The researchers would also like to acknowledge the support of Mr. Aristedes Zavaras and the staff at the Department of Corrections. This project would never have been possible without the commitment and contributions of management, mental health, case management, offender services, and security staff.This document is a research report submitted to the U.S. Department of Justice. This report has not been published by the Department. Opinions or points of view expressed are those of the author(s) and do not necessarily reflect the official position or policies of the U.S. Department of Justice.ii ABSTRACT One of the most widely debated topics in the field of corrections -the use of long-term administrative segregation (AS) -has suffered from a lack of empirical research. Critics have argued that the conditions of AS confinement exacerbate symptoms of mental illness and create mental illness where none previously existed. Empirical research has had little to offer this debate; the scant empirical research conducted to date suffers from research bias and serious methodological flaws. This study seeks to advance the literature in this regard.This study tested three hypotheses: (1) offenders in AS would develop an array of psychological symptoms consistent with the security housing unit (SHU) syndrome, (2) offenders with and without mental illness would deteriorate over time in AS, but at a rate more rapid and extreme for the mentally ill, and (3) inmates in AS would experience greater psychological deterioration over time than the comparison groups.Study participants included male inmates who were placed in AS and comparison inmates in the general population (GP). Placement into AS or GP conditions occurred as a function of routine prison operations. GP comparison participants included those at risk of AS placement due to their institutional behavior. Inmates in both of these study conditions (AS, GP) were divided into two groups -inmates with mental illness (MI) and with no mental illness (NMI). A third comparison group of inmates with severe mental health problems placed in San Carlos Correctional Facility, a psychiatric care prison facility, was also included. A total of 302 inmates were approached to participate in the study, and 55 refused to participate or later withdrew their consent. Participants were tested at 3-month intervals over a yearlong period.Standardized test data were collected through self-report, correctional staff and clinical staff measures. Tests with demonstrated reliability and validity were selected to assess the eight primary constructs of interest: (1) anxiety, (2) cognitive impairment, (3) depression-hopelessness, (4) hostility-anger control, (5) hypersensitivity, (6) psychosis, (7) somatization, and (8) withdrawal-alienation. Extensive analyses of psychometric properties revealed that inmates self-reported psychological and cognitive symptoms in remarkably reliable and valid ways.The results of this study were largely inconsistent with our hypotheses and the bulk of literature that indi...