The research literature on the psychometric properties of self-appraisals of work performance summarizes many results in terms of leniency, variability, halo, and construct validity. Comparisons with appraisals tend to show more leniency, less variability, more bias, and less discriminant validity. Different factor structures are found among self- and supervisor-ratings, while self-appraisals show generally less halo and significant correlations with other sources in some studies. Existing data do not allow any firm conclusions as to whether the quality of self-appraisals is a function of scale format, amount of rater training, type of judgment, or purpose of appraisal. Problems may exist when self-appraisals are used for administrative decision making, diagnosis of training needs, applied criterion measurement, measurement of constructs in basic research, or for selection purposes. (Author)