Individuals differ on multiple aspects of their job-role behavior; criteria are measures that attempt to capture these differences. Measures of criteria are used by several constituencies within applied psychology. Among them, researchers used criteria for the evaluation of theories of work behavior, the effective administration of human resources and the provision of feedback to individuals. One index of the importance of criteria is the observation that most, if not all, of the pioneers of industrial-organizational psychology addressed this issue during their careers. This article reviews conceptual and methodological developments pertaining to the criterion problem since 1917, using as an organizing device dimensions, methods of measurement and analysis, and categorizing frameworks. A shift away from an emphasis on brute prediction toward a balanced treatment of both empirical and conceptual issues is highlighted by calls for the validation of criteria and by increased attention to modeling performance, as well as a recognition of multiple perspectives and competing values from which to view performance and criterion measurement.People in organizations behave according to their role perceptions and the roles assigned to them by others. The resulting behaviors and associated outputs, which are often limited by situations (Viteles, 1925(Viteles, -1926b and may vary widely across individuals (Tiffin, 1942;Viteles, 1932), form a large part of the subject matter of industrial-organizational (I-O) psychology. Individual variability on criterion measures interests researchers and managers, who attempt to measure and influence it. Because criteria are essential for evaluation of individuals, programs, and organizational interventions (Schmitt & Klimoski, 1991), the practical significance of criterion measurement has been recognized since the beginning of industrial psychology (Munsterberg, 1913;Scott, 1917). However, the systematic study of criteria per se to expand scientific knowledge This article is dedicated to Robert L. Thorndike, 1910Thorndike, -1990, who made monumental contributions to the study of criteria during his long career at Teachers College, Columbia University. Both authors contributed equally to the preparation of this article.We thank the following individuals, who critically and constructively commented on earlier versions of this article:

The authors studied the relationship between the Big Five personality factors and rating elevation among 111 students making peer evaluations. It was hypothesized that Conscientiousness (C) scores would be negatively correlated with rating level and that Agreeableness (A) scores would be positively correlated with rating level. We further predicted that individuals who were low on C and high on A would produce the most elevated ratings. As predicted, A scores were positively related to rating level (.33, p < .01) and C scores were negatively correlated with rating level (-.37, p < .01). Using the strong hypothesis test (P. Bobko, 1986), the authors found that ratings by low C and high A individuals were more elevated than all other groups of students combined (p < .01).

Investigations of the learned helplessness model of depression have been hampered by the modest reliability of measures of explanatory style: the habitual tendency to explain bad events with internal, stable, and global causes. We describe a new measure of explanatory style, the Expanded Attributional Style Questionnaire, and its use in a preliminary study with 140 college students. Individual dimensions of explanatory style were reliable, were correlated with depressive symptoms, and predicted actual causal explanations for bad events 4 weeks later. Research implications are discussed.Contemporary psychology is interested in cognitive accounts of depression: theories that link the origin and maintenance of depressive symptoms to the ways people think about themselves

No abstract

We investigated the predictability of rating level and two measures of rating accuracy from rater Agreeableness (A) and Conscientiousness (C) scores of the Five Factor Model. One hundred and twenty-six students made peer ratings after participating in several group exercises under conditions designed to emulate the modal peer rating system in which raters had low accountability for their ratings. Scores were correlated with average rating level (r ¼.18, po.05) and both measures of rating accuracy (po.05) and C scores were correlated with rating level (À.20, po.05) and both measures of rating accuracy (po.05). As suggested by Bernardin, Villanova, and Cooke (Journal of Applied Psychology, 85, 232-236, 2000), raters who were both more agreeable and less conscientious made the most lenient and least accurate ratings. Contrary to Yun, Donahue, Dudley, and McFarland (International Journal of Selection and Assessment, 13, 97-107, 2005), more agreeable raters also tended to rate the least effective performers more leniently than did other raters.