SILJ 1121-0098 Part B

B. Collection of Information Employing Statistical Methods

1. Universe and Respondent Selection

1. Universe

The universe comprises 734,500 inmates held in 3,116 local jails in the United States that are publicly or privately operated facilities that generally have the authority to hold inmates beyond 72 hours and beyond arraignment. The frame for the SILJ is based on the 2019 Census of Jails. Facilities include county and city jails, detention centers, county or city correctional centers, special jail facilities (such as medical or treatment centers and prerelease centers), and temporary holding or lockup facilities that are part of a facility’s combined function. U.S. Immigration and Customs Enforcement (ICE), tribal, military facilities, and the combined jail and prison systems in Alaska, Connecticut, Delaware, Hawaii, Rhode Island, and Vermont are not included in the COJ. However, 15 locally operated jails in Alaska are included in the COJ. Inmates include persons in custody of jails including persons held for other jurisdictions (e.g., for federal, state, or tribal authorities). The frame incorporates facility-level information on variables such as number of confined inmates by race, sex, and adult or juvenile status, and average daily population.

1.2. Sample design

Sample enhancement

The targeted number of respondents to be selected for the 2024 national study will increase from 465 jails and 7,368 adult inmates in 2002, to 600 jails and 10,000 adult inmates. The larger inmate sample will allow BJS to produce detailed statistics, with more precision, for important subpopulations and will give BJS greater power to detect differences in these subpopulations.

Using the ratio of adult males-to-females from the 2002 SILJ, which was 5,169/2,199 = 2.35, we can allocate 7,000 male inmate interviews and 3,000 female inmate interviews (7,000/3,000 = 2.33). Based on the 7,000 males in sample and assuming that we will have an 80% inmate response rate, for a given estimate of 5% of the male population, we will achieve an approximate coefficient of variation (CV) of 8.2% on the estimate. Likewise for the sample size of 3,000 female inmates, we will achieve approximately a 12.5% CV for an estimate of 5% of the female inmate population having some characteristic. Based on the 2002 sample sizes of 5,169 males in sample and assuming that we will have an 80% inmate response rate, for a given estimate of 5% of the male population, we would have achieved an approximate coefficient of variation (CV) of 9.6 percent on the estimate. Likewise for the sample size of 2,199 female inmates, we would have achieved approximately a 14.6% CV for an estimate of 5% of the female inmate population having some characteristic.

A sample size of 600 jails and 10,000 inmates for the 2024 collection represents an improvement in the CV’s over the 2002 sample size. At present, the current sample size will produce acceptable error and can be supported with the current financial and staffing resources.

Stratified two-stage selection

The sample design is a stratified two-stage selection, in which jails are selected in the first stage and inmates to be interviewed are selected in the second stage. The formation of the strata is based on the number of confined inmates within the jail. Table 1 defines which stratum each jail is placed in.

Table 1. Definition of the stratum for the Survey of Inmates in Local Jails, 2024

Stratum	Female Population		Male Population	Additional Criteria
1	Confined Females >125	or	Confined Males >1500
2	101≤ Confined Females ≤125	or	501≤ Confined Males ≤1500	and not in Stratum 1
3	76≤ Confined Females ≤100	or	201≤ Confined Males ≤500	and not in Stratum 1 or 2
4	26≤ Confined Females ≤75	or	41≤ Confined Males ≤200	and not in Stratum 1, 2 or 3
5	0≤ Confined Females ≤25	or	0≤ Confined Males ≤40	and not in Stratum 1, 2, 3 or 4

First-stage design. BJS expects some facilities to refuse to participate (up to 20%). Therefore, we plan to oversample the number of jails selected with the expectation that 140 jails will not participate. BJS anticipates conducting initial outreach to 740 jails and obtaining final consent with up to 600 jails – the desired sample size of the national study. Using the target sample size of 600 jails, the number of jails within each stratum is presented in Table 2 below. Jails with a large number of male inmates or female inmates or both will be selected with certainty. The jail sample sizes for the remaining strata are proportionally assigned based on inmate population sizes. Jails within each stratum are selected using systematic sampling. Jails are first sorted by stratum code, FIPS state, FIPS county and Facility ID, and then jails are selected through take-every (TEs) nth unit based on the inmate populations within the strata. Each jail within a stratum has an equal probability of selection. The jail TE for a stratum is the number of jails in the stratum divided by the jail sample size of that stratum.

Second-stage design. The second-stage sample will be of inmates aged 18 and older within each sampled facility. Female inmates will be sampled at twice the level to allow for a meaningful analysis by sex. Unlike the 2002 SILJ, juveniles are excluded due to difficulties in receiving consent to interview juveniles, and the significant decline in the juvenile population held in local jails (from approximately 7,240 at midyear 2002 to 1,900 at midyear 2022). A representative sample of juveniles would require a much larger number of jails selected at the first stage, which is cost prohibitive.

In the second stage of the sample, inmates will be systematically sampled by sex. All inmates of a given sex in a sample jail have equal probabilities of selection. The inmate TEs for males within the strata were calculated as the overall inmate male TE divided by the jail TE for that particular stratum. The same is true for the female TE. For example, based on 621,350 male inmates and a sample size of 7,000 male interviews, the overall male inmate TE is 621,350/7,000 = 88.76. The stratum 2 within jail TE = 88.76/1.58 = 56.18 (Note: the jail TE used for this calculation is based on 600 jails in sample).

Table 2. Summary of the Sample Design for the Survey of Inmates in Local Jails, 2024
			Stratum 1	Stratum 2	Stratum 3	Stratum 4	Stratum 5	Totals
Universe: 2019 Census of Jails			205	180	442	1,225	1,064	3,116
Number of adult inmates (2019)
	Males		232,900	117,576	134,625	119,861	16,388	621,350
	Females		44,694	13,793	22,441	25,584	3,726	110,238
Stage 1: Selection of jails
	Jail sampling rate (TEs)		1	1.2	2.4	7.2	44.3
	Number of jails in sample		205	155	185	171	24	740
	Number of jails expected to participate		190	129	143	125	13	600
	Response rate		93%	83%	77%	73%	54%	81%
Stage 2: Selection of adult inmates
	Sampling rates (1 out of…)
		Males	88.8	56.2	27.3	9.2	1.5
		Females	36.3	23.0	11.2	3.8	1.0
	Number of inmates in sample
		Males	2,624	1,802	2,063	1,821	245	8,555
		Females	1,232	517	841	951	84	3,625
	Expected workload based on expected jail nonresponse*
		Males	2,432	1,504	1,598	1,329	132	6,995
		Females	1,142	432	652	694	45	2,964
	Expected response rates
		Males	93%	83%	77%	73%	54%	80%
		Females	93%	84%	78%	73%	54%	80%
*The expected number of interviewed inmates are less than the 7,000 for males and 3,000 for females due to nonresponse of the self-representing jails in stratum 1.



2. Procedures for Information Collection

The methods proposed for use in data collection are as follows:

1. Jail recruitment

In July 2023, the agency head (i.e., Sheriff or Jail Administrator) of each sampled facility will receive a written letter from the BJS Acting Director that will inform the facility of their selection to participate in the SILJ (see Attachment 10). Sampled facilities will also receive an informative flyer on why BJS is conducting the survey, who participates, when the study will begin, and how the data will be used (see Attachment 11). A week after mailing the letter, the Census Bureau, BJS’s data collection agent, will email the agency head to request access to a liaison from their office for the survey (see Attachment 12). The Census Bureau will work with the liaison to coordinate logistics for the data collection. If needed, the Census Bureau will send a reminder email to the agency head (see Attachment 13) and a final reminder email to the facility's point of contact for the Annual Survey of Jails, Census of Jails, or the National Inmate Survey (see Attachment 14), seeking approval to conduct inmate interviews in their facility. A nonresponse telephone call (see Attachment 15) will be made if no contact is established with the POC listed in the final email reminder. If the sampled facility refuses to participate, the BJS Jails and Community Corrections Statistics Unit Chief will seek to overturn refusal through email (see Attachment 16) or phone call if necessary (language from Attachment 16 will be used to facilitate the telephone conversation).

Once the jail confirms its willingness and ability to participate in the SILJ, the Census Bureau team will coordinate with the jail’s staff to schedule the field visits for the full data collection, based on the jail’s availability and staffing. Through email (see Attachment 17) and a phone call if needed, the Census Bureau will work directly with the liaison/POC at each sampled facility to finalize details for data collection, including the clearance process for conducting interviews, COVID-19 protocols, identifying appropriate space for interviewing, the need for bilingual interviewers, format of the roster which will be used to draw the sample of inmates, number of days and hours of each day when interviewing can be conducted, other interview protocols, and rules regarding items that may be brought into the jails, and instructions for arriving at the facility.

2. Sampling of Inmates

On the first day of data collection, the facility will provide a roster of all arraigned inmates age 18 and older (who are currently incarcerated there). A Lister field representative will work with the jail point of contact to verify that the listing includes only those inmates who are eligible for the survey based on the following criteria and re-number the inmates on the list as appropriate. The following inmates are not to be included:

• Inmates under age 18

• Inmates released from facility the prior night

• Inmates scheduled to be admitted but not officially at the facility

• Inmates under the jail’s jurisdiction but housed in another jail

• Inmates who had escaped or were absent without leave

• Inmates under jail supervision but not confined

The Lister will enter the current population of the jail into the automated listing instrument. The listing instrument will use predetermined values for the sampling interval within each jail to determine which line numbers from the listing are sampled to achieve the pre-determined expected number of sample cases. The Lister will then enter the names and inmate numbers associated with those line numbers into the Listing instrument. The selected sample of inmates will be transmitted into the data collection case management system and re-transmitted back to the laptops of each member of the team of interviewers assigned to the particular facility. The inmate sampling intervals are predetermined by the stratum the jail is in.

In order to determine if any bias is introduced due to inmate nonresponse, facilities will be asked to provide administrative record data for all inmates on the roster to compare demographic characteristics (i.e., date of birth, sex, race/ethnicity, admission date, offense type, conviction status, and sentence length if convicted) of responding inmates with those who do not participate if necessary. See Section 3 Methods to Maximize Response – Nonresponse Adjustments below for more information on the nonresponse bias analysis.

3. Data Collection

A team of interviewers from Census Bureau will visit the facility. They will ask correctional officers to bring each sampled inmate to a private interviewing area. Prior to the onset of the interview, interviewers will read the informed consent document to the inmate (see Attachment 8). If the inmate initially refuses, the interviewer will be trained to address any potential concerns to enhance participation. If the inmate is still hesitant to participate and expresses concerns about aspects of the records linkage, they will be informed that they can opt out of BJS reviewing their future criminal history records to conduct a recidivism study and linkage to other federal administrative data but can still participate in the survey. The interviewer will code the inmate’s decision in the case management system (CMS). If the inmate consents, the interviewer will administer the interview by asking questions and recording the inmate’s responses using a laptop. Inmates not interested in participating will be returned to their cell and a non-respondent worksheet (see Attachment 18) will be completed by the lead interviewer to collect information on inmate sex, date of birth, most serious offense, and status of inmate within the criminal justice process, as well as reason for refusal or other non-participation, from designated facility staff.

3. Methods to Maximize Response

Administration

Minimize burden on inmates. Every effort has been made to minimize the burden of the 2024 SILJ administration on inmates and high response rates is of great importance to BJS. The automated interview has been designed to make the interview concise, simple to administer, and easy for the inmates to respond. Computer-Assisted Personal Interviews (CAPI) also records the inmate’s responses quickly and maximizes the flow of the interview with the preprogramed skip sequences. Additionally, BJS cut a third (or 246 questions) of the questions after pretest to reduce the overall length for consent and the interview.

Whether interacting with facility staff or inmates, interviewers are trained to be courteous and professional in their behavior. They will adhere to the prescribed facility protocol and avoid engaging in extraneous conversations with inmates that could lengthen the interview unnecessarily. Interviewers will learn at training that an inmate’s movements and activities at the facility are typically structured with little room for deviations. The interviewers will work efficiently to ensure inmates are not delayed in getting to meals, counts, etc.

Protocols to maximize inmate response. Field interviewers will be trained on refusal avoidance to maximize response to the survey. During the consent process, they will tell inmates that their data is important to understanding the experiences of inmates. They will inform inmates that participation is voluntary and that any information that might identify them is confidential. Every effort is being made to make the survey materials clear and simple to use. The confidential nature of the data collected is clearly explained in the consent process and followed by several questions to verify respondent comprehension.

During training, interviewers will gain experience responding to common questions likely to be raised by inmates. During mock interviews, they will have an opportunity to practice answering questions, and addressing general objections so they become comfortable with the information. Interviewers will encourage hesitant inmates to start the interview and see how it goes. Inmates who are hesitant to allow BJS to link their responses with administrative records will be offered the option to opt out of this step. Interviewers will be trained to minimize the number of inmates who refuse to participate in the full study and the number who opt out of records linkage by addressing inmate concerns about this process and its implications.

Interviewers will also work with facility staff to arrange for inmates who must stop the interview before reaching the end (perhaps because they must go to court, to a meal, to a job, or return to their housing unit for a count) to come back at a later time to complete the interview. Interviewers will work closely with their facility contact to handle these restarted interviews efficiently. Interviewers are also encouraged to inquire about inmates who are scheduled for immediate release so that they can be interviewed first, thereby avoiding noninterviews.

A Spanish version of the CAPI questionnaire will be available for Spanish-speaking respondents. Interviewers will be available to answer any questions that respondents may have, including bilingual staff who can answer questions in Spanish. Arrangements with mental health staff at each facility can be made for any respondents interested in obtaining counseling services or assistance following the survey.

Protocols to minimize burden on facilities. The Census Bureau has extensive experience effectively working with a variety of jails through conducting the SILJ for BJS since 1972. Protocols have been established to minimize burden on facilities as much as possible, including a customized data collection schedule and minimizing the number of days in the facility to conduct data collection. The protocols allow for flexibility given that facilities are expected to vary in terms of interviewing space, number of days and hours of each day when interviewing can be conducted, specific rules regarding items that may be brought into the jail, and instructions for arriving at the facility. The number of days spent at each facility will vary primarily based on the number of expected sampled inmates, the available field staff and the number of interview rooms available at the jail. BJS expects to spend approximately 3-5 days in any facility.

Protocols to maximize facility response. BJS is providing challenge coins to correctional staff to show our gratitude for participating in the data collections. BJS will also utilize jail related conferences to promote SILJ awareness and secure participation.

Promoting Awareness of the SILJ

BJS has been and will continue to market the 2024 SILJ to garner support and generate interest among stakeholders to maximize response.

Expert review of questionnaire. BJS hosted the First Annual Jail Research Network (JRN) to convene a technical review panel of the SILJ questionnaire. The experts evaluated and helped shape the 2024 SILJ. Topics for review and feedback from the panel included: current offense and detention status, personal characteristics, pleas and bail, aggravating factors, mental and physical health, and socioeconomic status, among others.

Collaborative Promotion. BJS plans to promote the awareness of the upcoming SILJ through the American Jail Association (AJA) and National Sheriffs' Association (NSA). The goal is to ensure broad and informed participation in the upcoming SILJ, by utilizing the AJA's and NSA’s platforms and outreach channels.

Presentations. The SILJ Project Manager presents information on past and future SILJ reports and initiatives at the American Jail Association’s Conference in Jail Expo, National Sheriffs Winter Conference, and the Large Jails Network Meeting (hosted by the National Institute of Corrections).

SILJ Informative Flyer. To jails announcing survey and disseminating flyer at

conferences.

Dedicated BJS Website for 2024 SILJ. https://bjs.ojp.gov/survey-inmates-local-jails-silj-2024-2025.

Nonresponse Adjustments

With almost any survey, some of the selected subjects will not respond to the survey request (i.e., facility and inmate unit nonresponse) and some will not respond to particular questions (i.e., inmate item nonresponse). Weighting will be used to adjust for unit nonresponse in the 2024 SILJ. Created weights will allow for the analysis of the cross-section sample of inmates. In the event of unit nonresponse in the first stage (facility), a ratio adjustment will be applied to the stratum weights to account for the nonresponding jails. For inmate nonresponse at the unit level in the second stage, the nonresponse adjusted weights will be calculated within each stratum using administrative data from the sampling rosters. The weights will be adjusted using calibration software, which adjusts the sum of the respondent's weights to account for the weights of the nonresponding inmates.

For unit nonresponse at the inmate stage, BJS will assess whether responding inmates are different from nonresponding inmates regardless of response rate. We will examine the disposition codes among nonrespondents to determine if a particular type of inmate had significantly higher levels of nonresponse. Then, using the inmate characteristics we receive on the facility roster (i.e., age, sex, race and ethnicity, and conviction status), BJS will compare the distribution of respondents and nonrespondents by inmate characteristics. BJS will assess the level of potential nonresponse bias both before and after nonresponse adjustment and carry out statistical tests of weighted respondent vs. sample distributions to confirm the absence of a difference in the characteristics evaluated.

For inmate item nonresponse, imputation is preferred, as it preserves a single record per case, allowing for multivariable analysis. For SILJ, BJS will use hot-deck imputation after modeling the missingness distributions. This approach works well for large and diverse variable sets while providing a flexible toolset for controlling the underlying missingness mechanisms.

Post-Collection Outreach

After collection, a thank you letter, will be sent to thank the facilities for participating (Attachment 19).

4. Test of Procedures or Methods

Through cognitive interviews, BJS received helpful feedback on wording and flow, and estimated burden. From April to June 2023, BJS pretested the CAPI instrument in 5 jails in Florida and Texas with a sample of 169 jail inmates. As a result, a minimal number of skip patterns were revised, and one-third of the questions were not incorporated into the 2024 national study due to the substantial burden it would pose to respondents. See section 8 in Part A for information on consultation with persons outside the agency and section 15 for detailed information on changes to the survey instrument.

Cognitive Testing of Questionnaire Content. The 2024 SILJ questionnaire was designed in collaboration with Abt Associates (Award number 2015-R2-CX-K146 Survey of Inmates in Local Jails: Design and Testing), which also included a pretest study to evaluate the questionnaire. Prior to fielding the pretest, cognitive interviews were conducted through the design and testing project. Paramount to the successful development of the survey is ensuring that the new and enhanced questions themselves are understood by respondents and thus able to accurately capture data on the experiences of inmates during and before incarceration.

To evaluate the degree to which questions are understood by respondents and to inform enhancements to the survey, Abt Associates on behalf of BJS, cognitively tested questions with nine inmates in the Middlesex Jail & House of Correction in North Billerica, MA. Inmates received a notice form jail staff regarding the study and our interest in recruiting inmates to participate in cognitive interviews. Selection was based on a convenience sample. Abt recruited 9 adult male and female inmates, administered the consent and select questions (new and revised) from the survey to them in paper form, and conducted follow-up interviews with inmates to discuss their experience completing the survey questions.

The semi-structured interview included an introductory script that interviewers read verbatim to ensure that participants have a consistent understanding of the cognitive interview process. The interview guide also contained a list of probes that interviewers used during the cognitive interview.

Based on the cognitive test, minor changes were made to the wording of some questions to emphasize the intent of the question (e.g., highlighting of certain words to ensure emphasis during administration); definitions were added to clarify the meaning of a few response options; additional response options were added based on feedback; and 23 questions were cut from the questionnaire.

2023 SILJ Pretest. The main goals of 2023 SILJ pretest were to evaluate the procedures to recruit jails to participate in the survey, evaluate the SILJ questionnaire and functionality CAPI instrument, and measure the interview length to estimate respondent burden.

Abt Associates conducted a pretest of the full English and Spanish survey instrument using CAPI. The pretest helped BJS measure the average timing of the survey and to identify potential issues with question wording of the informed consent script and the survey instrument. The CAPI instrument was pretested successfully with a sample of 169 inmates in five local jails. The goal of the SILJ pretest was to complete pretest interviews in five preselected jail facilities in Texas and Florida from a total of at least 36 adult inmates (female and male) at each jail. Staff from BJS and Census Bureau were also on site on some of the interview days to observe the interviews and to document any problems with the survey instrument or the process.

The jail selection plan involved the organization of jails into geographic clusters—using Metropolitan Statistical Areas (MSAs)—and then by Average Daily Population (ADP) of the jail. The jail selection process was conducted in a data-driven manner to provide an opportunity to sample and interview both male and female inmates, English and Spanish-speaking inmates, and some that are pre- or post-arraignment across a range of medium- to larger-sized jails. To select jails to participate in the pretest, jails were first organized into geographic clusters by MSA, and then sorted by ADP based on the 2019 Census of Jails. We excluded jails with ADPs below 200 inmates, which are too small to support the target number of pretest interviews (N=36), and jails with ADPs above 1,500 inmates, because these jails will be automatically selected to participate in the full administration during national implementation.

Within jails that met the above size criteria, data on the percentage of female inmates, Hispanic inmates, and those awaiting trial/arraignment status were then used to identify geographic clusters of jails that would allow the team to sample and interview a diverse group of male and female inmates, English- and Spanish-speaking inmates, and inmates that are pre- or post-arraignment across a range of medium- to larger-sized jails. The COVID-19 pandemic resulted in delays to the administration of the SILJ pretest, which led to the project team finding ways to increase efficiencies in data collection costs for the pretest while supporting recruitment from among interviewer candidates living in two different regions of the country. The Abt project team, in collaboration with BJS, ultimately decided to select jails located in only two states that included jails with high proportions of Hispanic inmates, to best enable inclusion of Spanish-speaking respondents. The team therefore chose Texas and Florida as the two pretest states. Texas has numerous MSAs including jails with large Hispanic populations, and there were several Florida MSAs including jails with relatively large Hispanic populations (around 15% or higher).

Finally, as previously mentioned, the pretest schedule was significantly delayed because of COVID-19 and as a result coincided with the BJS National Inmate Survey (NIS). Therefore, the project team excluded any jails that were known to already be participating in the NIS. After this exclusion, a total of 10 jails throughout Texas and Florida were left in the sample for potential outreach to participate in the pretest effort.

Since the researchers assumed that not all jails would agree to participate, jail recruitment for the SILJ pretest began by reaching out to the 10 selected jails to identify their capabilities, operational feasibility to conduct CAPI, and willingness to participate. The BJS Director initially sent a letter to the sheriff of each selected jail, providing an overview of the project, describing data collection procedures, and asking for permission to conduct the pretest at the jail. BJS staff followed up with nonresponsive jails, and Abt then directly emailed seven jails that had agreed to participate, to schedule a recruitment call. During these recruitment calls, the Abt team gathered further information about each jail, regarding background check procedures for visitors, internet access in the jail, COVID-19 policies, and additional contact information. Based on the seven calls, the Abt team selected five jails that met all data collection criteria (e.g., availability of Spanish-speakers, both male and female inmates, and the capacity to accommodate at least three field staff for a week).

During these calls, the Abt team also reviewed all SILJ data collection procedures with the jail staff to make sure they fully understood what to expect, including that the Abt field team would need a copy of the jail’s roster of inmates housed at the facility at the time of data collection. The roster request specified that at a minimum the roster would need to include inmate name, sex, race and conviction status. The Abt team also asked various questions regarding possible interview locations for the interviewers at the jail, the availability of security escorts, whether field interviewers (FIs) would need to be the same sex as the inmates they interviewed, and the number of field staff the jail would be able to accommodate.

Once the jail had confirmed its ability to participate in the SILJ pretest, the Abt team coordinated with the jail’s staff to schedule the five-day field visit for data collection. Visits were scheduled based on the jail’s availability and staffing. The team scheduled an additional logistics call approximately two weeks before the field visit to discuss the protocol in more detail, including where the interviewers would be working, any prohibited items, the availability of Wi-Fi and outlets, and the shift hours that worked best for the jails. The Abt team also reiterated the importance of receiving a roster of inmates housed at the jail on the first day of data collection. After the call, the Abt team followed up with the jail contact via email to provide the specifics of what and whom the roster should include.

For jails that required security background applications or copies of drivers’ licenses to be provided in advance, Abt coordinated with the jail contact to set up a secure file transfer and submitted all materials by the date each jail had requested. The Abt team kept detailed checklists, emails, and call agendas that were drafted in advance to ensure consistent and thorough communications with all jails. In addition, the Abt team submitted all security clearance paperwork within the timeframe the jail had requested, to ensure a smooth check-in process upon the field team’s arrival.

For each visit, on the first day of data collection at each site, the Lead Interviewer (LI) obtained from the facility a list of adult (male and female) inmates who had had a bed assigned to them the previous night (i.e., slept at the jail the previous night or arrived that morning). The facility provided the roster in paper form and the LI used it to confirm the list of eligible inmates. At a minimum, the list included inmate name, facility ID number, and sex. The LI reviewed the list with jail staff to ensure the following inmates were not included:

• Inmates under age 18

• Inmates released from facility the prior night

• Inmates scheduled to be admitted but not officially at the facility

• Inmates under the jail’s jurisdiction but housed in another jail

• Inmates who had escaped or were absent without leave

• Inmates under jail supervision but not confined

The goal at each jail site was to conduct interviews with approximately 36 adult (male and female, pre- and post-conviction) inmates. The LI tracked selected inmates directly on the paper roster. The LI only used the roster for inmate selection and kept all other notes about interview results in a spreadsheet on her tablet computer. The roster was never taken outside of the facility. At the end of each workday, the LI secured the roster in a sealed envelope and gave it back to a staff member at the jail to be stored in a locked location. On the last day of the data collection week, the roster and any other paper materials were shredded at the facility.

Although the goal at each jail was to interview 36 inmates, the LI initially selected 50 inmates to allow for the possibility that not all selected inmates would be available and/or willing to participate. For purposes of random selection, every Nth inmate was selected for an interview (where N= the total roster number divided by 50—the number of inmates to be selected). If the initial 50 selected inmates were not sufficient to obtain the target number of interviews at that facility, the LI worked with the project team to determine how many additional inmates to randomly select from the roster.

The 2023 SILJ Pretest also allowed us to assess the SILJ instrument based on answers to the following questions.

• How long did the interviews take (average time, minimum and maximum times, etc.)? Instrument length was reviewed by section and in total to inform decisions about the addition or deletion of content and the length of the data collection period within each facility, and to accurately estimate respondent burden for the 2024 national study.

• Were there any survey items with unusually long administration times? Such items were reviewed to determine whether they needed to be clarified to reduce confusion or were creating excessive burden on respondents and therefore should be omitted from the instrument, especially in light of the length of the survey.

• Were there survey items with high rates of nonresponse? Item nonresponse (i.e., item-level assessment of “don’t know” and “refused” responses to survey questions) is an indicator of potential instrument or data quality problems. These types of items were candidates for deletion to reduce the length of the survey.

• How did inmates respond to a BJS’s interest in obtaining consent to link to respondents’ RAP sheets to collect data on current and/or future criminal history and other federal administrative data.

The interview length was the most serious challenge to minimize burden on inmates and facilities. Approximately 17% of the sampled inmates refused to participate in the study after hearing how long, on average, the interview was expected to take. The average duration of the SILJ pretest was 98.6 minutes, (7.2 minutes for the consent process and 91.4 minutes for the main survey interview). The median time was 94 minutes.

The findings from the pretest informed decisions to refine, change, and edit the questionnaire. The changes made since the pretest include reducing the overall questionnaire length to maximize survey response and minimize burden, refining the questionnaire items to reduce respondent burden and improve data quality, and improving the consent forms to streamline the consent process and maximize participation.

After this test, BJS made additional revisions to the questionnaire to reduce burden. The informed consent was also revised based on SPI consent, which was less than 3 minutes on average. In February 2024, BJS conducted internal testing and mock interviews to retest the timing of the revised instrument. The average duration of the revised SILJ is approximately 73 minutes, including the consent process.

5. Contact Information

The Jails and Community Corrections Statistics Unit of BJS takes responsibility for the overall design and management of the activities described in this submission, including data collection procedures, development of the questionnaire, and analysis of the data.

BJS contact for the Survey of Inmates in Local Jails:

Todd Minton

Unit Chief

Jails and Community Corrections Statistics Unit

Bureau of Justice Statistics

810 Seventh Street, N.W.

Washington, DC 20531

(202) 598-7226

[email protected]

21