A random sample of 3,000 women was selected from the three Dunedin (New Zealand) electoral rolls and an adjacent rural electorate. A questionnaire was mailed to each woman, soliciting information on any CSA. The interview sample was selected from the women under 65 years old in the urban electorates. All the women in this age group who reported any CSA on the mail questionnaire were invited for an interview. A majority of CSA incidents were reported at both mail and interview stages of the survey. Incidents that involved genital contact were most likely to be reported at both stages. A significant number of women reported abuse by a close family member in the mail questionnaire but not at the interview. In contrast, incidents mentioned only at the interview were more likely to be committed by a stranger. At interview, many women who replied negatively to a general screening question on CSA went on to report abuse in response to detailed descriptive questions. This was especially true for noncontact experiences. Increasingly restrictive definitions of CSA lowered the prevalence rate for CSA under the age of 16 years from 34.4 percent to 19.7 percent. The authors conclude that neither the written nor verbal responses on incidents of CSA offered clear advantages in the disclosure of abusive experiences. A self- administered format had the advantage of anonymity, and an interview provided an opportunity for clarification of detail. The likelihood of gaining comprehensive data from either source is dependent on detailed, clearly defined, and unambiguous questions.