Quality of prison operations in the US federal sector

Abstract

An examination of the quality of operations at private and public prisons is essential to making informed decisions about the desirability of using private and public prisons. Previous studies have used survey data collected from staff and inmates to compare prisons based on the proportions or percentages of staff making favorable evaluations of different items. As argued here and elsewhere, the previous studies have been flawed by methodological shortcomings. An approach is demonstrated here that allows for an examination of the aggregate measurement properties of survey items as well as for making comparisons of different prisons. The data are from surveys administered to staff at 96 federal prisons and two private prisons. The results demonstrate that not all survey items can be used indiscriminately to compare prisons, and when comparisons are made, it is essential to control for individual-level and institutional-level factors that are not related to institution performance but still influence survey results.