Inconsistent retrospective self-reports of childhood sexual abuse and their correlates in the general population

W. Langeland, J.H. Smit, H. Merckelbach, G. de Vries, A.W. Hoogendoorn, N. Draijer*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


Epidemiological research on childhood sexual abuse (CSA) and its consequences in adult life mainly relies on retrospective reports. This study explores their consistency and the correlates of inconsistent CSA self-reports in a random population sample. A stratified subsample of 2,462 subjects (selected from a large-scale (N = 34,267) representative sample of Dutch adults aged 40 and beyond) participated in a two-phase online questionnaire survey on extra-familial CSA which was conducted on a four- to six-week interval. Subjects reporting CSA were overrepresented. Participants with consistent and inconsistent responses were compared with regard to demographics, family background, abuse severity, and clinical characteristics. Potential correlates of inconsistency were identified using logistic regression analysis. An additional questionnaire (Phase III) administered to inconsistent respondents explored possible reasons for their inconsistency. Of the 1,992 respondents who had reported extra-familial CSA during Phase I, 707 (35.5 %) denied this in Phase II. Of the 2,462 respondents in Phase II, 727 (29.5 %; 9.2 % when considering sample stratification) gave a discrepant answer to the extra-familial sexual abuse item compared to their answers given in Phase I. Reports of less severe abuse, intra-familial CSA, and early parental separation predicted inconsistency. Reasons provided for inconsistency varied from misunderstanding (e.g., reporting intra-familial CSA rather than extra-familial CSA) to emotional motives (e.g., embarrassment, being overwhelmed) or practical considerations (e.g., lack of privacy while filling out the questionnaire). Inconsistent self-reports of extra-familial sexual abuse occur on a substantial scale and are associated with less severe forms of abuse (lack of salience) or classification difficulties (perpetrator being a family member or not). Consistency tests and probing for clarifications or corrections should be routinely conducted in order to increase the quality of CSA epidemiological research.
Original languageEnglish
Pages (from-to)603-612
Number of pages10
JournalSocial Psychiatry and Psychiatric Epidemiology
Issue number4
Publication statusPublished - Apr 2015


  • Adults
  • Childhood sexual abuse
  • Consistency
  • Reporting practices
  • Survey

Cite this