Empirical Study of Failure To Appear Rates Among Accused Offenders - Construction and Validation of a Prediction Scale

• NCJ Number: 81091
• Journal: Annual Journal of Pretrial Services Volume: 2 Dated: (1979) Pages: 105-117
• Author(s): C W Eskridge
• Date Published: 1979
• Length: 13 pages
• Sale Source: National Institute of Justice/
• Language: English
• Country: United States of America

Annotation

This study concerns the systematic development of three prediction devices that could be used to accurately identify failure-to-appear probability among potential recognizance release clientele. An attempt is also made to identify the impact of certain operation and management variables upon failure probability and to identify the impact of a failure-to-appear incident on subsequent case disposition and sanction severity.

Abstract

The analysis is based on two samples: an experimental group of 512 persons released on a recognizance bond under the Franklin County Pretrial Release Program (Ohio) in 1976 and a validation group of 103 randomly chosen persons released on bond under the same program during the same period. The typical program client is a young, unmarried, white male who has not completed high school. The study hypothesized that there were differences between individuals released on a reporting recognizance bond who failed on recognizance and those who did not fail as to (1) the severity of the present criminal charge, (2) the nature and extent of previous criminal activity, (3) past court appearance behavior, (4) community ties and socioeconomic background, (5) time lag from release to initial personal contact from program staff, and (6) sanction. Three analytical techniques were used to construct the three scales: an unweighted multiple regression technique, a weighted multiple regression technique, and a discriminant analysis technique. The scales derived from each analysis were applied to a validation sample, and three management tables were constructed from the validation sample, one for each scale validated. Employing a hypothetical case in order to apply the scales and tables, the study found that socioeconomic background and community ties were not significant indicators of failure probability. Moreover, as the severity of past criminal justice experience increased, the probability of failure on recognizance decreased. A failure incident tends to result in a more severe case disposition. Additional findings, a table, and 27 footnotes are supplied.