Understanding the dynamics of motivation to change among sexual offenders has significant implications for treatment design and delivery, as well as in enhancing outcome. Traditionally, sex offenders have not been considered amenable to treatment unless they acknowledge the offense, recognize their sexual offending as a problem they want to change, and are willing to enter into and fully participate in treatment. Subjects were 101 federally sentenced male sexual offenders on conditional release over a 7 year period. They were classified as belonging to one of three groups: pedophiles, nonpedophilic child molesters (largely incest-type offenders), or sexual aggressives (largely rapists and other sexual offenders against female adults). Data were collected from psychology files or from data maintained in the Offender Management System. Pretreatment assessment reports, institutional posttreatment reports, initial community assessment reports, and community progress reports were examined. Treatment reports were rated using the Goal Attainment Scaling protocol specifically designed for sexual offenders. The five motivational measures were acceptance of guilt, acceptance of responsibility, disclosure of personal information, participation of treatment, and motivation to change behavior. The results show that the five motivational measures changed over the course of treatment, which supports the conceptualization of motivation as a fluctuating variable. Motivation increased significantly from institutional assessment to institutional posttreatment, suggesting that the institutional treatment protocol had an impact on participants’ motivational levels. Other factors such as a desire to achieve conditional release may have biased the results to a degree. The influence of conditional release may be surmised from the finding that motivation decreased for all types of sexual offenders upon community release. Motivation is dynamic and may be impacted by internal, external, and alliance variables. These results show that clinicians in community settings should expect to have difficulty re-engaging offenders in the treatment process and should not assume that a positive institutional report will be reflected in a client’s attitude and behavior in the community. 5 figures 3 tables, 34 references