-fifths rule” or guideline to reach its decision. The paper also presents a guided senior thesis project to assess the statistical soundness of this “four-fifths rule”. The analysis reinforces a previous study that showed that the “four-fifths rule” guideline was not appropriate for the data in the Ricci case.