SPI Pilot Study Memorandum to OMB

MEMORANDUM

MEMORANDUM TO: Shelly Wilkie Martinez

Office of Statistical and Science Policy

Office of Management and Budget

THROUGH: Lynn Murray

Clearance Officer

Justice Management Division

William J. Sabol, Ph.D.

Acting Director

Bureau of Justice Statistics

FROM: Lauren E. Glaze

Statistician and SPI Project Manager

Bureau of Justice Statistics

DATE: April 15, 2013

SUBJECT: BJS Request for OMB Clearance for Pilot Study of Survey of Prison Inmates under the Survey of Prison Inmates: Design and Testing Project through the generic clearance agreement OMB Number 1121-0339

Introduction

The Bureau of Justice Statistics (BJS) is planning to conduct a national survey of prisoners within the United States through a project called the Survey of Prison Inmates (SPI) 2014. The purposes of this omnibus survey are to generate reliable, nationally-representative estimates of the characteristics of prisoners in the United States, track changes in the characteristics of prisoners over time, conduct studies of prisoners on special topics, and identify policy-relevant changes in the prison population. The survey will also be used to produce subnational estimates of prisoners within jurisdictions that have the largest prison populations (i.e., 50,000 or more) in the nation.

The 2014 SPI survey builds upon prior surveys and is organized around the concepts of harm, risk, and reentry. Domains related to specific offense conduct and criminal history are designed to measure harm and risk; domains related to pro-social connections are designed to capture dynamic risk elements, consistent with fourth generation risk instruments; domains related to inmates’ substance abuse, mental health, and medical problems as well as the programs in which they participated in prison are designed to address reentry issues (see Attachment A). A major enhancement of the 2014 SPI is the effort to link inmate survey data with administrative records, specifically to criminal history records (also known as the records of arrest and prosecution, or RAP sheets), Social Security Administration (SSA) data on benefits, and the basic department of corrections offender information system data for sampling, weighting, and nonresponse adjustments. (See Request for developmental work for additional information.)

Under Title 42, United States Code, Section 3732, BJS is directed to collect and analyze statistical information concerning the operation of the criminal justice system at the federal, state and local levels. To fulfill part of this mission, BJS has periodically fielded national omnibus surveys of prisoners since the 1970s.¹ The 2014 implementation of SPI will be the seventh national study of its kind that BJS has fielded. The scope of information collected solely through SPI, and the level of detail for some topics, is not available from any other single data source and the data are used by a diverse group of stakeholders to fulfill a variety of purposes. The continuing debate on the size of the prison population and crowding, the associated financial and other societal costs of incarceration, and alternative sanctions are enlightened through comparable and reliable data on prisoners provided by SPI and provide state and local administrators, officials, and policymakers national benchmarks with which to compare their prison populations. The U.S. Congress has cited estimates from the SPI to justify legislation to improve the management of offenders, services provided, and outcomes of offenders, specifically through the Mentally Ill Offender Treatment and Crime Reduction Act (MIOTCRA) of 2004 (P.L. 108-4140) and the Second Chance Act (P.L. 110-199). Estimates on female prisoners from SPI were used by the White House Council on Women and Girls in the report Women in America: Indicators of Social and Economic Well-Being. The Bureau of Justice Assistance and the National Institute of Justice have used the SPI data to justify the allocation of funding for research and grant programs as a result of MIOTCRA and the Second Chance Act. Other federal agencies, such as the Centers for Disease Control and Prevention and the National Institute of Mental Health, have relied on the SPI data to better understand the health status and service needs of the entire United States population, including inmates, and the differences between the inmate and general populations. The Centers for Medicare and Medicaid Services have stressed the need for the SPI data to better understand the impact and costs of healthcare reform as it relates to the future eligibility of inmates returning to the community and the potential eligibility of inmates while still incarcerated. Independent researchers have analyzed the SPI data to address a variety of topics, such as the impact of the prison boom on the characteristics of the prison population, changes in inequality in imprisonment, and the impact of incarceration on children, families, and labor markets.²³⁴⁵⁶

BJS has used the data from previous iterations of SPI to provide a profile of prisoner characteristics and to publish special reports on important substantive issues. For example, BJS has released reports that provide an understanding of the prevalence of physical and behavorial health problems among prisoners and their service needs and utilization, the extent and types of firearm use among prisoners in the commission of a crime and acquisition of those firearms, the educational background of prisoners and the variation in attainment across different prisoner characteristics, the rate at which parents are incarcerated and the number of children affected by parental incarceration, and the military and criminal backgrounds of incarcerated veterans as well as their health conditions. Specific reports that BJS has released on these topics include but are not limited to: Parents in Prison and Their Minor Children, Medical Problems of Prisoners, HIV in Prisons, Veterans in State and Federal Prisons, Lifetime Likelihood of Going to State or Federal Prison, Mental Health Problems of Prison and Jail Inmates, and Drug Use and Dependence among State and Federal Prisoners. The issues addressed in these reports: 1) describe the unique challenges correctional agencies face in managing these special prisoner populations and maintaining the safety of the inmates and prison staff; 2) illustrate the importance of understanding, accounting for, and reducing the costs associated with addressing the needs of these prisoner populations especially in light of current fiscal constraints that many states are facing; and 3) address the potential risks these populations pose to public safety upon returning to the community and the need to identify and implement ways to mitigate risk to improve outcomes.

Prior to 2007, SPI was the only national survey of prisoners conducted by BJS. However, the Prison Rape Elimination Act of 2003 (PREA or the Act; P.L. 108-179) resulted in BJS fielding the National Inmate Survey (NIS) in 2007, 2008-2009, and 2011-2012. The goals and methods of the NIS are different than those of the SPI, most notably, the NIS was designed to produce facility-level estimates so the sample sizes were much larger (average of 85,000 per wave) than SPI (average of 18,000) and the instrument was shorter in length because the content was limited to focus primarily on sexual victimization. Given the different goals and methods of the NIS compared to the SPI, the NIS is not the appropriate vehicle for collecting the variety of data covered by the domains in SPI and cannot be used to inform all plans related to the 2014 SPI national implementation.

The 2014 SPI national implementation will build upon the BJS’s Survey of Prison Inmates: Design and Testing Project which is currently underway with RTI International (RTI) serving as the data collection agent, under a cooperative agreement (Award 2009-BJ-CX-K054). The main goals of the SPI design and testing project are to enhance the instrument to measure new constructs and improve the measurement of existing constructs, develop a sample design that will generate reliable national and subnational statistics, explore the feasibility of using administrative records to supplement survey data and thereby avoid increasing respondent burden, and conduct a pilot study of the survey instrument and protocols. The findings from the pilot study will be used to revise and enhance the plans for the national implementation of the 2014 SPI.

Request for developmental work

The BJS plans to conduct developmental work for the national implementation of the 2014 SPI through the Survey of Prison Inmates: Design and Testing Project under the generic clearance agreement (OMB Number 1121-0339). Specifically through the SPI design and testing project, BJS is requesting clearance for a pilot study of the SPI questionnaire, sampling procedures, and interviewing and other logistical procedures. The goal of the SPI Pilot Study is to fully test all aspects of the redesigned survey to determine how well it performs in the correctional environment in advance of fielding the main study in 2014. All information gathered from the pilot study will be integrated into the full information clearance package for the 2014 SPI national study that will be submitted to OMB in the fall of 2013.

The SPI Pilot Study will involve interviewing a random sample of 480 state and federal prisoners across 6 prisons (i.e., 4 state prisons and 2 federal prisons). The target sample size of inmates per facility (80) is based on the sample design planned for the national implementation of the 2014 SPI. For the pilot study, facilities will be selected purposively to represent the diversity that exists in the size, gender of inmates housed, and security level of prisons in the United States and will be located on the east coast to minimize data collection costs and increase efficiencies (e.g., using a small number of interviewer teams). Facilities will be chosen collaboratively with selected departments of correction and the federal Bureau of Prisons (BOP) with which BJS and RTI have strong working relationships. BJS will contact facilities directly to gain cooperation (see Design of the Pilot Study: Sample Design for more information). While no two prisons are exactly the same, we believe the plans to pilot study in 6 prisons and implementing the second-stage sampling procedures planned for the national study, along with our experience interacting with and collecting data within prisons in previous iterations of the SPI and the NIS, will provide the information required to identify any necessary changes to our survey procedures prior to fielding the 2014 SPI.

The pilot study will also be used to further explore the feasibility of using administrative records to supplement the inmate survey data to provide additional information on the pre-prison experiences of inmates without increasing respondent burden. As part of this effort, BJS plans to link the survey data with RAP sheets of inmates to learn more about their criminal histories other than the information collected through the questionnaire. The unique, fingerprint-supported State Identification (SID) Number or the Federal Bureau of Investigation’s (FBI) unique fingerprint identification number will be obtained from the prisons to conduct the linking through another BJS project currently underway to develop a software system to obtain and standardized RAP sheet information from all 50 states and the federal system.⁷ (See Data Collection Procedures for more information about the data that will be collected from the prisons.) In addition, inmates will be asked to provide their Social Security Numbers (SSNs) so their survey data can be linked to pre-prison beneficiary records from the SSA. The primary goals of both of these efforts through the pilot study are to establish and implement the procedures to obtain the administrative records, analyze the data to determine the utility of producing reliable national statistics on quantities of interest, and develop analytic plans for the national study; the SSA effort will also be used to determine what percent of respondents provide their SSN. (See Data Collection Procedures, Analysis Plan, and Informed Consent, Data Confidentiality and Data Security for additional information.)

Key questions concerning the operational/logistical components to be assessed from the pilot study include the following:

• Were our processes for contacting facilities, planning and scheduling data collection, facilitating background checks of interviewers, and sampling inmates within facilities sufficiently efficient and effective? And, what feedback did facilities give to increase efficiency and/or reduce burden?

• Will our current plans related to the operational/logistical components of SPI enable us to conduct the national data collection during 2014 successfully and within budget? If not, what changes are necessary or might help?

The SPI Pilot Study will allow us to assess the effectiveness and efficiency of our data collection procedures which have been modeled on previous iterations of the SPI and the NIS procedures and adapted where necessary. Specifically, the data will allow us to answer the following key questions, overall and for each type of facility:

• How many cases were we able to complete during the scheduled data collection period?

• What was the overall response rate?

• Are there any indications of nonresponse bias?

• Was the staffing plan adequate? That is, were we able to meet with each sampled inmate within the time period agreed to by the prison? If not, what changes would have been needed and would those changes have been possible given the constraints of the facility?

• What common questions or concerns were raised by respondents? Were interviewers sufficiently prepared to answer them? If not, are revisions needed to the content of our interviewer training program?

The SPI instrument will be evaluated based on answers to the following questions:

• How long did the interviews take (average time, minimum and maximum times, etc.)? Instrument length will be reviewed by sub-section and in total to inform decisions about the addition or deletion of content, the length of the data collection period within each facility, and to accurately estimate respondent burden for the 2014 SPI.

• Were there any survey items with unusually long administration times? Such items will be reviewed to determine whether they need to be clarified to reduce confusion or are 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) will be an indicator of potential instrument or data quality problems. These types of items will be candidates for deletion to reduce the length of the survey.

• Are there any unusual patterns in the data and will these have implications for data quality in the 2014 SPI? Having a preliminary understanding of data patterns will enable anticipation of data outliers for data quality monitoring during the 2014 SPI.

• Did the instrument work as intended, in a live data collection situation? And, what, if any, limitations emerged? The pilot study will be the best opportunity to fully test the complex navigation and capabilities of the Computer-Assisted Personal Interviewing (CAPI) instrument in a real-world situation prior to the fielding the national study.

RTI’s Institutional Review Board has recently reviewed the questionnaire, consent forms, and pilot study plans and has granted approval for this human subjects’ research. We are currently waiting to receive the final documentation of approval.

Cognitive Testing

Two rounds of cognitive interviewing have already been completed with the redesigned questionnaire. Each round of testing was conducted with individuals who had recently been released from prison. As these individuals were no longer incarcerated, the cognitive interviewer asked each participant to answer the questions as though he/she was still incarcerated in the facility where he/she had most recently been held. The first round of cognitive interviewing included a selection of questions from 9 of the 11 questionnaire modules. A total of 9 participants were included in Round 1. The second round of cognitive interviewing included a selection of questions from the remaining 2 modules of the questionnaire and again 9 participants were interviewed. All interviews were conducted in Durham, NC. Based on the two rounds of cognitive interviewing, revisions were made to the questionnaire to reduce ambiguity, facilitate recall, and/or reduce respondent burden.

Use of Incentives in the Pilot Study

The plan is to offer a small incentive to each inmate who participates in the SPI Pilot Study to maximize response to the survey request. Based on BJS’s and RTI’s experience conducting the NIS, each participant will be given a small bag of Chips-Ahoy 100 calorie snack cookies. The inmate will not be allowed to leave the interview location with the cookies but rather will eat them during the interview. The interviewer will collect all trash and dispose of it according to facility procedures. This small incentive increased response rates by approximately 5 percent in the NIS.

Design of the Pilot Study

Sample Design: For the SPI Pilot Study, the 6 prisons will be selected purposively as we do not need to test the facility sampling component (i.e., first stage of sample design) of the redesigned SPI.⁸ The 6 prisons will be selected to include both state and federal BOP facilities, both male and female facilities, and facilities of varying security levels to represent the variation that exists in prisons across the United States. The experience of BJS and RTI conducting three rounds of the NIS have provided a wealth of knowledge regarding how to effectively work with a variety of prisons to schedule and conduct data collection. While the NIS and SPI serve different purposes, there are many similarities related to the logistical/operational components of the two, including the need to identify private interviewing space at each facility, provide information on the interviewers that allows the facility to conduct background checks on the team in advance of data collection, and determining how best to manage the flow of inmates to and from the interviewing location. The logistical/operational plans to conduct the SPI Pilot Study have been modeled after the NIS and adapted where necessary. Although no two prisons are exactly the same, we believe these plans along with implementing the second-stage sampling procedures planned for the 2014 SPI and our experience interacting with and collecting data within a variety of prisons in the NIS (and previous iterations of the SPI) will provide the information required to identify any necessary changes to our operational, sampling, and data collection procedures prior to fielding the 2014 SPI. The pilot study for the NIS included a purposive sample of 8 facilities (prisons and jails) and provided more than a sufficient environment for testing the NIS procedures. Given the similarities described between the NIS and the SPI, we believe the 6 facilities planned for the SPI Pilot Study will be equally adequate.

Facilities will be chosen collaboratively with selected departments of correction and the BOP with which BJS and RTI have strong working relationships. BJS will first contact the commissioners of the departments of correction to gain cooperation and then will directly contact the facilities to notify them of their selection and that RTI’s logistics manager will be in contact to begin planning for the pilot study (see Attachments B and C). Pilot facilities will be located on the east coast of the United States to minimize data collection costs and increase efficiencies (e.g., using a small number of interviewer teams).

Sampling of prisoners (i.e., second stage of sample design). RTI staff will work with each of the 6 prisons to obtain a complete roster of the inmates housed in the facility. Based on experience with the NIS, RTI will request that the roster be faxed or emailed to RTI a few days prior to when data collection is scheduled to begin. Receiving the roster as close as possible to when data collection begins ensures the sampling frame is as accurate as possible. RTI statisticians will use the rosters to draw a random sample of 80 inmates within each facility; the number of inmates sampled is based on the second-stage sampling plan for the 2014 SPI.⁹ This list of sampled inmates will be transmitted to the interviewers’ laptops the day before they begin work at the prison.

Second-stage response rate. We have assumed a conservative response rate for the survey request of about 80%.¹⁰

Data Collection Procedures

The SPI Pilot Study data collection methodology is modeled after the approach that was used to conduct three waves of the NIS, for which over 250,000 inmates were interviewed across over 1,200 correctional facilities. RTI will work with prisons well in advance of data collection to gain all necessary approvals and work out all relevant logistics. The facility will be asked to provide a complete enumeration (roster) of their prisoners, from which 80 will be randomly sampled. Facilities will be asked to include on the roster each inmate’s name, unique fingerprint-supported State Identification (SID) Number, the FBI’s unique fingerprint identification number, date of birth, sex, race/Hispanic origin, admission date, offense type, sentence length, and housing unit where the inmate resides within the facility.¹¹ RTI will then send a small team of trained field interviewers to the facility on the agreed-upon dates who will work with facility staff to have each of the 80 sampled prisoners brought to private interviewing areas within the facility.

Prior to the onset of the interview, interviewers will read the informed consent document to the inmate and will offer a copy of the document for the inmate to keep as well (see Informed Consent and Data Confidentiality for more information). Once the prisoner agrees to the survey request, the interviewer will begin the interview. It is expected that at least 80% of sampled prisoners will be interviewed and thus we are anticipate interviewing between 384 and 480 inmates across the 6 prisons selected for the SPI Pilot Study. The data collection will last for 3 days within each facility and will run for approximately 5 weeks, start to finish. The interviewer training is planned for June 24 – 28, 2013. Data collection will begin on July 8, 2013 and continue through August 9, 2013.

The SPI Pilot Study survey instrument will be administered using CAPI technology. Interviewers will read the survey questions from the laptop computer screen and enter the inmates’ responses directly into the laptop computer. Based on internal testing of the instrument by RTI staff, the average length of the SPI Pilot Study survey instrument is estimated to be about 80 minutes (range: 70 to 90 minutes).¹² The domains covered by the survey instrument include:

• Demographics

• Criminal justice

• Socioeconomic status

• Mental health conditions and treatment

• Physical health conditions and treatment; disabilities

• Alcohol use

• Drug use

• Alcohol and drug treatment

• Facility programs and services

• Social support from outside the facility

• Rule violations within the facility

At the end of the survey, inmates will be asked to provide their SSNs to allow BJS to link their survey data to beneficiary records from the SSA (see Analysis Plan and Informed Consent and Data Confidentiality for more information). For inmates who can provide SSNs, BJS will request their records from the SSA. At the conclusion of data collection in each facility, interviewers will transmit survey data back to RTI where the data will be cleaned and analyzed. Feedback will be collected from interviewers and facility staff on the survey instrument, data collection methods, consent procedures, etc. so these paradata can be analyzed. Within a week of completing data collection, a thank you letter will be sent to each facility and the commissioner in the state thanking them for their participation in the pilot study (see Attachments D and E).

Burden Hours for Pilot Study

We request 658 hours for the SPI Pilot Study data collection (see Table 1). This estimate covers time spent by sampled prisoners listening to interviewers read the informed consent document, providing consent to the survey, and completing the survey (total time of 512 hours). We have conservatively assumed that all inmates who do not participate will meet with the interviewer and thus hear the survey request read to them which will take approximately 5 minutes. However, it is likely that at least a portion of the nonrespondents will be unavailable for interview and thus will experience no burden. As such, the estimated 8 hours of burden for nonrespondents is the upper bound of burden for this group. The last component of the total burden estimate includes contact with facility staff (total time of 138 hours). This estimate includes 18 hours of staff time (i.e., 3 hours per facility) to talk with RTI staff to figure out which days data collection will occur, describe the background check process, pull the roster together, verify the roster, and send the roster of prisoners to RTI. The estimate of facility contact burden also includes about 120 hours of staff time (i.e., .25 hours per respondent) to locate prisoners and escort them to and from the interview site.

Table 1. Estimated Burden of the SPI Pilot Study

Interview	Number of Responses	Average Time per Response	Total Time
Inmate Respondent	384	80 minutes	512 hours
Inmate Nonrespondent	96	5 minutes	8 hours
Facility Contact	6	23 hours	138 hours
TOTAL	486		658 hours

Analysis Plan

The goal of the SPI Pilot Study is to answer key questions related to the operational and logistical components of data collection in correctional environments, data collection procedures proposed for the 2014 SPI, and the updated SPI questionnaire including the quality of the data. The survey data and paradata will be analyzed and the analytic plan will be structured to answer the specific research questions listed under Request for developmental work. In addition, the RAP sheets and SSA records will be analyzed to determine the feasibility of relying on administrative records in the national study to produce reliable and valid statistics on pre-prison criminal histories and benefits of inmates. If the methods are determined to be feasible, the pilot study will also be used to establish linkage and analytic plans for the administrative records in the national study.

Together, these sources of information and analyses will be used to inform revisions to the plans for the national study and ensure that the national implementation will be as efficient and as effective as possible. There are no plans to archive the pilot study data for public use or to produce a substantive report presenting survey estimates. However, depending on the findings from the pilot study, BJS may choose to publish a methodological report on an important issue(s) in statistical or survey methodology, such as the results from the efforts to explore the use of administrative records in supplementing the inmate survey data to produce national statistics on pre-prison experiences of inmates.

Informed Consent and Data Confidentiality

At the start of the interview, the interviewer will read the informed consent document to the inmate and offer a copy of the document for the inmate to keep as well (see Attachment F). The consent document provides the purpose of the survey, an explanation of how the information will be used, the voluntary nature of the study, how the inmate was sampled, and an address to write with questions about the study (a phone number is not provided as the majority of inmates will not have access to a phone). In addition, prisoners will be told how important their participation is to this data collection and that identifying information collected is strictly confidential. At the end of the survey, prisoners will be asked to provide their SSN and interviewers will explain to inmates how their SSN will be used to link their survey data to external sources of beneficiary data. (In Attachment A, see page 157 in section “Interviewer Closeout Screens” for more information.)

The data collected for this project are protected under the Bureau of Justice Statistics statutory protection. This protects the data from potential subpoena (42 USC 3789g).

Data Security

Data will be transmitted from the field each night using an encrypted transmission system. Data will be housed within RTI’s Enhanced Security Network (ESN) at RTI International in Durham, NC until all identifying information is removed from the data records. Identifying information includes the inmate’s SID, the inmate’s FBI fingerprint ID number, the inmate’s name, the inmate’s date of birth, and his/her SSN. The ESN provides a secure infrastructure that meets the compliance standards for securing and handling data containing personal identifiers or data that require additional security. The identifiers will remain in the ESN until they are delivered to BJS for use in obtaining the administrative records from the SSA. RTI will transfer the data to BJS using a Pretty Good Privacy (PGP) Self Extracting Archive. The archive can be decrypted using a single password that RTI will provide to BJS separately from the data. This method of encryption and data delivery will maintain Federal Information Processing Standards (FIPS) compliance. Once the identifiers have been stripped from RTI’s file and RTI’s copies of the identifiers are destroyed, the remaining questionnaire data will be moved to RTI’s standard secure computer system which is password protected and protected by access privileges which are assigned by the appropriate system administrator. All systems are backed up on a regular basis and are kept in a secure storage facility. To protect the identities of the respondents, no identifying information will be kept on the final survey file. With respect to personnel, all RTI employees are required to sign a pledge of confidentiality. This pledge requires employees to maintain confidentiality of project data and to follow the above procedures when handling confidential information.

All inmate survey data sent by RTI and beneficiary records from the SSA will be physically stored at BJS which is located in a secure building that includes the Department of Justice’s (DOJ) Office of Justice Program (OJP) offices. All OJP Servers are backed up on a regular basis and stored in a secure location, specifically a locked room with access limited to only information technology personnel from the DOJ’s OJP, and require a badge swipe to enter. Technical control of the survey data and SSA records will be maintained through a system of firewalls and encryption.  The data will be stored on an isolated, secure server behind the DOJ’s firewall. Data will be encrypted to FIPS 201 standards using PGP encryption to ensure the integrity of data contained therein and to safeguard the confidentiality of participants’ personally identifiable information (PII). The inmate survey data and SSA records will be protected by access privileges assigned by the BJS information technology specialist. Access to the SSA records will also be limited to BJS staff that have signed SSA’s confidentiality agreement certifying that they understand SSA’s security, confidentiality and ethics requirements. All agreements must be sent to SSA prior to delivery of the administrative records. BJS can provide on-site access to the SSA records to RTI staff for linkage and analytic purposes if those RTI staff have signed and submitted SSA’s confidentiality agreement and have been cleared through OJP’s contractor clearance process. Audit trails are maintained to identify users, authenticate users, and trace users’ actions on the secure server which allows BJS to maintain individual accountability of all data users.

The identifiable SSA records will be stored on BJS’s isolated, secure server at OJP for up to 5 years after the project’s completion, at which point the identifiable elements will be stripped from the files and destroyed. Once the SSA data have been stripped of identifiers, linked to survey data, and extracted into a data file, the data will remain on BJS’s isolated, secure server and will continue to be PGP encrypted and protected by access privileges assigned by the BJS information technology specialist. 

Contact Information

Questions regarding any aspect of this project can be directed to:

Lauren Glaze

Statistician and SPI Project Manager

Bureau of Justice Statistics

801 7^th St, NW

Washington, DC 20531

Office phone: 202-305-9628

E-mail: Lauren.Glaze@usdoj.gov

Attachments

See the list below for additional attachments.

Attachment A: Survey questionnaire and specifications

Attachment B. Initial contact letter to state commissioners/BOP

Attachment C: Initial contact letter to selected facilities

Attachment D: Thank you letter to state commissioners/BOP

Attachment E: Thank you letter to selected facilities

Attachment F: Prisoner informed consent to interview

Attachment G: Reports from both rounds of cognitive testing

Attachment H: Agenda for interviewer training

Attachment I: Table of contents for the interviewer training manual

1 Prior iterations of BJS’s national survey of prisoners were known as the Survey of Inmates in State and Federal Correctional Facilities (SISFCF). The first survey of state prisoners was fielded in 1974 and periodically thereafter in years 1979, 1986, 1991, 1997, and 2004. The first survey of federal prisoners was fielded in 1991 along with the survey of state prisoners, and both have been fielded at the same time since 1991.

2 Wildeman, Christopher. (2009). “Parental Imprisonment, the Prison Boom, and the Concentration of Childhood Disadvantage.” Demography, 46(2): 265–280.

3 Petti, Becky & Sykes, B. (2008). “The Demographic Implications of the Prison Boom: Evidence of a ‘Third Demographic Transition’?”, University of Washington, Working Paper.

4 Pettit, Becky & Western, B. (2004). “Mass Imprisonment and the Life Course: Race and Class Inequality in U.S. Incarceration.” American Sociological Review, 69: 151-169.

5 Western, Bruce, Pattillo, M., & Weiman, D (2006). “Introduction.” In Mary Pattillo, David Weiman, and Bruce Western (Ed.), Imprisoning America: The Social Effects of Mass Incarceration. (pp.1-18). New York: Russell Sage Foundation.

6 Bushway, Shawn, Stoll, M., & Weiman D (Ed.)(2007). Barriers to Reentry? The Labor Market for Released Prisoners in Post-Industrial America. New York: Russell Sage Foundation.

7 For more information about this BJS project, see: Durose, Matt. A Standardized Approach to Collecting and Processing Multistate Criminal History Information for Statistical Analysis available on the Justice Research and Statistics Association’s website at http://www.jrsa.org/pubs/forum/forum_issues/for28_2.pdf.

8 Prisons will be sampled with probabilities proportionate to size (as measured by the number of prison inmates reported through BJS’s most recent Census of State and Federal Correctional Facilities for which data collection is near completion). This was the first-stage sampling method applied in the NIS and previous iterations of the SPI.

9 Through the SPI Design and Testing Project, BJS and RTI explored a number of different sampling options with the goal of producing national and some subnational estimates on key indicators with levels of precision that were equal to or better than the precision levels yielded from the last iteration of the SPI. The second-stage sampling plan of 80 sampled inmates for the 2014 SPI proved to yield equivalent or better precision than prior iterations of the SPI for key indicators without resulting in significant increases in costs.

10 Prior iterations of the SPI have yielded second-stage response rates of 90% or higher. However, since the last SPI, the NIS was in the field in 2007, 2008-2009, and 2011-2012 and second-stage response rates for NIS averaged about 70%. While the NIS covered a sensitive topic which may be why response rates were lower than prior iterations of SPI, to account for any potential adverse impacts from the NIS and changes in the correctional environment related to survey requests from the Department of Justice, we have assumed a more conservative response rate than was achieved in prior iterations of the SPI.

11 The data provided by the facilities will not only be used to sample inmates but will also be used to develop post-survey adjustments and to link the survey data to prisoners’ RAP sheets using the SID or FBI ID number.

12 The length of the interview in prior iterations of the SPI averaged around 60 minutes. The data collected through the SPI Pilot Study will be analyzed to make decisions about deletions to the instrument to reduce the average length of the instrument to be consistent with prior iterations of SPI. For example, problematic questions (e.g., questions that appear to be causing excessive cognitive burden, have long administration times, or result in poor data quality) will be identified and discussed for deletion. Timing data obtained through the pilot study will be used to identify the modules that take the longest to administer and that information will be used to scale back some modules.

8