NCJ Number
221978
Journal
Journal of Quantitative Criminology Volume: 24 Issue: 1 Dated: March 2008 Pages: 33-49
Date Published
March 2008
Length
17 pages
Annotation
This paper examines various approaches for studying the mean length of stay in prison, providing a systematic analysis of existing measures and developing two new methods for estimating mean length of stay in prison.
Abstract
The authors first describe the best method for estimating mean length of stay, i.e., the life table, and then profile three other estimation methods, each of which relies on the assumption of a stationary population. They show what biases result in these measures when a population is not stationary and develop a method for correcting the biases in two of the three cases. The paper then demonstrates the performance of the four existing methods and two improved methods in populations that are not stationary. One set of such populations is termed "stable" populations in terms of demographics. Next, the authors use population simulation, which allows the imposition of shocks stemming from growth or decline in prison admissions and in changes in the length of stay inside prison. Both of these approaches permit the assessment of the sensitivity of the estimates to violation of assumptions and to error in data. The paper concludes with recommendations about the measures to be used for future examination of length of stay in prison. 7 tables, 27 references, and appended sources and processing of data and derivations