Structural and Criminogenic Risk Factors Associated with Probation Violations

Background: Approximately 1 in 53, or 4.6 million, U.S. adults are under correctional community supervision (i.e., probation, parole, or post-release supervision) at any given time. Estimates suggest that 1.8 million of these individuals (27%) experience mental illnesses and approximately 20% of the adult correctional population are early adults aged 18-29. Considering the large proportion of early adults, and adults with mental illnesses, this research aims to identify the impact of age cohort and mental health status on the number of violations experienced by probationers.

Methods: This observational, cohort study uses administrative data from 57,747 probationers in one southeastern state to identify the relationships among age group, gender, mental health symptomatology, and criminogenic risk and need, and probation violations.  The administrative data source contained data on participants’ socio-demographic backgrounds, a measure of criminogenic risk level and criminogenic needs, and probation event details. Socio-demographic variables included gender, date of birth, race, and possession of a high school diploma. Zero-inflated negative binomial regression is used to determine the variables impacting the number of violations experienced by the sample.

Results: 18,474 adult probations received no violations during their probation sentence. Multiple variables were associated with having a higher likelihood of being certain zeros for number of violations, however two variables were more highly associated than the others: individuals identified as low risk of reoffending by a criminogenic risk scale (b=-0.83, p<.001) and individuals of a racial group other than African American (b=-0.92, p<.001). For the early adult probationers, being identified as low risk of reoffending was associated with 2.32 higher likelihood of being in the certain zero group for probation violations (b=-.84, p<.001). Similarly, individuals with a high school diploma or equivalent experienced 1.77 higher odds of being in the certain zero violations group (b=.57, p<.001) and belonging to a racial group other than African American was associated with 3.00 higher odds of certain zero violations (b=-1.10, p<.001).

Implications: The findings from the zero-inflated negative binomial regression provide important considerations in regards to treatment targets for individuals on probation. Although individual risk factors (i.e., criminogenic risk and need) are predictive of both receipt of any violation and the number of violations incurred, other factors provide stronger correlations. The findings also provide evidence that more consideration is needed for structural factors potentially influencing decisions about probation violations. Structural variables (e.g., poverty, socioeconomic status, employment opportunities) are an important consideration of criminal justice involvement. In regards to race, though being African American does not imply certain violation, there are contextual assumptions that should be considered in regards to social and criminal injustice. Race could be used as a proxy for predictors such as neighborhood context and socio-economic status. Additionally, educational attainment was also a significant predictor of violation status. Considering these variables together, the findings give credence to the need to explore how structural factors are influencing decisions around criminal justice practices, including probation violations.