2019Journal of Studies on Alcohol and DrugsRequires access

Gender-Specific Predictors of Recidivism Among DUI Offenders

Angela A. Robertson, Qian M. Zhou, Andrew Tatch, Melanie Walsh

Open publisher page 16 citations

Abstract

OBJECTIVE: Women are less likely than men to be arrested for driving under the influence (DUI) of alcohol or another drug, yet their proportion of DUI offenders is growing. Understanding how DUI recidivism risk varies for men and women is of practical utility for DUI assessment and intervention programs. The goals of the current study are to determine if there are different sets of predictors for men and women and whether gender-specific DUI recidivism risk scales perform better than a single recidivism scale for both men and women. METHOD: We rely on statistically driven techniques to develop gender-specific and total sample recidivism risk scales. We then test the ability of the scales to predict recidivism within 12 months among a large sample (N = 10,827, 22.3% female) of DUI offenders court mandated to a DUI intervention in Mississippi. RESULTS: Predictors of recidivism were drawn from measures of criminal history, substance use disorders, driving behaviors, and accidents. Gender-specific models yielded different sets of recidivism risk factors for men and women, with minimal overlap between the two. Male risk factors were criminal history and heavy alcohol consumption. For women, evidence of a substance use disorder was a unique predictor. Having a prior DUI arrest, driving behaviors, and a physical health consequence of alcohol or drug use (i.e., weight loss) were shared predictors for both sexes. CONCLUSIONS: Findings suggest that within broad categories of risk factors, the predictive validity of specific assessment items may vary by sex. Our methods represent progression toward more efficient prediction of DUI recidivists.

About this research paper

What this paper is about

OBJECTIVE: Women are less likely than men to be arrested for driving under the influence (DUI) of alcohol or another drug, yet their proportion of DUI offenders is growing. Understanding how DUI recidivism risk varies for men and women is of practical utility for DUI assessment and intervention programs. The goals of the current study are to determine if there are different sets of predictors for men and women and whether gender-specific DUI recidivism risk scales perform better than a single recidivism scale for both men and women. METHOD: We rely on statistically driven techniques to develop gender-specific and total sample recidivism risk scales. We then test the ability of the scales to predict recidivism within 12 months among a large sample (N = 10,827, 22.3% female) of DUI offenders court mandated to a DUI intervention in Mississippi. RESULTS: Predictors of recidivism were drawn from measures of criminal history, substance use disorders, driving behaviors, and accidents. Gender-specific models yielded different sets of recidivism risk factors for men and women, with minimal overlap between the two. Male risk factors were criminal history and heavy alcohol consumption. For women, evidence of a substance use disorder was a unique predictor. Having a prior DUI arrest, driving behaviors, and a physical health consequence of alcohol or drug use (i.e., weight loss) were shared predictors for both sexes. CONCLUSIONS: Findings suggest that within broad categories of risk factors, the predictive validity of specific assessment items may vary by sex. Our methods represent progression toward more efficient prediction of DUI recidivists.

Why it matters

OpenAlex reports 16 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

OBJECTIVE: Women are less likely than men to be arrested for driving under the influence (DUI) of alcohol or another drug, yet their proportion of DUI offenders is growing. Understanding how DUI recidivism risk varies for men and women is of practical utility for DUI assessment and intervention programs. The goals of the current study are to determine if there are different sets of predictors for men and women and whether gender-specific DUI recidivism risk scales perform better than a single recidivism scale for both men and women. METHOD: We rely on statistically driven techniques to develop gender-specific and total sample recidivism risk scales. We then test the ability of the scales to predict recidivism within 12 months among a large sample (N = 10,827, 22.3% female) of DUI offenders court mandated to a DUI intervention in Mississippi. RESULTS: Predictors of recidivism were drawn from measures of criminal history, substance use disorders, driving behaviors, and accidents. Gender-specific models yielded different sets of recidivism risk factors for men and women, with minimal overlap between the two. Male risk factors were criminal history and heavy alcohol consumption. For women, evidence of a substance use disorder was a unique predictor. Having a prior DUI arrest, driving behaviors, and a physical health consequence of alcohol or drug use (i.e., weight loss) were shared predictors for both sexes. CONCLUSIONS: Findings suggest that within broad categories of risk factors, the predictive validity of specific assessment items may vary by sex. Our methods represent progression toward more efficient prediction of DUI recidivists.

Key concepts: Recidivism, Driving under the influence, Poison control, Injury prevention, Human factors and ergonomics, Psychology, Occupational safety and health, Suicide prevention

Related papers

Back to paper searchBrowse research topicsOriginal source
Gender-Specific Predictors of Recidivism Among DUI Offenders — Research Paper | ScholarLens