A widely accepted method of job analysis, the critical incident technique, was applied in order to determine dimensions of performance among psychology interns. All training directors of Association of Psychology Internship Centers were contacted regarding participation, and 46 (90%) completed the critical incident interview. The 270 critical incidents were analyzed with a standard procedure for determining behaviorally defined job dimensions. Seven dimensions were identified: clinical skills, commitment to own professional development, crisis and emergency situations, interpersonal relationships, knowledge, professional and ethical behavior, and programming and consultation skills. The results are presented within the context of other definitions of intern performance and of entry-level psychological practice. In addition, suggestions for improving selection and evaluation of psychology interns are presented. A recent issue of The Counseling Psychologist is devoted to a discussion of the predoctoral internship in psychology. In this issue, Holloway and Roehlke (1987) and the responding authors raised the issue of selection as a problematic aspect of the internship process. In fact, throughout the published literature discussing the internship as a component of predoctoral training in psychology, selection is identified as being a key to successful internship training. In relevant articles authors have offered suggested questions for sites and interns in order to enhance the fit between intern and site (e.g., Brill, Wolkin, & McKeel, 1985; Hersch & Poey, 1984). Other authors have described factors considered by desirable candidates in making an internship choice (e.g., Burstein, Schoenfeld, Loucks, Stedman, & Costello, 1981). In yet other articles authors have recounted the process from an anecdotal perspective (e.g., Burt, 1985), hoping to provide information to prospective interns on how internship selection can influence both decision making and satisfaction with the decision. Some research has been conducted to determine which criteria are weighed heavily in the selection decision. For example, Spitzform and Hamilton (1976) surveyed 91 training directors of the 116 existing APA-approved internships with questions regarding the selection procedures. Their results indicated general agreement among training directors that letters of recommendatio n and experience in clinical