Measuring Behavior Problems in Children from Low-Income Families: A Rasch Analysis of the Child Behavior Checklist for Ages 1½–5

Abstract

The Child Behavior Checklist for Ages 1.5–5 (CBCL/1½–5) has been widely used by researchers and clinicians in the field of special education and psychology. The purpose of this study was to examine the psychometric properties of the CBCL/1½-5 with a sample of preschool children from low-income families using the Rasch model. Participants included 244 children enrolled in a Head Start program. Findings suggested that both the Internalizing and Externalizing subscales are unidimensional and demonstrated local independence successfully, after misfit items were removed to fit the Rasch model. Both subscales operated well with high item reliability and low to medium person reliability, indicating that both subscales have stable item difficulty orders from sample to sample but weaker capacity to distinguish children with mild problem behaviors from those with more severe problems. Differential item functioning was found for a few items across child gender. Considering the length of the item sets, it is appropriate to use the subscale scores to compare the differences in problem behaviors between boys and girls. Overall, the CBCL/1½–5 has adequate psychometric properties for detecting problem behaviors in preschool children from low-income families. Implications were discussed.