This study of prison rapes used an ethnographic, culturally relativistic methodology and was conducted between April 2004 and September 2005. The study was conducted in 30 correctional institutions, 23 men's and 7 women's, in 10 states. All 23 men's institutions were the highest-security level men's prison available in each state. When women's institutions were multi-security level and housed minimum, medium, and high-security inmates, they were selected from the highest-security level housing units within the institution. A total of 564 (409 male and 155 female) inmates were interviewed. The inmates to be interviewed were selected from the general prison population using a probability sample design. Average interview length was just under an hour. The sole mode of data collection was an open-ended, semistructured inmate interview. To ensure comparability of answers, surveys were designed with each query resting on a particular concept or variable. The same interview instrument was used for both male and female inmates. Questions were asked about inmate prison history, mental health, rape, social process, domestic violence and relationships, staff, institutional factors, and perception of social roles, and demographic information. Also included are lexical responses and free list questions such as "Why do inmates have sex with other inmates?"