This study explores the construct validity of the Child Behavior Checklist (CBCL), which was created as a standardized tool for recording the behaviors of children aged 4 through 16 as reported by their parents or caretakers. Since its inception in 1983, the CBCL has been applied and tested on numerous clinical samples, yielding an impressive body of quantitative support. Due to the liberal application of the CBCL in child mental health settings, however, it is critical that the validity of the instrument be supported for specific child sub-populations. In this study, the factor structure of the CBCL is explored for different sex and age variations in a sample of 2,628 children presented for treatment at a mental health center in a midwestern city. Results suggest that the subscales of aggression and depression are invariant across sub-samples of the population. No single source of data is sufficient in diagnostic assessment of individual children or adolescents, however. For this and other reasons, cautious use and interpretation of the CBCL is recommended. It is also recommended that the CBCL be used as auxiliary data, employed in conjunction with other data. Includes 10 tables. (KW)