An internal consistency reliability generalization of the Child Behavior Checklist

A reliability generalization study of the Child Behavior Checklist (1991) was conducted to examine variability in internal consistency reliability estimates reported in the literature. Relevant articles from three databases were retrieved, resulting in 2107 articles to analyze. Seventy-seven percent of the articles retrieved were unusable due to no reliability reported, 20% of the articles were unusable due to reliability induction, and 2% revealed usable internal consistency reliability data, leaving 40 (2%) articles providing 49 estimates. Overall reliabilities were adequate for the overall scale and many subscales. Moderator variables such as gender, ethnicity, respondent, age and population were not able to be analyzed due to the poor reporting of data among the studies. Overall reliabilities, while acceptable, were typically lower than manual-reported reliabilities. Results indicate recommended use of the CBCL in all settings, with modification or shortening of scales not recommended. Overall, the CBCL can be considered a sound, highly used, extremely validated instrument. When used in research, efforts should be undertaken to adequately report the demographic information related to reliability to determine the nature of the variability in the future.