2021 Survey Methods
Starting in early 2021, the Correctional Leaders Association (CLA) Restrictive Housing Committee joined with Liman Center researchers to draft, streamline, and clarify questions for the 2021 CLA-Liman Survey. The drafting committee did a pilot survey. After feedback from correctional agencies, the researchers created a revised 170-question questionnaire with a dropdown menu.
That questionnaire (the 2021 CLA-Liman Survey) was distributed to all CLA members on July 1, 2021, via Qualtrics survey software linked to an online version which most respondents used; a few opted to fill in answers by hand, scan, and email completed surveys. When clarifications were needed, Liman Center researchers followed up and integrated answers into the data set; thereafter, layers of review aimed to ensure an accurate compilation. As in past reports, after compiling and analyzing the data, the research team circulated a draft report to the CLA membership for comments and corrections, and that feedback contributed to the published Report.
This research method did not provide direct access to jurisdictions’ internal data and recordkeeping but relied on survey responses provided by each jurisdiction. The 2021 Survey asked how jurisdictions collected the data they reported as well as whether and how frequently they reviewed data on restrictive housing in the course of their regular operations. The full 2021 CLA-Liman Survey is reproduced in Appendix A of the Report, Time-in-Cell: A 2021 Snapshot of Restrictive Housing.
The 2021 Survey collected information in reference to one date in time by asking jurisdictions to report, as of July 1, 2021 (or July 15, or another date in July 2021), both the total custodial populations under their direct control, and the number of people held in restrictive housing in those facilities. The Survey defined people under a jurisdiction’s "direct control" as people "sentenced to and received by" a jurisdiction. The point was to exclude "people who are sent out of the jurisdiction or held under local or county authority" because if a jurisdiction did not have control over the facility, it would be unlikely to have specific information on how people were housed. For example, in one state, almost half of post-conviction, sentenced individuals were not in that state’s prisons but held in local jails. Thus, the total number of people described by the thirty-five jurisdictions was less than the total number of people incarcerated under the legal authority of these jurisdictions. Thirty-five responding jurisdictions reported a total of 731,202 people under their direct control as of July 2021, while data from the Vera Institute of Justice (Vera) indicated that these jurisdictions had legal authority over almost seventy thousand more people, 802,821, who were incarcerated as of March 2021 when that data was collected.
To provide comparisons across years, many questions in the 2021 Survey used the same wording as prior surveys. The 2021 Survey defined restrictive housing, as had the 2017 and the 2019 Surveys, as "separating prisoners from the general population and holding them in cell for an average of 22 or more hours per day, for 15 or more continuous days."
The 2021 Survey defined “total custodial population” as “the total number of people sentenced to and received by your department.” (As noted above, the Survey specified that respondents should limit their responses to people under jurisdictions’ “direct control.”) People in restrictive housing, general population, and any other levels of custody including individuals in cell for twenty-one hours per day are all part of the total custodial population.
This method was built on those in use in previous surveys. For more information, see Time-in-Cell 2019 (published Sept. 2020), Reforming Restrictive Housing (published Oct. 2018), Aiming to Reduce Time-in-Cell (published Nov. 2016), and Time-in-Cell 2014 (published Sept. 2015).
Research Challenges and Caveats
First, the data, based on CLA-Liman Reports, come from information reported by responding jurisdictions which answered a questionnaire. No third-party site visits were done, nor were incarcerated individuals surveyed. In addition, responses to some questions posed challenges for providing cross-jurisdictional comparisons and for aggregating data. In the published versions of the materials summarized on this website, detailed explanations are provided as well as endnotes with explanatory caveats when needed. In addition to survey materials, this website draws on a range of materials including statutes, regulations, and other analyses related to solitary confinement.
Second, given that the Survey’s definition focused on people held 22 hours or more on average, 15 consecutive days or more, the data do not provide a full picture of isolation because people who are in for fewer than fifteen days or in for fewer than 22 hours per day would not have been included in answers. Another reminder is that the data were collected about people in prison and not about people held in solitary confinement in other settings including local jails, immigration detention, military confinement, and juvenile facilities. To be clear, tens of thousands of people are in such facilities. According to Vera, in the spring of 2021, when approximately 1,193,930 people were held in state and federal prisons, an additional 647,200 people were in jails around the United States. Moreover, in 2019, Vera estimated that 5.64% of the individuals were isolated in jails in cells for at least twenty-two hours per day.
For additional information on research challenges, see Time-in-Cell: A 2021 Snapshot of Restrictive Housing, Time-in-Cell 2019 (published Sept. 2020), Reforming Restrictive Housing (published Oct. 2018), Aiming to Reduce Time-in-Cell (published Nov. 2016), and Time-in-Cell 2014 (published Sept. 2015).