Goodman et al. (2003) interviewed adolescents and young adults who had been victims of documented CSA that led to criminal prosecution of the alleged perpetrator. The sample was a subset of those who had previously (approximately 13 years earlier) participated in a study of court testimony (1992). As part of the follow-up study, participants were asked to report any experiences of CSA. Failure to report the target incident could indicate lack of memory for the abuse. The research concluded that forgetting is relatively infrequent among CSA survivors whose alleged perpetrators were prosecuted after a criminal investigation. A second major finding was that age, severity of abuse, and maternal support were related to disclosure, but gender, ethnicity, relationship with the perpetrator, and extent of legal involvement were not. The researchers used this latter finding to discount a variety of theories related to memory for trauma and abuse. There are many possible reasons why the nondisclosure rate might have been lower in this study. It can be expected that there would be better memory for abuse among survivors who had the opportunity to talk repeatedly about their victimization experiences and who were believed by caregivers and authority figures. It is also important to assess the analyses related to the second finding, especially given the researchers' broad arguments for rejecting several theories related to memory for trauma. The current critique argues that because of low statistical power, sparseness of data, and possible multicollinearity between predictors, Goodman et al.'s analyses should be interpreted with caution, since their substantive conclusions may be premature. The political, legal, and clinical implications of conclusions about memory mechanisms related to CSA are too significant to be based in flawed and preliminary research. Additional research that improves on the weaknesses of research to date is needed to inform debate and policymaking. 14 references