This study examined the effects of poverty and State capital (State-sponsored support) on recidivism among women offenders, as well as whether criticisms of actuarial risk tools, such as the failure to take into account gender-related factors, have merit.
The study used longitudinal data from a community corrections sample of 134 female felony offenders. the participants were women from one county in Oregon and from the cities of Minneapolis and St. Paul, MN. The primary criterion for participant eligibility was a recent felony conviction in the counties selected as research sites. Two self-report recidivism measures were used in the analysis. Rearrest was coded as a dummy variable; it indicated whether the subject reported that she had been arrested between the initial and second interview. The second outcome measure, "violation," was also coded as a dummy variable; it reflected whether the subject reported that she had violated her probation or parole conditions during the study period. Independent variables were "minority," "education," "poverty" (annual legal household income below the designated poverty threshold), and "risk." Logistic regression was selected as the primary modeling technique. The findings show that poverty status increased the odds of rearrest by a factor of 4.6, and it increased the odds of a supervision violation by a factor of 12.7. In contrast, risk scares did not predict recidivism once poverty status was taken into account. Among poor women offenders, the study found that State-sponsored financial support to address short-term needs (e.g., housing) reduced the odds of recidivism by 83 percent. These findings suggest that the LSI-R, a commonly used actuarial risk assessment tool, does not give sufficient emphasis to the economic marginality of women offenders. Community corrections agencies should encourage their officers to refer and advocate for their poor female clients to receive available sources of State capital during the early stages of supervision. Welfare policy reforms should provide State-sponsored resources for child care, education, health care, and job training, all of which are factors that contribute to economic marginalization among women in general. 7 tables and 61 references
810 Seventh Street NW, Washington, DC 20531, United States
Dataset may be archived by the NIJ Data Resources Program at the National Archive of Criminal Justice Data