Back to Annual Meeting Page		American Public Health Association 133rd Annual Meeting & Exposition December 10-14, 2005 Philadelphia, PA

Abstract #107293

Adam C. Carle, MA, PhD, Statistical Research Division, U.S. Census Bureau, Center for Survey Methods Research, Washington, DC 20233-9100, 301-763-1863, adam.c.carle@census.gov

Differential item functioning (DIF) is the extent to which items that make up a scale are systematically related to exogenous variables, e.g., gender, after equating individuals on the latent trait. Differential test functioning (DTF) refers to the extent that DIF affects the summary score. Before comparing groups' observed scores, it is necessary to establish a measure's equivalence across the groups. Despite findings that suggest the prevalence and course of depressive symptomatology differ across the developmental lifespan of boys and girls, no published studies have employed IRT to examine measurement invariance in a measure specifically designed to assess childhood depression. To address this issue, data from a larger, longitudinal study employing a multi-method multi-trait design to examine depression and anxiety among children and adolescents (see Cole, et al. 1998), were used to assess the validity of the Children's Depression Inventory (CDI: Kovacs, 1992) across 365 boys and 415 girls in the third and sixth grades. Analyses employed a partial credit Rasch model to examine DIF and DTF across boys and girls on the CDI. The general mathematical CFA model is developed and a method for exploring measurement bias is described. Results supported the use of the CDI in cross-gender comparisons generally, and provided evidence that similarities in depression for this age group likely reflect true similarities. Implications regarding the extent and impact of DIF and DTF with respect to the CDI, public health, prevention, and intervention are discussed.

Learning Objectives: