Content area
Full text
The effects of rating scale formats on several indices of the usefulness of performance appraisal for employee development were examined. The job performance of 96 police officers was rated using simple graphic scales or one of two behaviorally oriented rating formats: behaviorally anchored rating scales (BARS) and behavior observation scales (BOS). As predicted, ratees' satisfaction with performance appraisal was highest and their perceptions of performance goals most favorable when using BOS. In addition, performance improvement goals for officers rated using BOS were judged by experts to be most observable and specific. Contrary to the authors' predictions, graphic rating scales were generally as good as BOS and as good as or better than BARS when evaluated in terms of ratee attitudes and goal characteristics. The results suggest that different behaviorally-oriented rating formats can enhance or inhibit the developmental applications of performance appraisal.
Researchers and experts in performance appraisal have suggested two broad uses of appraisal in organizations (McGregor, 1957; Wexley, 1979). First, it serves administrative purposes in areas such as reward allocation (salary increases, bonuses) and assignment decisions (promotions; transfers, demotions). Second, it contributes to employee development in that it makes possible the identification of their strengths and weaknesses, provides performance feedback, and facilitates exchanges with supervisors (Bernardin & Beatty, 1984; Cummings, Schwab & Rosen, 1971; Dorfman, Stephan, & Loveland, 1986; Murphy & Cleveland, 1995).
A wide range of rating scale formats have been used in performance appraisal, and there is substantial literature dealing with the psychometric characteristics of rating scales (Bernardin & Beatty, 1984; Fay & Latham, 1982; Feldman, 1981; Kingstrom & Bass, 1981; Latham & Wexley, 1981; Latham, Fay, & Saari,1979; Murphy & Constans,1987; Murphy & Pardaffy, 1989; Steiner, Rain, & Smalley, 1993; Tziner, 1990). On the whole, differences in rating scale formats seem to have little impact on the reliability or validity of performance ratings (DeCotiis, 1977; Gosselin & Murphy, 1994; Landy & Fan, 1980; Murphy & Cleveland, 1995; Schwab, Heneman & DeCotiis, 1975). However, it is possible that differences in rating scale formats could influence other important dependent variables. In particular, different methods of obtaining and recording information about job performance may be more or less conducive to the developmental uses of performance appraisal. The purpose of this article is to...