From information obtained on a psychosexual history interview, 68 child molesters were categorized as either repressed, nonrepressed, or exploitive. Findings show that a significantly larger proportion of therapy failures were offenders who were sexually repressed. There were also significant differences between repressed and nonrepressed offenders on some of the special research scales of the Minnesota Multiphasic Personality Inventory, the Burt Rape Myth Scales, some of the Multiphasic Sex Inventory scales, and the Mosher's Sex Guilt Scale. Differences on these measures are consistent with the characteristics of intrafamilial child molesters reported in the clinical and research literature. The results of this investigation are, however, speculative due to the small sample size of the repressed group. It is also conceivable that differences in treatment effectiveness between repressed and nonrepressed groups may be due to differences in social skills rather than to sexual repression. 4 tables, 3 notes, and 25 references