Study
Knoth and Ruback (2021) conducted a quasi-experimental design to assess the impact of criminal labeling on individuals who commit their first-time driving-under-the-influence (DUI) offense in the state of Pennsylvania. In Pennsylvania, individuals who commit their first-time DUI offense may be sentenced to a guilty conviction disposition or they may be sentenced to participate in the Accelerated Rehabilitative Disposition (ARD) Program. Both sentencing outcomes involve similar treatment conditions and requirements (i.e., up to 1 year of probation, 90 days of license suspension, and mandatory drug or alcohol treatment in a community facility). The major difference between the guilty disposition and the ARD program is that the ARD program allows participants the opportunity to avoid a criminal record on completing the program successfully. The study authors examined whether the ability to expunge a criminal record following completion of the ARD program influenced future criminal behavior.
Future criminal conviction or recidivism is a dichotomous variable operationalized as rearrest within 4 years of the final disposition. The main independent variable of interest is a dichotomous variable indicating whether those who committed a first-time DUI offense were diverted to an ARD program (59.9 percent of the sample) or sentenced to a guilty conviction (40.1 percent of the sample). Other predictors in this study included factors about the person who offended, the offense, the individual’s criminal history, and the county where the individual was sentenced. Those who committed a first-time DUI offense and were sentenced to a guilty conviction (35 percent) were more likely than those receiving ARD (28 percent) to recidivate.
Several data sources were used to conduct the analyses for the study. Data from the Administrative Office of Pennsylvania Courts (AOPC) were used to select the initial list of statewide DUI cases. Only those who were sentenced to their first DUI offense, did not have a child younger than 14 in the vehicle at the time of arrest, and were not involved in a crash that caused serious injury or death were eligible for the ARD program. To clearly assess racial differences, the sample included only those who identified as either white or Black. Additionally, the sample was limited to those who were sentenced during 2006 or 2007 in a Common Pleas Court, were sentenced to either a conviction or ARD (not both), and who had data available for all 4 years of the follow-up period. Data from the Pennsylvania State Police were used to obtain records of arrest and prosecution (RAP sheets), which provided necessary information on arrests before and after the primary DUI offense for those included in the sample.
The final sample size was 34,135 and included slightly more cases from 2007 (n = 20,357) than 2006 (n = 13,778). Those in the sample were mostly male (76.4 percent), white (92.2 percent, with the remainder of the sample Black), and between the ages of 21 and 45 (the average age was 33.3 years). Those who committed a first-time DUI offense and were sentenced to ARD (the treatment group) were statistically significantly different from those who were sentenced to a guilty conviction disposition (the comparison group). Females (69 percent), whites (61.0 percent), and older individuals (32.05) were more likely to be sentenced to ARD than males (57 percent), Blacks (44.0 percent), and younger individuals (31.8). Individuals sentenced to ARD had fewer prior arrests (0.38) and were older at the time of their first arrest (29.67). To account for differences between the treatment and comparison groups, the study authors statistically controlled for confounders in a Multilevel Logistic Regression model assessing the impact of ARD on the odds of rearrest 4 years after the final disposition. The study conducted a subgroup analysis assessing differences in recidivism by diversion/guilty conviction for white males, Black males, white females, and Black females.