EXPLAINING U.S. COURTS OF APPEALS DECISIONS
INVOLVING PERFORMANCE APPRAISAL:
ACCURACY, FAIRNESS, AND VALIDATION
JON M. WERNER, MARK C. BOLINO
Department of Management
University of South Carolina
Accuracy and due process perspectives were used to extend policycapturing research concerning employment discrimination case law. TWo-hundred ninety-five usable U.S. Circuit Court decisions concerning performance appraisal were located from 1980-1995. In both chisquare and multivariate LOGIT analyses, decisions were explained by: use of job analysis, provision of written instructions, employee review of results, and agreement among raters. Contrary to hypotheses, appraisal frequency and type (traits vs. behaviors or results) were unrelated to judicial decision. Rater training approached significance in chi-square analysis. Of other variables checked (e.g., type of discrimination claim, statutory basis, class action status, year of decision, circuit court, type of organization, purpose of appraisal, evaluator race and sex), only circuit court approached significance. We conclude that issues relevant to fairness and due process were most salient to Judicial decisions; issues pertaining to accuracy were important, yet validation was virtually ignored in this sample of cases.
Today there is no dispute that performance appraisal practices are
subject to employment legislation such as Title VII of the 1964 Civil Rights Act. Further, many researchers and practitioners view performance appraisal as an employment "test" covered by the Uniform Guidelines on Employee Selection Procedures (1978). The Guidelines were adopted in 1978 by the Equal Employment Opportunity Commission and other regulatory agencies, and emphasize the need for employers to validate all employment criteria, both "objective" and "subjective," where adverse impact has been found. This is the official position of We would like to thank John HoUenbeck, Sandra Gleason, Hoyt Wheeler, Brian Klaas, Martin Malin, John Grego, Samantha Wolf, Tbm Ruprecht, Bob Strack, Scott Keyes, and the anonymous reviewers for their valuable assistance at various stages of this research. An earlier effort at similar research was presented at the National Academy of Management Convention in San Francisco, August, 1990.
Correspondence and requests for reprints, or a list of Courts of Appeals cases used in this analysis, may be obtained from Jon M. Werner at the Department of Management, University of South Carolina, Columbia, SC 29208 or e-mail firstname.lastname@example.org.
COPYRIGHT © 1997 PERSONNEL PSYCHOLOGY. INC
Division 14 of the American Psyehological Assoeiation as well (Amicus Curiae Brief, 1988).
Initially, however, it was unelear whether the eourts would view performance appraisal as an employment test (Schneier, 1978). Acting on the assumption that appraisals should be viewed in this manner, Feild and Holley (1982) utilized quantitative analyses to determine which variables most influenced judicial decisions in employment discrimination cases. They studied legal cases from 1965 to 1980 where performance appraisal was an issue in an employment discrimination charge. Sixty-six cases were located, including two state court, 46 federal district court, 16 court of appeals, and two U.S. Supreme Court decisions. Five variables distinguished between decisions for plaintiffs and decisions for defendants: use of job analysis, behavior-oriented appraisal systems, provision of written instructions for evaluators, review of appraisal results with employees, and type of organization (industrial vs. non-industrial). When the legal ramifications of performance appraisal are discussed, recommendations similar to theirfindingsare often made (Barrett & Kernan, 1987; Martin & Bartol, 1991; Milkovich & Boudreau, 1994; Veglahn, 1993).
Recently, the strong measurement emphasis in performance appraisal has been likened to a...
Please join StudyMode to read the full document