A multi‐rater assessment of organizational commitment: are self‐report measures biased?
研究通过自我、同事和主管三种评估者来源，检验了组织承诺自我报告测量的偏差来源，发现自我报告主要受自身观察影响而非防御性反应偏差，且自我和同事测量在预测工作绩效上基本可互换。
Abstract Most investigations of organizational commitment have been conducted using self‐report measures, however, the veracity of self‐reports is often questioned. In a sample of 79 public‐sector administrative staff, we assessed two types of organizational commitment (affective and continuance) from the perspective of three different sources of raters (self, peer, and supervisor) to test three explanations of the factors influencing self‐report measures (observational opportunities, simple defensiveness, and moderated defensiveness). The pattern of correlations among the measures, analysed using the composite direct product multitrait–multirater approach, suggested that self‐report commitment measures are affected mainly by observations or experiences of the self‐reporter rather than by systematic bias related to defensive responding. This increases our confidence that scores from self‐report measures of affective and continuance commitment are veridical. Further, self‐ and peer‐based measures of commitment were largely redundant in the prediction of a job‐performance criterion whereas supervisory measures added unique predictive variance. Implications are discussed. Copyright © 2001 John Wiley & Sons, Ltd.