Measurement Bias in Caregiver-Report of Early Childhood Behavior Problems across Demographic Factors in an ECHO-wide Diverse Sample

Abstract

BackgroundResearch and clinical practice rely heavily on caregiver-report measures, such as the Child Behavior Checklist 1.5-5 (CBCL/1.5-5), to gather information about early childhood behavior problems and to screen for clinically significant behavior problems. While studies have shown that demographic variables influence ratings of behavior problems, the extent to which the CBCL/1.5-5 functions equivalently at the item level across diverse samples is unknown.MethodsItem-level data of CBCL/1.5-5 on a large sample of young children (N=7827) were drawn from 26 cohorts in the Environmental influences on Child Health Outcomes (ECHO) program. Factor analyses and the alignment method were applied to examine measurement invariance and differential item functioning (DIF) across child- (child age, sex, bilingual status, and neurodevelopmental disorders), and caregiver-level characteristics (caregiver sex, education level, household income level, and depression). Child race was examined in sensitivity analyses.ResultsItem sets with the most impactful DIF across child and caregiver groupings were identified for Internalizing, Externalizing, and Total Problems. The robust item sets, which excluded the high DIF items, showed good reliabilities and high correlations with the original scales for all three constructs. Language version of CBCL administration, caregiver education, and respondent sex showed the most significant impact on the measurement invariance, followed by child age. Sensitivity analyses revealed that child race has a unique impact on DIF over and above socioeconomic status.ConclusionsThe CBCL/1.5-5, a parent-report measure of early childhood behavior problems, showed bias across demographic groups. Robust item sets with less DIF can measure Internalizing, Externalizing, and Total Problems equally as well as the full item sets.