Many studies of U.S. Supreme Court decisionmaking focus on the relationship between the judge and the decision. One variation of this method uses judges' past votes to infer their attitudes. The attitudes, in turn, are used to predict the judges' future voting behavior. Another variation assumes that judges' precourt experiences will shape their attitudes and values, thus linking background characteristics with decisionmaking. This line of decisionmaking analysis does not consider the facts or legal principles involved in particular cases to be relevant. Other studies focus on the link between the case and the decision. The author proposes combining judge-decision and the case-decision methods and hypothesizes that the combined methods would be more successful in predicting court decision than would either method used alone. She studied equal protection cases by the U.S. Supreme Court during the 1981, 1982, and 1983 terms to test the utility of combining case characteristics and judges' attributes in a single analysis. The author discusses background and case variables and reports that stepwise discriminant function analysis was used to assess the impact of case characteristics and background characteristics on justices' votes in equal protection cases. She found that two background characteristics, political party identification and judicial experience, emerge as predictors of justices' votes in equal protection cases. Of the nine case characteristics used, the strongest predictor, the author found, is the presence of the Federal Government as defendant. The author concludes that combining background variables and case variables produces an improvement in predictive ability. 40 footnotes.