Managing Violent Misconduct in a Maximum Security Prison: Processes and Outcomes in a Restrictive Status Housing Program

The use of restrictive housing in prisons is at the forefront of national discussions on crime and punishment. Civil and human rights activists have argued that its use should be limited due to harmful effects on the physical and psychological health of inmates as well as its limited ability to reduce subsequent offending. Stacked against this is the need for correctional administrators to respond to institutional violence in a manner that ideally curtails future violence while doing no further harm to the well-being of those housed in these environments. The current project explores the effectiveness of a Restrictive Status Housing Program (RSHP) designed for inmates who commit violent assaults within the Arizona Department of Corrections. The program, as designed, moves beyond exclusively punitive approaches to segregation by encouraging behavior modification that is influenced by cognitive behavioral training. This study advances the literature and informs correctional policy by: 1) examining the effects of program participation on future behavioral outcomes, and 2) exploring mechanisms through which the program works (or does not work) by interviewing former RSHP participants and staff. The current research uses a mixed-method research design and was carried out in two phases. For Phase 1, quantitative data on behavioral outcomes of program participants (N = 240), as well as a carefully constructed comparison group (N = 1,687), will be collected and analyzed using official records over a one-year follow-up. Phase 2 will examine qualitative data derived from semi-structured interviews with former RSHP participants (n = 25) and correctional staff who oversee the day-to-day management of the program (n = 10). Results from the current study suggest that placement in the RSHP has null, and at times, an adverse effect on subsequent levels of institutional misconduct. Policy implications and recommendations based on these findings are discussed.