Who Are the Compliant Correctional Clients? New Evidence on Protective Factors among Federal Supervised Releases

Towards more accurate classification of risk of arrest among offenders on community supervision: An application of machine learning versus logistic regression

BACKGROUND

Although there is general consensus about the behavioural, clinical and sociodemographic variables that are risk factors for reoffending, optimal statistical modelling of these variables is less clear. Machine learning techniques offer an approach that may provide greater accuracy than traditional methods.

AIM

To compare the performance of advanced machine learning techniques (classification trees and random forests) to logistic regression in classifying correlates of rearrest among adult probationers and parolees in the United States.

METHOD

Data were from the subgroup of people on probation or parole who had taken part in the National Survey on Drug Use and Health for the years 2015-2019. We compared the performance of logistic regression, classification trees and random forests, using receiver operating characteristic curves, to examine the correlates of arrest within the past 12 months.

RESULTS

We found that machine learning techniques, specifically random forests, possessed significantly greater accuracy than logistic regression in classifying correlates of arrest.

CONCLUSIONS

Our findings suggest the potential for enhanced risk classification. The next step would be to develop applications for criminal justice and clinical practice to inform better support and management strategies for former offenders in the community.

The association between self-reported traumatic brain injury, neuropsychological function, and compliance among people serving community sentences

BACKGROUND

Traumatic brain injury is overrepresented in incarcerated samples and has been linked to a number of poor correctional outcomes. Despite this, no research has explored the impact of a recent TBI on compliance outcomes for individuals serving community-based.

METHOD

We screened for a history of TBI in 106 adults on community sentences and collected compliance (arrests, sentence violations) and related variables (e.g., risk scores, substance use) over 6 months. Sixty-four participants also completed the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS), the Comprehensive Trail Making Test and Color-Word Inference Test.

RESULTS

A TBI in the last year predicted a significantly higher likelihood of arrest, even when controlling for risk of reconviction and current substance use, but was not associated with non-compliance with sentence conditions nor with performance on the neuropsychological tests. In addition, no significant associations were found between performance on neuropsychological tests and measures of non-compliance.

CONCLUSIONS

TBI in the last year was an independent predictor of arrest. This result suggests that those with a recent TBI on a community sentence may need additional monitoring or support to reduce the risk of reoffending.

The p factor, crime, and criminal justice: A criminological study of Caspi et al.'s general psychopathology general theory

The general psychopathology general theory or p Factor is an influential theoretical development in the social and behavioral sciences, but has yet to gain traction in criminology and criminal justice. Drawing on data from a sample of 1722 federal pretrial defendants, we created a 22-item composite indicator or additive index of the p Factor containing externalizing, internalizing, substance use, paraphilic, and forensic indicators. Negative binomial regression models found that age, sex, and diverse forms of trauma exposure are associated with higher p Factor scores. Higher p scores strongly predicted total, violent, sexual, property, weapon, and drug arrest charges net the effects of demographic features and adverse childhood experiences. There is broad heterogeneity in psychopathology within this sample with nearly 29% of clients exhibiting zero psychopathology, nearly 61% showing average psychopathology or less, and nearly 40% evincing average to exceedingly high psychopathology. As a general theory, the p Factor has considerable potential to inform the assorted morbidities that often accompany criminal activity, including self-harm, reduced global functioning, substance use, and social dysfunction and thus is a parsimonious conceptual framework to understand the overlapping and systemic personal problems that typify chronic and serious criminal offenders.