Toward a Theory of Bail Risk

• NCJ Number: 70884
• Journal: Criminology Volume: 18 Issue: 2 Dated: (August 1980) Pages: 147-161
• Author(s): M R Ozanne; R A Wilson; D L Gedney
• Date Published: 1980
• Length: 13 pages
• Sponsoring Agency: Pennsylvania Governor's Justice Cmssn
• Publication Format: Article
• Language: English
• Country: United States of America

Annotation

Criteria which are commonly employed in assessing bail risk (failure to appear) are examined; this article reports the findings of an analysis of the Vera Institute criteria performed in Pennsylvania.

Abstract

In the attempt to assess the usefulness of the Vera criteria as a predictor of failure to appear (FTA), researchers performed three separate analyses. The first involved extracting variables. The original sample consisted of 1,842 defendants who were interviewed by the Philadelphia Release on Recognizance (ROR) Program in 1973, between April and June. Information was gathered on elements directly related to the Vera criteria and on other demographic and personal characteristics of the defendants. In all, 150 variables were employed. The 1,842 defendants were followed for a period of 6 months; during this followup period, information was gained on whether the defendants appeared for all court appearances and whether bench warrants were issued. Variables that were significantly related to FTA were identified using a Goodman's Tau B. This procedure pinpointed 46 variables that were statistically significant in predicting failure to appear. From these 46 variables, 16 'predictor' variables were selected for further analysis based on the principles of accessibility, policy relevance, and legality. The followup study was conducted in Philadelphia (a sample of 2,226 defendants) and in Delaware County (a sample of 500). Relationships among the 16 variables were analyzed. Findings show that the original Vera criteria are not universal predictors of FTA; moreover, the underlying factors, which help explain FTA, appear to have a different mathematical (factor) structure than that implied in the Vera conceptualization of community ties. A significant implication of the findings is that differences in the ties between the communities and defendants must be considered in developing ROR criteria. Tables, footnotes and 15 references are provided.