A Comparison of Content-Balancing Procedures for Estimating Multiple Clinical Domains in Computerized Adaptive Testing: Relative Precision, Validity, and Detection of Persons with Misfitting Responses

摘要: This simulation study sought to compare four different computerized adaptive testing (CAT) content-balancing procedures designed for use in a multidimensional assessment with respect measurement precision, symptom severity classification, validity of clinical diagnostic recommendations, and sensitivity atypical responding. The were (a) no content balancing, (b) screener-based, (c) mixed (screener plus balancing), (d) full balancing. In balancing following administration the screener items, item selection was based on whether target number items item’s subscale reached information function. Mixed provided best representation from each main subscales Internal Mental Distress Scale. These also resulted higher CAT full-scale correlations Trauma Homicidal/Suicidal Thought ...