A Multilevel Analysis of County and State Variation in the Severity of Sentences Imposed in Large Urban Courts

This study explored the structural sources behind variability in the sentences applied to felons convicted in state courts located across the U.S. Multilevel regression models were used to explore whether various state and county-level attributes help to account for why defendants experience a significantly higher probability of incarceration versus probation in certain jurisdictions. Drawing upon a broad theoretical landscape, the analyses test several hypotheses derived from macro level theories of social control which predict that the legal and organizational culture of courts, and the socioeconomic and political attributes of the communities they serve, influence sentencing outcomes. This study sought to fill two important gaps in the existing research. First, it broadened the theoretical framework used to interpret community variation in punishment to include the impacts of state sentencing policies that have been linked to the increase in mass incarceration among U.S. states. The second major goal of this study was to bring new data to bear on the issue of whether social and cultural attributes of communities are associated with the severity of the sentences their courts impose. The analysis examines this issue by linking individual sentencing outcomes to aggregate-level General Social Survey (GSS) responses that capture community variation in public sentiment. The sentencing data used to test these hypotheses are derived from the State Court Processing Statistics (SCPS) for the years 1998, 2000, 2002, and 2004. Information on a sample of 26,000 felony cases in the SCPS were appended to a unique county and state-level database containing measures that capture variation in sentencing policy, criminal statutes, correctional resources, crime rates, court case load pressures, GSS survey responses, and census-derived demographic attributes. The findings indicate that the average probability of being sentenced to incarceration varies significantly across court jurisdictions, and that differences in the types of cases courts process do not account for this variation. Consistent with previous studies, the analyses reveal that commonly considered attributes such as county racial composition, levels of crime, and adverse economic conditions, exert weak or null effects on the outcomes of criminal cases. Analyses of the effects of legal policy reveal that defendants processed in jurisdictions with certain punitive sentencing policies do not face significantly higher odds of being incarcerated. This finding contradicts much of the theoretical and policy literature, which highlights the role of more punitive sentencing policies as a key factor responsible for the growth in mass incarceration. Certain aspects of the organizational context in which courts operate, including the amount of state correctional spending and higher monthly probation supervision fees, are associated with a significantly lower likelihood of going to prison. Models examining the effects of community social climate indicate that defendants convicted in communities with higher levels of anti-Black resentment among whites are significantly more likely to receive a prison versus a jail sentence, and that the odds of being sentenced to prison are significantly lower in jurisdictions where religious fundamentalism is more widespread. A series of supplementary models separating out drug versus non-drug felony cases suggests that the relationship between community context and sentencing outcomes is complex, and that certain aspects of the broader social and legal climate only impact the odds of incarceration for drug offenders. The implications of these findings for advancing contextual explanations of sentencing are discussed.