Statistical evidence may be relevant when a plaintiff tries to rebut a legitimate, non-discriminatory reason given by an employer for an adverse employment action, but often such evidence is not persuasive.  The numbers are often mixed, statistically insignificant, and not tied to the employment action at issue.  However, a recent case in which statistics played a major role was Dominguez v. Department of Education.[1]

In Dominguez, an Assistant Principal of a New York public high school, who was of Hispanic descent, claimed that his most recent favorable positive performance evaluation ("satisfactory") was down-graded (to "unsatisfactory"), that he was not retained under his five-year probationary service agreement, and that he was fired after four years of exemplary service, all by an African-American principal who came to the school shortly before his discharge.  The new principal asserted that the actions were due to the plaintiff's unsatisfactory work record and violation of the conflicts of interest policy, accusations the Assistant Principal denied.

The Court ruled against the employer, allowing the plaintiff to proceed with his case, in part because of evidence offered by the Assistant Principal disproving the principal's stated reasons for action.  But the Court also relied heavily on statistical evidence.  Before the new principal's arrival, four of the 11 Assistant Principals at the high school were not African-American (three Latinos, one West Indian, and one Caucasian).  Under the new regime, all four non-African-Americans were let go and replaced by African-Americans.  Likewise, all the African-American Assistant Principals were retained or replaced by other African-Americans.  This changed the percentage of African-American Assistant Principals from 64% to 100%.