The sample design employed was a stratified two-stage
selection with the probabilities proportional to the size of the
correctional facility. The sample was selected independently from two
frames, one for males only and a second that allowed an oversample of
females. Within each frame, facilities were stratified by type (prison
versus community corrections facility) and by census region. In the
second stage, interviewers visited each selected facility and drew a
sample of inmates using predetermined sampling procedures. There were
275 facilities selected, yielding a sample of about 15,000 inmates. Of
these, 14,649 were contacted and 13,711 were interviewed successfully.
All adult inmates housed in correctional facilities
operated by a state government in the United States.
survey data