Ethnic differences using behavior rating scales to assess the mental health of children : a measurement equivalence study

Cultural competence in measurement has become increasingly important in child welfare and juvenile justice systems where disproportionate numbers of children and families from diverse ethnic groups are served. However, there is a present gap in the literature regarding the assessment of children from different ethnic groups using behavior-rating scales. This study examined the cross-ethnic measurement equivalence of the SAC (Glisson et al, 2001) using a sample that included 562 African American (AA) and 692 Caucasian-American (CA) children. These data were part of a three panel longitudinal survey of families referred to juvenile court in one state. Specifically, baseline and first time follow-up data (at least six months) on these children were used to conduct the analyses presented here. Caregivers completed several assessment instruments and a family interview questionnaire. Two types of reliability, as well as multiple sources of post-dictive and predictive criterion-related validity were analyzed. Furthermore, the factorial equivalence of the SAC was analyzed using a structural equation modeling procedure. Alpha reliabilities for the full scale and for each sub-scale (i.e., externalizing and internalizing) were high for both groups. Test-retest reliabilities were also good for both groups. There were substantially more similarities than there were differences between the two groups in terms of the criterion validity evidence. The factorial equivalence evidence was mixed, and suggested that while the factor structure of the SAC was non-equivalent across the two groups, it worked similar and equally well for both groups. Practical implications of these results and directions for future research are forwarded.