IDEOLOGICAL, LEGAL, AND CASELOAD EFFECTS ON COURTS OF APPEALS JUDGES’ VOTING IN RACE-BASED TITLE VII CASES BROUGHT BY AFRICAN-AMERICAN PLAINTIFFS IN K-12 PUBLIC EDUCATION

Abstract

This study investigated how effectively U.S. Courts of Appeals judges’(1) political ideology, as measured by party of the appointing president or judges’ DW-nominate scores; (2) legal developments, as measured by the 1991 Amendments to Title VII of the Civil Rights Act of 1964; (3) judicial caseload carried by courts of appeals three-judge panels; and (4) plaintiffs’ gender predict liberal-pro-employee and conservative-pro-employer voting in K-12 Title VII racial discrimination cases brought by African-American plaintiffs decided between 1964 and 2015. The study used logistic regression as its primary statistical tool. The principal findings for this group of K-12 decisions are: (1) political ideology as measured by party-of-the-appointing president and DW-Nominate scores does not effectively predict court of appeals judges’ voting when all other variables are held constant; (2) changes brought about by the 1991 Title VII amendments brought about a substantial reduction in pro-plaintiff voting compared to the pre-1991 period when all other variables are held constant, irrespective of whether the P-A-P or the DW-Nominate measures were used in the model; (3) judicial caseload was not a significant predictor of judicial voting when all other variables are held constant; and (4) plaintiffs gender was not a significant predictor of judicial voting within these regression models. It was suggested that the strong reduction in liberal-pro-plaintiff voting after the 1991 amendments were brought about by incentives provided by the combination of compensatory and punitive damages and attorney’s fee awards becoming available after the amendments: these appeared to be brought about a marked increase in meritless claims due to those incentives during the post-1991 period. Possible effects of the amendments on the ideological measures were considered as well.