Date of Award
Doctor of Philosophy (PhD)
This study examined characteristics of youth with sexual behavior problems (N = 251) followed over approximately two years and applied Classification Tree Analysis (CTA) via Optimal Data Analysis (ODA) to produce a hierarchically optimal classification model to predict recurrence of SBP over time. This study was designed to address problems with clinical judgment leading to inaccurate diagnosis and in some cases, incorrect treatment or placement of the youth by ascertaining if an actuarial model designed to optimize classification accuracy might be more helpful in directing diagnosis and treatment for these youth. Results suggest initial event severity influences the potential for problem recurrence, with lower-severity (i.e., less invasive) SBP actually being more likely to recur than higher-severity SBP. The factor of Narcissism by self-report on the Antisocial Process Screening Device and the placement at time of initial event also influenced the prediction of whether or not SBP would recur. Classification performance was fair but lacks the specificity necessary for clinical usage. The classification accuracy of the model could be improved significantly in future studies with a larger sample size and a longer time of follow-up.
Morgan, David J., "Predicting Repeated Sexual Behavior Problems among Youth: A Hierarchically Optimal Classification Tree Analysis" (2007). Dissertations. 3909.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
© 2007 David J. Morgan