Attribution Processes in the Use of Performance Feedback in an Organization.

Abstract

This study examines the impact of performance feedback to individuals on subsequent leadership ratings. Using Bem's self-perception model, it was hypothesized that leadership ratings would be responsive to performance cues. This hypothesis was tested in a field setting, using ten bank branches participating in a 13-month time series quasi-experimental field test of a monthly performance and attitude feedback system. Within-subjects repeated measures experimental design models were estimated using performance feedback to predict subsequent leader ratings. The results were mildly supportive of the hypothesized self-perception process, with performance feedback predicting future leader ratings. However, the pattern of responses to performance feedback was more complex than originally anticipated. Groups that actively used the feedback data for problem solving were less responsive to the feedback data than groups that received feedback data without using it for problem solving. Performance feedback cues were more actively used in self-perception (1) when they reflected the activities of the leader and (2) when other plausible explanations for performance cues and leader behavior were unavailable to raters. The self-perception process was moderated by the availability of internal salient cues. The self-perception model did not predict satisfaction and intrinsic motivation ratings, where internal salient cues are available to the rater, but did predict leadership and group process ratings, where internal salient cues are unavailable. Correlational analyses showed occupation and work groups responding differently (in direction and magnitude) to performance feedback. Implicit leadership-performance theories (measured by correlating ratings of leadership and performance) changed across time, indicating that they may be affected by the intervening performance feedback. The results indicate individual processing of performance feedback is dynamic, with changes in beliefs and implicit theories resulting from processing feedback. Theoretical suggestions and revisions are offered to deal with the complexity of responses to performance feedback. The areas focused on are defining knowledge representations and their complementary inference processes, greater sensitivity to changing information needs, and the generation of new information by persuasive argumentation. Suggestions are offered for designing management information systems, and diagnosing readiness for organizational change.