1974, 1979, 1986, 1991, 1997, 2004, 2016
The Survey of Prison Inmates (SPI) is a periodic, cross-sectional survey of the state and sentenced federal prison populations. Its primary objective is to produce national estimates for the state and sentenced federal prison populations across a variety of domains, including but not limited to demographic characteristics, current offense and sentence, incident characteristics, firearm possession and sources, criminal history, socioeconomic characteristics, family background, drug and alcohol use and treatment, mental and physical health and treatment, and facility programs and rule violations. From January through October 2016, data were collected through face-to-face interviews with prisoners using computer-assisted personal interviewing (CAPI).
Prior iterations of the SPI were known as the Survey of Inmates in State and Federal Correctional Facilities (SISFCF), which was renamed with the 2016 implementation. The first survey of state prisoners was fielded in 1974 and thereafter in 1979, 1986, 1991, 1997, and 2004. The first survey of federal prisoners was fielded in 1991, along with the survey of state prisoners, and since then both surveys have been conducted at the same time using the same questionnaire and administration.
The target population for the 2016 SPI was prisoners age 18 and older who were held in a state prison or were serving a sentence to federal prison in the United States during 2016. Similar to prior iterations, the 2016 survey was a stratified two-stage sample design in which prisons were selected in the first stage, and prisoners within sampled facilities were selected in the second stage. The SPI sample was selected from a universe of 2,001 unique prisons (1,808 state and 193 federal) that were either enumerated in the 2012 Census of State and Federal Adult Correctional Facilities or had opened between the completion of the census and July 2014 when the SPI sample of prisons was selected. A total of 364 prisons (306 state and 58 federal) participated in the 2016 survey out of the 385 selected (324 state and 61 federal). The other 21 prisons were not included, either due to nonresponse or ineligibility. The first-stage response rate (i.e., the response rate among selected prisons) was 98.4%: 98.1% among state prisons (all but 6 out of 312) and 100% among federal prisons. A total of 24,848 prisoners participated (20,064 state and 4,784 federal) in the 2016 SPI based on a sample of 37,058 prisoners (30,348 state and 6,710 federal). The second-stage response rate (i.e., the response rate among selected prisoners) was 70.0%: 69.3% among state prisoners and 72.8% among federal prisoners.
Responses from interviewed prisoners in the 2016 SPI were weighted to provide national estimates. Each interviewed prisoner was assigned an initial weight corresponding to the inverse of the probability of selection within each sampled prison. A series of adjustment factors were applied to the initial weight to minimize potential bias due to nonresponse and to provide national estimates.
For more information on the 2016 SPI methodology, see Methodology: Survey of Prison Inmates, 2016 (NCJ 252210).