BAYESIAN DECISION THEORY FOR TREE-BASED ADAPTIVE SCREENING TESTS WITH AN APPLICATION TO YOUTH DELINQUENCY

• 成果类型: Article
• 署名作者: Krantsevich, Chelsea; Hahn, P. Richard; Zheng, Yi; Katz, Charles
• 署名单位: Arizona State University; Arizona State University-Tempe; Arizona State University; Arizona State University-Tempe
• 刊物名称: ANNALS OF APPLIED STATISTICS
• ISSN/ISSBN: 1932-6157
• DOI: 10.1214/22-AOAS1657
• 发表日期: 2023
• 页码: 1038-1063
• 关键词: adolescent problem behaviors; protective factors; measuring risk; communities; BIAS

摘要

Crime prevention strategies based on early intervention depend on accu-rate risk assessment instruments for identifying high-risk youth. It is impor-tant in this context that the instruments be convenient to administer, which means, in particular, that they should also be reasonably brief; adaptive screening tests are useful for this purpose. Adaptive tests constructed using classification and regression trees are becoming a popular alternative to tradi-tional item response theory (IRT) approaches for adaptive testing. However, tree-based adaptive tests lack a principled criterion for terminating the test. This paper develops a Bayesian decision theory framework for measuring the trade-off between brevity and accuracy when considering tree-based adap-tive screening tests of different lengths. We also present a novel method for designing tree-based adaptive tests, motivated by this framework. The frame-work and associated adaptive test method are demonstrated through an appli-cation to youth delinquency risk assessment in Honduras; it is shown that an adaptive test requiring a subject to answer fewer than 10 questions can iden-tify high-risk youth nearly as accurately as an unabridged survey containing 173 items.