Social withdrawal in children is a risk factor for maladjustment. The Revised Class Play (RCP; Masten, Morison, & Pelligrini, 1985) has often been used to identify children’s behavioral difficulties with peers. However, in previous studies the sensitive-isolated scale of the RCP appeared to measure a mixture of different types of withdrawal, including withdrawal from peers and exclusion by peers. In the present study the original RCP was modified to more clearly distinguish withdrawal from the peer group from behavior associated with exclusion by peers. Two studies in Dutch samples of 8–13-year-old children were conducted to examine the reliability and validity of this modified Revised Class Play. Both studies revealed three scales: Sociability-Leadership, Aggressive-Disruptive, and a third factor labeled Anxious-Withdrawn. Anxious-Withdrawn primarily reflected social reticence. Stability of the scales after two years was high. High scores on the Anxious-Withdrawn scale were best predicted by teacher ratings of timid and anxious behavior and by negative self-perceptions of social competence. The results support the reliability and validity of the modified RCP in a Dutch sample.