The child sexual behavior inventory: Reliability and validity in a Dutch normative and clinical sample.

Background: The Child Sexual Behavior Inventory (CSBI) is one of the few instruments available to assess sexual behavior in children. Because of cross-cultural differences in the definition of normal and atypical sexual behaviors, the CSBI is not generalizable to other countries.

Objective: The current study examined psychometric properties of the Dutch version of the CSBI. Participants and setting: The study represents 3206 children from a normative sample, and 932 children from clinical samples.

Methods: Parents ratings on the CSBI, and questionnaires for psychosocial problems were obtained to assess psychometric properties.

Results: Results showed that Dutch parents interpret more CSBI items as developmentally appropriate than USA parents. Reliability coefficients indicated internal consistency (α = .39-.82), test-retest (r = .86, p < .000) and interrater reliability (Cohen's d =0.47, p >.05). Positive correlations between the CSBI total scale and the subscales demonstrated content validity (r = .32-.96, p < .01). Differences between normative and clinical samples on the CSBI total (ηp2 = .01-.65, p < .000), DRSB (ηp2 = .00-.03, p = .00-.38) and SASI scales (ηp2 =.00-.18, p < .01) indicated construct validity. Within the clinical samples, highest associations were found between the CSBI score and convergent measures (r = .64, p < .000). Lowest associations were found between the CSBI scores and divergent measures r = .14-.54, p = n.s.-<.000.

Conclusions: Cross-cultural differences in the perception of developmental appropriate sexual behavior underscore the importance of country-specific normative data. Good psychometric properties of the Dutch version of the CSBI were supported by this study.