The effects of substance type and use on public safety: A machine learning analysis



Journal Title

Journal ISSN

Volume Title



Drug use has been shown to have an effect on recidivism, but the extent of that effect, the types of drugs that produce that effect, and what types of crimes are being affected has not been addressed. This study attempted to answer some of those questions. The purpose of this dissertation is dissertation is to determine the effect drug type and intensity has on drug crime and property crime recidivism. Also, thresholds for identifying criminogenic need of substance use were explored. Using a criminal justice framework, the most accepted factors for criminogenic risk/need were assessed on 323,146 offenders in Texas Department of Criminal Justice who had been referred for drug treatment. A machine learning analysis revealed that risk was comparable to other studies on criminal reconviction. The analysis identified substances that were important in predicting recidivism of drug crimes and property crimes. Substance use and type were identified as significant variables in predicting recidivism, but they were not the most important. Variables ranging from arrests, convictions, and incarceration behavior to tattoos and age were more predictive of drug crimes and property crimes reconviction. Logodds plots identified the increased and decreased risk associated with the different levels of use of the important substance use variables. Logodds analysis also created a threshold at an individual level for identifying need. The results of this study revealed methodological, practice, and policy implications along with a variety of opportunities for future research.



Substance Use, Criminal Justice, Risk-Need-Responsivity, Risk Assessment, Policy