The Revised Class Play (RCP) is presented along with evidence of its reliability and validity. This measure of peer reputation was designed to improve the assessment of social competence as well as the psychometric properties of the class play method. The RCP was administered to third through sixth graders in two urban schools (N = 612). Three scores were derived based on factor analysis that revealed three dimensions: "Sociability-Leadership," "Aggressive-Disruptive," and "Sensitive-Isolated." Data are presented relating the three scores, which proved to be reliable and stable at intervals of 6 and 17 months, to SES, IQ, achievement, and teacher ratings. Positive reputation was associated with other aspects of social and intellectual competence, whereas isolated reputation was linked to difficulties at school. Multiple regression analyses suggested that a pattern of low-positive, high-aggressive-disruptive, and high-isolated peer reputation is the least competent. The advantages of multidimensional scoring and analysis of peer reputation are demonstrated by the results. Further directions for research are discussed.