Abstract
This study investigated the independent effects of security level on rates of violent misconduct, and other forms of misconduct in Federal prison. With first-time federally incarcerated people, this study tested the effect of security level in a sharp intention-to-treat regression discontinuity to separate the effect from individual characteristics. Higher security was independently associated with higher rates of violent misconduct at medium (vs. low) and high (vs. medium) security. This study refers to this result as an “inflator effect.” Policy implications are discussed in the wake of the ban on Federal private prisons, and implications for risk assessment design are discussed.
Keywords
Get full access to this article
View all access options for this article.
