Measurement properties of the Child Behavior Checklist-10: an ultra-brief screening updated in longitudinal cohort

To develop a brief behavioral screening tool for youth, we sought to reduce the Child Behavior Checklist (CBCL) by selecting the most informative items, establish its structural model, and evaluate comprehensive psychometric properties for potential attention deficit and hyperactivity disorder (ADHD) screening across diverse populations.

Longitudinal data from the Future of Families and Child Wellbeing Study (FFCWS) were analyzed across three time points. Using a split-sample approach, we aimed to reduce the 34-item CBCL by integrating exploratory graph analysis (EGA) and graded response model (GRM). The final instrument would be evaluated for measurement properties.

Using a sample of 1,786 youth (49.5% female; 51.4% Black/African American), we systematically reduced the 34-item CBCL to a 10-item version (CBCL-10) with a three-factor structure. The CBCL-10 delineated excellent structural validity, supportive measurement invariances, and high internal consistency. The receiver operating characteristic analysis revealed good diagnostic efficiency for ADHD (area under the curve [AUC; 95% CI] = 0.843 [0.818-0.867]), with optimal sensitivity of 0.797 and specificity of 0.736 at a cutoff score of 3.

The present study successfully developed a psychometrically sound CBCL-10 with a diagnostic efficiency for measuring youth attention/behavioral problems. Without sacrificing measurement quality, CBCL-10 offers clinicians and researchers a feasible assessment tool that captures the most diagnostically relevant attention and behavioral manifestations of ADHD.