The first study involved 20 incarcerated male child molesters, who volunteered to participate in the study. All participants were in the assessment phase of the Kia Marama Sex Offender Treatment Program at Rolleston Prison, and none had undergone previous sexual offending treatment. During the assessment phase, each child molester undergoes various tests and interviews. The Offense History Interviews were used as a data source for this study. The second study involved 25 incarcerated male child molesters in the assessment phase at the Kia Marama Sexual Offender Treatment Program. Ten of these offenders had participated in the first study, but this was not believed to bias or contaminate the second study, because a different method of data collection was used. For the second study, an independent, university-based researcher assisted the men in completing a written description of an offense structured around relapse-prevention construct. The model developed from the first study consists of four categories: offense chain, cognitive operations, cognitive content, and meta-variables. To determine its content validity and reliability, the model was applied to the offense descriptions of the sample of 25 incarcerated child molesters. Results suggest that the model has provisional validity and adequate interrater reliability. The authors discuss the theoretical and clinical implications of the model. 29 references