Strategies for enhancing the prediction of job performance from job satisfaction

Two moderator variables, performance to reward contingency and self-esteem, were incorporated into multiple regression equations in an attempt to increase the predictability of job performance ratings from job satisfaction information. 251 employees, occupying 3 organizational levels of a nationwide corporation, responded by mail to a questionnaire package. Ss received scores on 4 self-report inventories covering the areas of job satisfaction, self-esteem, and contingency of rewards on job performance (e.g., the Job Description Index, the Faces scale of job satisfaction, and the Texas Social Behavior Inventory). Additionally, each S was rated by his or her immediate supervisor on the quality of overall job performance. Results indicate that the moderator variable approach, operationalized via moderated regression, substantially increased the relationship between satisfaction and performance. (15 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).