Assessing the Impact of County-level Factors on Pretrial Detention Decisions

Document Type



Doctor of Philosophy


Criminology and Criminal Justice

Date of Defense


Graduate Advisor

Michael Campbell


Richard Rosenfeld

Lee Slocum

John Wooldredge


Scholarship devoted to understanding bail decisions and outcomes suggest that legal factors hold the blunt of the power in explaining these punishment decisions. However extra-legal factors, such as race, can still be at play in judicial decision-making, though results are mixed. It has been argued that cultures and norms of localities help to shape punitive attitudes, which can then in turn shape criminal justice actors’ decisions. Therefore, the factors that make up the locality can influence the context of which individuals are punished, and may lead to differential treatment across jurisdictions and defendant race/ethnicity. Drawing on two theoretical perspectives—focal concerns and group threat theory—the current study attempts to address the following research questions: First, how do county-level contextual factors related to group threat and violent crime rates inform bail decision outcomes for all felony defendants across jurisdictions? Second, are pretrial decisions more punitive in jurisdictions that have higher levels of politicization within the courts? Lastly, do county-level factors related to group threat, crime, and politicization shape pretrial decisions and outcomes differently for African-Americans or Hispanics than whites? To address these research questions, this dissertation utilizes the State Court Processing Statistics, which contain felony case filing data for individual defendants across 75 of the most populous counties for 1990, 2000, and 2009. County-level administrative data are then be linked, to allow for multilevel modeling, to assess the impact that county-level factors have on pretrial detention decisions and outcomes (bond amount and held on bail).

Results suggest that county-level factors related to group-threat, system constraints, and politicization are associated with significant differences between jurisdictions on average bail amounts and odds of a defendant being held on bail. Most importantly is the association between residential segregation and bail decisions, which finds that higher levels of segregation are associated with lower bail amounts for all defendants, regardless of race or ethnicity. Results also suggest that county-level factors influence bail decisions and outcomes differentially for defendants of different race/ethnic backgrounds. These results may better help to understand the use of criminal justice decisions to maintain the racialized social ordering within communities.

This document is currently not available here.