SSV Part B_final

2. Collection of Information Employing Statistical Methods

1. Universe and Respondent Selection

The Survey of Sexual Victimization (SSV) collects information on allegations and substantiated incidents of sexual victimization that occur in adult correctional and juvenile justice facilities. BJS uses a series of sampling frames to identify the universe of facilities covered by PREA. This universe is fluid due to changes in the operational status of facilities, including openings, closings, new contracts (i.e., privately operated facilities), and ended contracts (i.e., no longer privately operated).

Over 6,000 facilities are covered by PREA. This includes about 4,650 adult prisons and jails and 1,780 juvenile facilities. PREA requires that BJS collect data from a sample of at least 10% of these facilities. Table 1 shows the number of facilities and reporting units, by facility type, covered by PREA and sampled in the 2022 SSV using the previous sampling methods.

Table 1. Number of facilities and reporting units in 2022 covered by PREA and included in the SSV.
			Facilities		Reporting Units
Facility type		Form	Universe		Universe	Sampled	Percent sampled
	Total		6,426		4,930	1,532	31.1%
Prisons
	Public – federal system	SSV-1	111		1	1	100%
	Public – state systems	SSV-2	1,155		50	50	100%
	Private	SSV-4	372		372	155	41.7%
Jails
	Public – local jurisdictions	SSV-3	2,867		2,867	700	24.4%
	Private	SSV-4	20		20	15	75%
Other adult facilities
	Indian country jails	SSV-4	58		58	25	43.1%
	Military systems	SSV-4	27		4	4	100%
	ICE	SSV-4	39		39	39	100%
Juvenile facilities
	Public – state/DC systems	SSV-5	336		51	51	100%
	Public – local	SSV-6	616		616	244	39.6%
	Private	SSV-6	804		804	227	28.2%
	Indian country	SSV-6	21		21	21	100%

Because of the low numbers of reported sexual victimizations to correctional authorities and the centralized authority at the system level that governs responses to the surveys, BJS elected to conduct a complete enumeration at the system level – including the Federal Bureau of Prisons (BOP), all state department of corrections (DOC), all state juvenile justice systems, and each branch of the U.S. military. In each of these cases, both the information systems and the authority to report are centralized. Moreover, this annual enumeration minimizes burden on the respondents (rather than selecting a sample from the more than 1,600 facilities operated by these systems). About 2 months prior to launching the survey each year, BJS receives an update of the listing of state and federal prison administrators from the Correctional Leaders Association (formerly the Association of State Correctional Administrators) and state administrators from the Council of Juvenile Justice Administrators, who serve as the current year’s data collection respondents. Each year, the four military branches provide BJS with a single point-of-contact for their centralized reporting.

BJS also conducts a complete enumeration of all facilities operated by or under contract with the U.S. Immigration and Customs Enforcement (ICE) primarily housing detainees for ICE, called dedicated ICE facilities, and juvenile justice facilities in Indian country housing exclusively juveniles. Unlike the full enumeration of facilities with a central reporter, a respondent at each of the dedicated ICE adult facilities and Indian country juvenile facilities provide BJS data. BJS annually obtains a list of the dedicated ICE facilities from a point-of contact at ICE and identifies the full list of Indian country juvenile facilities from the Annual Survey of Jails in Indian Country (ASJIC) (described below under “Other jails.”).

Finally, private prisons, local jail jurisdictions, private jails, Indian country jails, and local and private juvenile facilities are completely decentralized. As part of the 2022-2023 SSV redesign, researchers from the Complex Survey Methods and Analysis branch of the Economic Statistical Methods Division at the U.S. Census Bureau conducted an evaluation of the previous sample designs and made recommendations of how the samples for each could be improved (See Attachment 7 for the final report and recommendations). BJS evaluated the recommendations and has chosen to adopt those most feasible and appropriate. The following sections include information about each facility type and how they were previously sampled, then a summary of the new proposed designs for SSV 2023-2025. Annually, BJS will review and evaluate the precision of each of these revised designs. If changes are needed, BJS will submit a request for a substantive change for SSV 2024 and 2025.

Adult Correctional Facilities

The BOP, all state prisons, all military facilities and all facilities operated by or under contract with the ICE are included in the SSV each year. Below, there are descriptions of the previous methods used for sampling of private prisons, public jails, private jails, and Indian country jails in 2022 followed by new proposed plans for these facility types in 2023-2025.

Private Prisons – Previous Methods, 2022

Each year, the sample of private prisons has been drawn from the most recent available Census of State and Federal Adult Correctional Facilities (CCF, OMB Control No. 1121-0147), which is conducted every 5-7 years by BJS. The most recent CCF was conducted in 2019 and used for the 2022 SSV. The 2019 CCF has been updated with out of scope and closed facilities that are identified during the data collection process each year of the SSV.

In 2022, a sample of 155 (29%) privately operated state and federal prison facilities was drawn from the 372 private prisons identified in the 2019 CCF. This list of private prisons has been continuously updated based on identified closures or out of scope facilities since the 2019 CCF was conducted. The facilities were first sorted by average daily population (ADP) and a pre-determined cut-off (ADP of 258 or more) was used to determine 64 facilities as certainty-due-to-size units. The facilities remaining in the frame at this point were sorted by region (i.e., Northeast, Midwest, South, or West), state, and ADP, and 91 facilities were sampled systematically with probability proportional to size. A CV was produced to estimate sample precision based on the total frame ADP. For males, the CV was 1.1% and for females the CV was 8.0%.

Private Prisons - New Proposed Sampling Plan for 2023-2025

The private prison sample for 2023, 2024 and 2025 will be drawn from the universe identified through a version of the 2019 CCF that has been updated annually after each iteration of the SSV. If a new CCF is completed during this period, the sample will be drawn from the updated CCF. A sample of 155 private prisons from a population of 372 would result in a relative standard error (RSE) for an unknown population estimate of about 6% with a 95% confidence level. Decreasing the RSE to 5% would require a sample size of 190. Data collection including another 45 facilities would require increased resources as well as data quality and nonresponse follow-up which would not be effective given the marginal decrease in variability of the estimate. Therefore, the same total sample size as previous iterations of the SSV, 155 facilities, will be selected from the frame. Certainty units will be selected due to size. The ADP cut-off for certainty units will be determined using the cumulative square root of frequency (cum-root-f) method (Cochran, Sampling Techniques, 1997 edition, p. 129). In a simulation of this method using the 2019 CCF frame at the time of the 2019 SSV, the cutoff threshold was an ADP of 1,040 with 42 private prisons in the certainty strata, which would have resulted in 113 units being selected as noncertainty units.

The remaining noncertainty units will be selected using stratified random sampling, with the number of strata determined using the cum-root-f method. There will be three total strata, one of all certainty units and two strata of noncertainty units. Because this is a data driven method that will be informed by the number of units on the frame and the ADPs of those units, the number of units in each strata will not be known until all closed and out of scope facilities from the previous year have been identified. The number of units sampled from each stratum will be determined using optimal allocation. This number will also depend on the updated frame.

Public Jails – Previous Methods, 2022

Starting with the 2019 SSV, the Census of Jails (COJ, OMB Control No. 1121-0100) was used as the sampling frame for public jails. The COJ is a periodic data collection of all local jail facilities conducted every 5-6 years. The most recent COJ was collected in 2019 and was used to draw the sample of public jails for the 2022 SSV. If the SSV data collection process identifies sampled public jails closed or out of scope, they have been removed from the frame.

For the 2022 SSV, a sample of 700 (24%) publicly operated jail jurisdictions was selected from the 2,867 jails on the updated 2019 COJ. They were stratified into two certainty strata and three noncertainty strata. For the first stratum, the largest jail jurisdictions in 45 states and the District of Columbia were selected to meet the PREA requirement that at least one jail per state is selected each year.¹ For the second stratum, all other jail jurisdictions with ADPs greater than or equal to 1,000 inmates were selected with certainty. A total of 105 jail jurisdictions were selected as certainties-due-to-size. The remaining 549 jail jurisdictions on the frame were grouped into three strata. The cum-root-f method was used to determine the boundaries of these three strata based on ADP as the measure of size (Cochran, 1997). The jail jurisdictions in each of these three strata were then sorted by region, state, and ADP and selected systematically. Again, CVs were produced to estimate sample precision based on the total frame ADP. For males, the CV was 0.8% and for females the CV was 1.3%.

Public Jails - New Proposed Sampling Plan for 2023-2025

The public jail sample for 2023, 2024 and 2025 will be drawn from the universe identified through the most recent COJ available. A sample of 700 public jails from a population of approximately 2,867 on the frame would result in an RSE for an unknown population estimate of 3.3% with a 95% confidence level. Decreasing the RSE to 2% would require a sample size of 1,336. Data collection including another 540 facilities would greatly strain data collection efforts for a marginal decrease in variability of the estimate. At present, the sample size of 700 produces acceptable error and can be supported with the current financial and staffing resources. The largest jail jurisdictions in 45 states and the District of Columbia will be selected to meet the PREA requirement that at least one jail per state is selected each year. A certainty-due-to-size stratum will be selected using an ADP cutoff determined using the cum-root-f method. The noncertainty strata will use the cum-root-f method to form three size classes of ADP, and their sample size will be calculated with optimal allocation based on the number of confined persons (as of 12/31 in a given year). There will be a total of four strata, one with all certainty units (certainties of largest in each state and certainties by size) and three that contain the non-certainty units. Within the noncertainty strata, units will be serpentine sorted by region, state, and ADP. Sequential random sampling will be used to select units from each of the strata. Because the cum-root-f method is a data driven approach, the number of certainty units and non-certainty units in sample as well as the ADP cutoffs for each type of strata will be impacted by the number of units on the updated COJ frame and the most recent ADP of the units. While these specific values will change each year, a simulation study that included the proposed cum-root-f methods using the original 2019 COJ frame concluded the certainty size threshold would have been an ADP of 1,595 and 60 units would be certainties due to size at that threshold, with 640 noncertainty sampled units having then been drawn from the remaining 2,822 units on the frame (See Attachment 7, page 29).

Private Jails – Previous Methods, 2022

In 2022, a sample of 15 of the 20 (75%) privately operated jails was selected from the 2019 COJ file (described above under “Public Jails”). Two were selected with certainty due to size (ADP of 1,000 or more). Nine additional jails met the statistically derived certainty due to size criteria and were selected with certainty. The remaining private facilities were sorted by region, state, and ADP, and 4 jails were systematically sampled with probability proportional to size. Given the large standard errors, estimates for private jails are combined with public jails. The separate sample is used to ensure inclusion of private jails in the SSV. Again, CVs were produced to estimate sample precision based on the total frame ADP. For males, the CV was 4.9% and for females the CV was 8.3%.

Private Jails - New Proposed Census Plan for 2023-2025

The 2023, 2024 and 2025 private jail respondents will also be drawn from the most recent COJ, which includes an indicator that identifies privately operated facilities. Rather than drawing a sample, all private jails will be included each year representing a full enumeration of the facility type.

Indian Country Jails – Previous Methods, 2022

In 2022, the Indian country jail sample was drawn from the ASJIC. BJS maintains the ASJIC, which is updated annually and includes all known Indian country adult correctional and juvenile justice facilities operated by tribal authorities or by the U.S. Department of the Interior, Bureau of Indian Affairs. The prior year of ASJIC has been used for the SSV sample each year.

For jails in Indian country, 25 (43%) of the 58 jails in Indian country housing adult inmates were selected. Ten facilities were sampled with certainty due to their size (ADP of 60 or more). The remaining facilities on the frame were sorted by state and ADP, and 15 facilities were sampled with probability proportionate to size. Facilities in Indian country housing exclusively juveniles were excluded from the adult sample. CVs were produced to estimate sample precision based on the total frame ADP. For males, the CV was 3.4% and for females the CV was 6.2%.

Indian Country Jails - New Proposed Sampling Plan for 2023-2025

BJS will continue to use the prior year of the ASJIC for each year of the SSV. A sample of 25 Indian country jails from a population of approximately 58 on the frame would result in an RSE for an unknown population estimate of nearly 15% with a 95% confidence level. In order to achieve an RSE of 5%, 50 of the 58 tribal jails would need to be sampled, nearly a full census of this facility type. All of these facilities are contacted annually for the ASJIC data collection and adding a burden to each to complete the SSV every year would put additional strain in these facilities. BJS will continue to sample 25 jails under the proposed plan. The largest facility in each region (4 facilities) will be selected with certainty to ensure all regions are represented. Additional facilities will be selected with certainty-due-to-size. The ADP cut-off for these certainty-due-to-size units will be determined using the cum-root-f method. All certainty units will be in a single stratum. In a simulation of this method using the original 2019 ASJIC frame, this method would have resulted in cutoff ADP of 86, with 8 units included as certainties and the remaining 17 being selected as noncertainty units from the remaining 52 units on the frame. The remaining noncertainty units will be contained in one stratum. The sample size for the noncertainty stratum will be determined using optimal allocation, and units will be randomly selected within that stratum. Because the cum-root-f method is a data-driven approach, the number of certainty units and non-certainty units in sample as well as the ADP cutoffs for each type of strata will be impacted by the number of units on the updated ASJIC frame and the ADPs of those units.

Juvenile Justice Facilities

All juvenile state central reporters and facilities that report separately are included in the SSV each year. All facilities in Indian country holding exclusively juveniles are also included each year. Descriptions of the previous sampling methods from 2022 for local and private juvenile justice facilities, including privately-operated facilities, local facilities and detention centers are described below, followed by a description of the new proposed methods for 2023-2025.

Local and Private Juvenile Justice facilities – Previous Methods, 2022

In 2022, the local and private juvenile facility sample was drawn from two frames which are used in alternating years. The Census of Juveniles in Residential Placement (CJRP), which collects population data, is conducted by the Census Bureau for the Office of Juvenile Justice and Delinquency Prevention (OJJDP) in even-numbered years. The Juvenile Residential Facility Census (JRFC), which collects data on facility characteristics, is conducted, again by the Census Bureau, for OJJDP in odd-numbered years. The most recent frame is from the year preceding the SSV collection.

The 2021 CJRP contained a total of 1,758 non-tribal juvenile facilities. All state central reporters (51 potential respondents with 46 that were active and eligible) that oversee 336 facilities were included in the sample with certainty. Thirty-six units were selected with certainty as the largest locally operated facility in each state, 49 units were selected with certainty as the largest privately operated facility in each state, and an additional 119 detention, local, or private facilities were determined to be certainties-due-to-size. In total, 540 units were sampled with certainty, excluding Tribal facilities. The remaining facilities were serpentine sorted by geographic region (Midwest, Northeast, South, and West), state, facility type (detention, local, or private), and size. The remaining detention facilities were sampled with probabilities proportionate to size from 4 strata based on geographic region, the remaining local facilities were divided into 2 strata based on commitment status (commitment and non-commitment), and sampled with probabilities proportionate to size, and the remaining private facilities were sorted by region and state, then sampled with probabilities proportionate to size. A total of 326 additional facilities were sampled at this stage, for a total of 866 sampled facilities, excluding Tribal facilities (49% of frame). For the juvenile samples, CVs were calculated for juvenile offenders and non-offenders. The CV for juvenile offenders was 0.8% and the CV for juvenile non-offenders was 1.9%.

Local and Private Juvenile Justice Facilities - New Proposed Sampling Plan for 2023-2025

The most recent frame is from the year preceding the SSV collection. All strata definitions will remain consistent with previous sampling. The following strata will be included: state and central reporters and facilities that report separately, largest locally operated facility in each state, largest privately-operated facility in each state, detention facilities, local facilities, and private facilities. All state and central reporters and facilities that report separately, largest locally operated facilities in each state, and largest privately operated facilities in each state will be included in the sample with certainty.

In the 2022 SSV, there were 1,337 detention centers, locally operated and privately operated facilities and from that total, 540 were included in the sample as either certainty or non-certainty units. This sample size of 540 results in an RSE for an unknown population estimate of 3.3%. To decrease an RSE to 2%, it would require 871 local and private facilities to be sampled. There is significant turn-over of these facilities, with approximately 60 being closed or out of scope in a given year of the SSV. Because the frame is very fluid, including another 331 facilities in the sample could greatly increase the amount of time and effort required in nonresponse follow-up with a minimal decrease in outcome estimate variability. Therefore, BJS will continue to sample about 540 facilities under the proposed plan.

For Detention facilities, certainty-due-to-size units will be selected based on an ADP cut-off determined using the cum-root-f method. The remaining noncertainty units will be stratified based on region, then randomly selected using optimal allocation to determine sample size.

For local facilities, certainty-due-to-size units will be selected based on an ADP cut-off determined using the cum-root-f method. The remaining noncertainty units will be stratified based on commitment status (commitment or non-commitment), then randomly selected using optimal allocation to determine sample size. If there are two or fewer local commitment or noncommitment facilities in a specific region, those facilities will be included with certainty.

For private facilities, certainty-due-to-size units will be selected based on an ADP cut-off determined using the cum-root-f method. All certainty units will form one stratum. The remaining units with be included in one stratum and randomly selected using optimal allocation to determine sample size.

For 2023-2025, BJS’s data collection agent, the U.S. Census Bureau will provide the information in Table 2 as part of the sampling documentation to describe the frame and methods that will be used for each sampled facility type.

Table 2. Frames, strata, certainty and noncertainty units for sampled facility types

Facility Type	Frame source	# of facilities in Frame*	# of requested units	Types of certainty units	# certainty units*	How certainty units are defined	# of strata (including certainty strata)*	How strata are defined	# of sampled units per noncertainty strata*
Public jails	Census of Jails		700	Certainty by size Largest in each state		Cum root f clustering method Facility with largest MOS in the state		Cum Root F
Private prisons	Census of State and Federal Adult Correctional Facilities		155	Certainty by size		Cum root f clustering method		Cum Root F
Tribal jails	Annual Survey of Jails in Indian Country		25	Certainty by size		Cum root f clustering method		Cum Root F
Local and private juvenile facilities	Census of Juveniles in Residential Placement; Juvenile Residential Facility Census		530	Certainty by size Largest in each state		Cum root f clustering method		Cum Root F

*These values will change each year based on the total number of facilities in each frame.

While BJS is mandated to collect data under PREA (P.L. 108-79), overall, participation in the SSV has fluctuated in recent years. In 2016, 95% of sampled systems and facilities responded, in 2017, the overall response rate declined to 85%. This was due to delays in launching the collection which led to the shortening of the survey cycle. In the 2018 collection, nonresponse follow-up was hindered by COVID-19. As a result, the response rate remained at 85%. Response rates then improved to 91% for 2019 and again back to 95% in 2020. The overall response rate for the 2021 collection was 90%. Across many disciplines and types of data collections, response rates have been trending downward. However, given the history of this establishment collection and the mandate for BJS to collect these data, we aim to continually reach response rates above 90%. Table 3 below for more detail on response rates for 2021, the most recently completed SSV collection. The 2022 SSV collection is ongoing.

Table 3. Sampled reporting units by type of facility and response status, 2021.
Facility type		Form	Sampled	Active/Eligible	Responded	Response rate
Prisons
	Public – federal system	SSV-1	1	1	1	100%
	Public – state systems	SSV-2	50	50	50	100%
	Private	SSV-4	155	142	129	90.9%
Jails
	Public – local jurisdictions	SSV-3	700	696	634	91.1%
	Private	SSV-4	15	10	10	100%
Other adult facilities
	Indian country jails	SSV-4	25	25	20	80.0%
	Military systems	SSV-4	4	4	4	100
	ICE	SSV-4	31	31	31	100%
Juvenile facilities
	Public - state/DC systems*	SSV-5	51	46	46	100%
	Public – local	SSV-6	278	266	244	91.7%
	Private	SSV-6	252	200	163	81.5%
	Indian country	SSV-6	18	16	12	75.0%
Note: Arkansas, Connecticut, Montana, and South Dakota did not have state-operated juvenile facilities in 2021. *Public jails that are operated jointly by state departments of correction are captured on their state’s SSV-2 form. The states that operate integrated prison and jails systems are Alaska, Connecticut, Delaware, Hawaii, Rhode Island, and Vermont.

2. Procedures for Information Collection

Collection Procedure

BJS will administer the 2023-2025 SSV as a web survey. The website developed for previous iterations of the survey will continue to be used, however there are several edits to the website that will accommodate changes to the instruments (see Attachment 4 for current website screenshots that have not yet accounted for the changes to the instruments as this full clearance is pending). The website is hosted by the Census Bureau and is located on a secure server. Each respondent has an individual username and a password to enter the website. To improve data quality and lower cost, all respondents are encouraged to submit data online. Paper copies of the survey will not be mailed to respondents at the onset of data collection but will be made available upon request. If necessary, respondents will be able to submit data by email, fax, or mail.

The SSV collects data on sexual victimizations occurring in a calendar year. The implementation of the SSV 2023 will begin in September 2024. Data collection, nonresponse follow-up, data quality follow-up will last 7 months and will include a variety of mailings and telephone contacts. A brief description of all steps in the data collection protocol are provided below.

• Pre-notification letter. In early September 2024, a pre-notification letter will be emailed to each central respondent and respondents at selected facilities announcing the impending data collection and describing PREA and the importance of the survey (see Attachment 9). PDF survey forms will be made available on the SSV website for preview at that time.

• Invitation letter. In late September, each respondent will receive a survey invitation letter in the mail (see Attachment 10) announcing the start of the SSV data collection and directing them to the SSV website with login information. The PREA coordinators for states will also be emailed directly the contents of this letter. Respondents are asked to submit their data within 6 to 8 weeks of the survey launch.

• Nonresponse reminder letters and telephone contact. Two weeks prior to the survey due date a reminder email will be sent to respondents who have not yet submitted their data as notification of the impending due date. Personalized nonresponse follow-up via email and phone will be repeated about every two weeks after the submission due date and continue until the data collection closes (see Attachment 101for a sample script for these contacts). Outreach to nonrespondents in the form of emails and phone calls may occur up to 10 times during a given data collection cycle.

• Final reminder. Three weeks prior to the close of data collection, which will occur in May, a letter, requiring receipt signature, will be mailed to non-respondents stressing the importance of the collection and encouraging participation (see Attachment 12).

• Telephone and e-mail data quality follow up. Data review will begin as soon as data are submitted and will continue through data collection closeout. During the review, discrepancies or missing data values on incident forms are discovered, outreach via e-mail or phone will be conducted to clarify responses or obtain missing forms (see Attachment 13 for call specifications for these contacts).

• Final submission page. After respondents have successfully submitted their data, a final landing page thanks them for their submission and provides instructions for viewing their submitted data (see Attachment 14). If the respondent provides data through other means such as over the phone or through email, they are thanked by the program manager or call center representative.

Data Processing Procedure

If a respondent submits data by mail or fax, responses are keyed into the data processing system by a Census Bureau employee. Following the closure of data collection and data review by the Census Bureau, data are reviewed by BJS statisticians. Form submissions are reviewed to ensure the number of substantiated incidents reported on summary forms match the number of incident forms submitted, item responses are within bounds, and text responses are reviewed to ensure all data is contained in item categories.

3. Methods to Maximize Response Rates

Over time, significant effort has been made to make the SSV data collection materials clear and straightforward, in an effort to yield high response rates. The SSV questionnaires have been designed to make collection of the data as concise and easy for the respondents as possible. Uniform definitions of terms and concepts, as well as counting rules for items to be reported are included on the forms. Item response rates for the SSV summary forms is near or at 100%. BJS receives incident-based reports on all substantiated allegations of sexual victimization.

During each iteration of the SSV, the materials are reviewed to determine if further improvement can be made to increase response rates, reduce respondent burden, save cost, and enhance data quality. To address the decline in response rates, BJS has added additional correspondence to the usual respondent contact materials. BJS will continue to encourage online data submission for best data quality but will also provide other modes of data submission (fax, e-mail, or telephone) to accommodate respondents that have limited access to the internet or simply prefer to work with hardcopy survey forms.

Throughout the data collection period, the data collection agent will monitor data submission and track survey para-data, e.g., respondent log-in timestamps, contacts made with each respondent, and comments respondents left on the survey form. The para-data will be used for tailored nonresponse follow-up and data quality follow-up. For these outreach activities, respondents receive emails from the collection team from a designated SSV project email address and are also provided the contact information of the SSV project manager at the Census Bureau. This project manager serves as a consistent “agency liaison” throughout data collection.

Table 4. Anticipated response rates for SSV 2023-2025
Form	Facilities included	Anticipated response rate
SSV-1	BOP	100%
SSV-2	Adult state prisons	100%
SSV-3	Public jails	90.0%
SSV-4	Adult private prisons, private jails, Indian country jails, Military, and ICE facilities	95.0
SSV-5	Juvenile central responders	100%
SSV-6	Juvenile public, private and Indian country facilities	88%

Weighting and nonresponse adjustments

BJS implements nonresponse weight adjustment procedures that account for unit nonresponse to produce national estimates of sexual victimizations. BJS doesn’t perform imputation in cases of item nonresponse on summary or incident forms. Data from BOP and all state prison systems, all state central reporters for juvenile justice, and U.S. military facilities, are given a weight of 1.00 because they were all selected with certainty and all respond to the survey. Additionally, data from all dedicated ICE facilities are given a weight of 1.00 as it is a full enumeration of such facilities.

Public jails, private jails, Indian country jails, private prisons, and local and private juvenile facilities are assigned an initial sampling weight equal to the inverse probability of selection. After the revised sampling designs go into effect during the 2023 reference year data collection, nonresponse adjustment calculations will be the same for all sampled facility types. In each survey year, weights for responding facilities will be adjusted for nonresponse by multiplying initial sample weights by the ratio of the sum of initial weights of active facilities in each stratum to the sum of weights for participating facilities. After applying the nonresponse adjustment, the sum of the final weights in each stratum equals the sum of weights for active facilities in each stratum. The final weights applied to each facility are used to produce national estimates of sexual victimization.

4. Final Testing of Procedures

The US Census Bureau and BJS conducted extensive testing of the SSV as part of the redesign efforts. See Attachment 8 for detailed explanations of the respondents included in each stage of testing, revisions to summary and incident forms, and overall recommendations following testing. Early-stage scoping interviews were conducted with nine in-sample respondents to understand how respondents viewed the format and language of the SSV instruments, their process for reporting incidents, and their views on what improvements should be implemented.

The SSV forms were revised based on respondent feedback. The following changes were made to summary forms:

1. Definitions of sexual victimization were standardized to match the PREA Standards. Inmate-on-inmate nonconsensual sexual acts and abusive sexual contact were consolidated into a new category called “inmate-on-inmate sexual abuse.” Staff-on-inmate sexual misconduct was changed to “staff-on-inmate sexual abuse”.

2. Check boxes were removed for cases where there were “none” or “zero” allegations and instructions were changed to indicate a “0” should be written in these cases.

The incident forms underwent many revisions after early-stage scoping interviews. The overall structure was altered to contain four main sections: basic incident information, victim information, inmate/youth-perpetrator information, staff-perpetrator information. Revisions were also made to specific items, including changes to the wording and layout. See Attachment 8 for detailed information on revisions made to the incident forms at this stage of testing.

Cognitive interviews were conducted with 15 respondents in sample for the SSV. Respondents were first sent a recruitment email, then participating respondents were asked for feedback to understand their comprehension and ability to answer specific questions on the SSV summary and incident forms.

After cognitive interviews were completed, 500 facilities were identified to participate in unmoderated cognitive testing. Respondents were invited to complete a web survey in which they were asked about their response process and comprehension of questions. Respondents were first presented with a consent form that outlined the purpose of the testing, an outline of respondent burden, and the confidentiality authority of the data provided. Respondents gave active consent to participate before providing data. Approximately 38% of invited participants completed the web survey.

A second round of cognitive interviews were conducted with 16 respondents. Respondents were again sent a recruitment email with an invitation to participate. Respondents were asked for feedback on the revised SSV forms, including the text and format of specific items.

Recommendations from all stages of testing were considered and final revisions were made to the summary and incident forms. Definitions of sexual victimization were changed to correspond to PREA definitions and items on the summary and incident forms were revised to improve respondent comprehension and ease of reporting. Incident forms were further revised to include four main sections, which will allow data analysis to link incident outcomes to person-level details and characteristics. For detailed information on the final revisions to summary and incident forms, see Attachment 8, pages 105-114.

4. Contacts for Statistical Aspects of Data Collection

BJS takes responsibility for the overall design and management of the survey, including sampling procedures, development of the questionnaires, and the analysis and publication of the data. The BJS contact is—

Emily Buehler, Statistician

Bureau of Justice Statistics

810 Seventh St., N.W.

Washington, DC 20531

(202) 598-1036

emily.buehler@usdoj.gov

The Economic Reimbursable Surveys Division at the Census Bureau is the collection agent and is responsible for the collection of all data. The Economic Statistical Methods Division is responsible for drawing the samples. The Census Bureau contact is—

Greta Clark, Survey Statistician Criminal Justice Statistics Branch Economic Reimbursable Surveys Div.

U.S. Census Bureau

4600 Silver Hill Road

Washington, DC 20233-6800

(301) 763-2586

[email protected]

4