Investigating the Stigma of Prison Classification:An Experimental Design

Abstract

In spite of the widespread use of objective classification systems as a tool for effective prison management, little is known about the relationship between the classification label and inmate behavior. This study uses an experimental design to assign inmates to either maximum security (control group) or medium security (experimental group). Inmate disciplinary activity was monitored for a period of 1 year. Results indicate that there is no statistical difference between the experimental and control groups for disciplinary activity. These findings suggest that the classification label may be a determinant of behavior rather than a reflection of inmate characteristics.During the past three decades, objective classification systems have been widely touted as the foundation for the operation of the American prison system (Proctor, 1994). Prompted by rising prison populations and litigation that established the need for offenders to be classified into security and custody designations based on fair, objective criteria (Beard v.

Individual and group IQ predict inmate violence

“…Some studies have shown higher custody-level facilities to be positively related to violent misconduct (Camp et al, 2003;, while others fail to find an effect of custody type (Bench & Allen, 2003;Camp & Gaes, 2005;Cunningham & Sorensen, 2006a). Finally, sentence length is often studied as a potential predictor of misconduct.…”

Unintended consequences: experimental evidence for the criminogenic effect of prison security level placement on post-release recidivism

“…There has been an abundance of theoretical and psychometric research on the classification instruments that have been designed to predict and classify risk for individuals involved in different criminal justice processes, including prison (see volumes by Champion 1994, andTonry 1987), but only a few papers have evaluated the impact of security level placement on subsequent inmate behavior (Bench and Allen 2003;Berecochea and Gibbs 1991;Berk and de Leeuw 1999;Chen and Shapiro 2007;Camp and Gaes 2005). Instruments such as the level of supervision inventory (Bonta and Motiuk 1992), the salient factor score (Gottfredson et al 1978) and the Federal Bureau of Prisons inmate classification system (Kane 1986) have demonstrated that risk assessment instruments can be developed that predict prison misconduct, post-release recidivism, and other criminal justice behaviors.…”

“…Had some of the low risk level IV inmates remained in level IV prisons and been compared with their level IV counterparts placed in the high supervision level III prisons, the natural experiment would have been more relevant to theory and policy. Bench and Allen (2003) were able to assign maximum security inmates randomly to both maximum and medium security facilities. This occurred in a double blind experiment where neither the inmate nor the supervising staff knew the actual security risk score.…”

“…Recently, scholars have questioned whether classification assignments actually meet those goals and whether they have unintended post-release consequences (Bench and Allen 2003;Chen and Shapiro 2007). At a broader level, the concern is that prisons may make individuals more criminal upon release, the so-called criminogenic effect (Vieraitis et al 2007).…”

Prison gang integration and inmate violence

“…Prison crowding has been linked with inmate violence (e.g., Franklin et al, 2006;), as has sentence length (e.g., Camp et al, 2003;Craddock, 1996;Jiang & Fisher-Giorlando, 2002). Inmates' custody level, perhaps more indicative of "deprivation" than any other variable, has also been linked with violence, although not in a clear manner (Bench & Allen, 2003;Berecochea & Gibbs, 1991;Worrall & Morris, 2011).…”