An analysis of psychometric data from a sample of 341 UK child abusers who had completed a probation-based sex offender treatment programme was carried out in order to assess the effectiveness of therapeutic treatment. A cluster analysis was also undertaken to examine the pattern of pre-treatment problems. Three clusters of offenders, with distinctly different psychometric profiles, were identified depending upon the number and extent of their offence-specific and social adequacy problems. These were labelled Low need, Medium need and High need. Pre-post analyses revealed clinically significant treatment effects for the entire sample, with differing effects found across the clusters as follows: 50-81% of the Low need group scored within the cut-off (or normative range) at the post stage, while between 3% and 26% had shifted to a clinically significant degree; 34-75% of the Medium need group scored within the cut-off range after treatment, while between 9% and 100% had shifted to a clinically significant degree; 16-52% of the High need sample scored within the cut-off at the post-treatment stage, and between 15% and 80% had shifted to a clinically significant degree.