Skip navigation
Please use this identifier to cite or link to this item:
Title: Closing the Revolving Doors: Designing Prison Educational Programs to Reduce Recidivism
Authors: Katticaran, Julu
Advisors: Piehl, Anne
Department: Woodrow Wilson School
Class Year: 2015
Abstract: Background: Correctional facilities in the United States house over 1.5 million prisoners, a quarter of the world’s prison population, and cost taxpayers over $30,000 per inmate per year. Repeat offenders form a substantial proportion of our prison population. As a result, correctional policymakers are increasingly looking to education programs in prisons as a means to reduce the number of offenders who recidivate. While a large body of research has concluded that correctional education programs as a whole can help reduce post-release recidivism, almost no research has attempted to identify the particular types of programs that are most effective at achieving this outcome. Objectives: This thesis seeks to compare the impact on recidivism of two widely utilized correctional education programs – a high school program and a GED program – which are offered in facilities run by the New Jersey Department of Corrections. Hypothesis: In the absence of research comparing high school and GED programs in correctional settings, I turn to the literature on these two programs in non-correctional settings. This body of evidence strongly suggests that high school programs are better at promoting favorable employment and labor market outcomes. Given that favorable labor market outcomes aid in reducing recidivism, I hypothesize that inmates enrolled in the high school program will be less likely to recidivate relative those enrolled in the GED program. Dataset: I use a rich and unique administrative dataset – the New Jersey Department of Corrections Youth Correctional Education Program dataset – which I collected form the New Jersey Department of Corrections over the course of four months. Methods: I estimate the causal effect of the GED program relative to the high school program on recidivism by exploiting a discontinuity in the assignment of inmates to each of these programs. I arrive at my preferred estimates using an Instrumental Variables (IV) model that makes use of this discontinuity. Findings: Contrary to expectations, I find that inmates in the GED program are less likely to recidivate post-release relative to inmates in the high school program. Policy Recommendations: Based on this result and other findings about how the program structures interact with prison population characteristics, I suggest that the New Jersey Department of Corrections reconfigure the mechanism by which inmates are assigned to educational programs. I estimate that adopting this policy recommendation will result in cost savings to the New Jersey Department of Corrections of over $14,000 per inmate. In addition to this substantial economic benefit, the state of New Jersey will also stand to benefit from lower levels of crime.
Extent: 114 pages
Type of Material: Princeton University Senior Theses
Language: en_US
Appears in Collections:Woodrow Wilson School, 1929-2017

Files in This Item:
File SizeFormat 
PUTheses2015-Katticaran_Julu.pdf3.31 MBAdobe PDF    Request a copy

Items in Dataspace are protected by copyright, with all rights reserved, unless otherwise indicated.