U.S. flag

An official website of the United States government, Department of Justice.

The Relationship between Prison Length of Stay and Recidivism: A Study using Regression Discontinuity with Multiple Break Points

NCJ Number
251410
Date Published
October 2017
Length
86 pages
Agencies
BJS-Sponsored
Grant Number(s)
2016-BJ-CX-K044
Annotation
This BJS grant-funded report evaluates the relationship between prison length of stay and recidivism post-release.
Abstract
This BJS grant-funded report examines Federal sentencing structure to determine if increasing the length of prison terms increases or decreases recidivism, and seeks to measure if the effect of increasing the length of prison differs across individual inmates. The authors contend that imprisonment is an expensive sanction and justifying its use often rests partly on its presumed utility to reduce post-release reoffending. Most scholars separate the research on imprisonment effects into two subsets: imprisonment in contrast to an alternative sanction and prison length of stay. The authors attempt to answer if prison is expected to deter offenders from future reoffending, then how does it compare to a sentence of probation, home confinement or other alternative sanction? Likewise, if prison is chosen as a preferred sanction, can a deterrent effect of imprisonment be achieved with a shorter sentence? A secondary goal of this study was to measure length of stay treatment heterogeneity; specifically, does the effect of increasing the length of prison differ across individuals? Scholars have argued the effect of prison may depend on a host of factors, such as, the characteristics of the offender, institution, and sentence. Incarceration treatment heterogeneity is important because unpacking treatment effect dependence may lead to an explanation of why various outcomes are observed across imprisonment studies found in systematic reviews.
Date Created: November 29, 2017