Factor structure and internal consistency of the ISPCAN Child Abuse Screening Tool Parent Version (ICAST-P) in a cross-country pooled data set in nine Balkan countries

Abstract

Objective: Researchers are increasingly using parental report measures in population-based surveys of violence against children. No research thus far has examined the factor structure of the International Society for the Prevention of Child Abuse and Neglect (ISPCAN) Child Abuse Screening Tool Parent Version (ICAST-P), a non-commercialized measure for prevalence studies.

Methods: This paper uses secondary data from the Balkan Epidemiological Study on Child Abuse and Neglect (BECAN) which included 25,202 primary caregivers of school-going children in three grades (aged 11, 13 and 16). Primary caregivers completed the ICAST-P 2010-2011, which measures children's exposure to physical and psychological violence and neglect by caregivers and sexual violence by any perpetrator. Confirmatory factor analysis (CFA) was used to assess the reliability and the factor structure of the ICAST-P for a four-dimensional model and a bifactor model, with one general factor and four group factors representing the specific components of each set of items. Measurement invariance could not be tested.

Results: Lifetime prevalence for physical violence was 66.7 %, for psychological violence 66.6 %, for sexual violence 3.1 % and for neglect 9.1 %. CFA showed adequate fit for the hypothesized four-dimensional model, however, improved fit was shown for a bifactor model with abuse as the general factor. Internal consistency was good for physical and psychological violence, but not neglect and sexual violence. Hierarchical omega showed good internal consistency for the general factor.

Conclusions: Parents report that their children experience high levels of violence. The factor structure of the ICAST-P is best captured using a bifactor modelling approach.

Keywords: Bifactor; Child abuse; Child maltreatment; ICAST; Instrument; Measurement; Psychometrics.