With the 2016 implementation, the survey was renamed the Survey of Prison Inmates.
The sample for the 2004 survey was selected from the universe of all federally-owned and -operated prisons as provided by the Federal Bureau of Prisons. The sample design was a stratified two-stage selection, in which prisons were selected in the first stage and inmates to be interviewed were selected in the second stage. In the first sampling stage, one female and two male facilities were selected with certainty based on the size of the facilities. The remaining facilities were separated into seven separate strata based on the security level of the facility and the male and female populations in each prison. Due to the large proportion of drug offenders in the Federal system, the second-stage sampling for the Federal survey involved a two-step process to ensure increased representation of nondrug offenders within the sample while still including enough drug offenders to produce reliable estimates on the drug offender population. First, an oversample of inmates was selected to ensure a large enough sample of non-drug offenders was selected; from this initial list, 1 of every 3 drug offenders was retained. The weighting procedure consisted of a base weight for each inmate, and four adjustment factors that produced the final weight for the survey. Data were collected through face-to-face interviews with inmates using computer-assisted personal interviewing.
The questionnaire is modified with each iteration in order to capture information on evolving issues in criminal justice and topics of the greatest interest to users. Recent modifications include enhanced questions on medical and mental health histories of inmates.