Participants were involved in a 94-bed juvenile male residential sex-offender specific treatment program in a midwestern community. Treatment was based on the Jesness Model of Interpersonal Maturity. The sample consisted of 78 juvenile child molesters (ages 12 to 17 years) who completed intake psychological testing from 1993 to 1995. For the purposes of this study, child molesters were defined as offenders who were at least 3 years older than their victims and who had used either bribery, coercion, manipulation, or a combination of these factors, rather than physical force, to sexually molest their victims. The independent variables measured in relation to progress in treatment were pretreatment demographic variables, including age and extent of sexuality knowledge, and psychometric test variables that pertained to psychological maladjustment, impulsivity, defensiveness, criminal prosecution, and preoccupation with sexual thoughts. "Noncompleters" were defined as participants who were prematurely terminated from treatment because of chronic failure to complete sexuality-specific treatment goals, decampment, or the threat of violent behavior. Program "completers" were defined as those who successfully completed 10 sexuality-specific goals to the satisfaction of a multidisciplinary team. Forty-one participants completed the program, and 37 were noncompleters. A second sample collected between 1995 and 1996 was used for the purpose of cross- validation. Age and impulsivity produced the best-fit model, correctly classifying 76.9 percent of the sample (younger juveniles were more likely to complete the program). Cross- validation of this two-factor model on a different sample of 35 juvenile offenders with similar demographic characteristics, including age, ethnicity, IQ, criminal prosecution, and grade, correctly classified 71.43 percent of the sample. 4 tables and 23 references