NPRP Memo to OMB

MEMORANDUM

MEMORANDUM TO: Shelly Wilkie Martinez

Official 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: Howard Snyder

Deputy Director

Bureau of Justice Statistics

DATE: December 17, 2013

SUBJECT: BJS request for OMB Clearance to conduct a survey on select criminal justice agencies responsible for the processing and management of pretrial defendants within the U.S through the National Pretrial Reporting Program (NPRP) under the generic clearance agreement OMB Number 1121-0339.

Introduction

The Bureau of Justice Statistics (BJS) is planning to conduct a survey on select criminal justice agencies responsible for the processing and management of pretrial defendants within the United States (U.S.) through a project titled the National Pretrial Reporting Program (NPRP). Prior to the implementation of this project, BJS intends to assess the feasibility of collecting nationally representative data about the pretrial population. The assessment will be conducted through the fielding of a prototype survey. Not only will this project inform BJS about the feasibility of developing nationally representative statistics about pretrial case processing, but it will also help BJS to identify challenges and strategies to successfully and efficiently field such a data collection program at the national level.

Background

The number of individuals in jail in the U.S. has risen dramatically over the past two decades; between 1990 and 2008, the daily jail population grew from about 400,000 inmates to approximately 800,000 (Minton 2011) ¹. Much of this growth has been driven by the increasing number of defendants held in jail while awaiting trial. Although jail populations were split fairly evenly between sentenced and pretrial inmates until the mid-1990s, pretrial defendants now comprise 61 percent of all inmates in local jails (Minton 2011). This burgeoning jail population has resulted in a growing burden to taxpayers. Housing pretrial defendants in jail is a major contributor to these escalating costs; according to the Florida Sheriff’s Association, for example, keeping pretrial defendants in thirty of Florida’s county jails costs taxpayers almost one billion dollars annually.

In addition, the pretrial stage of the criminal justice system is an important process in which the outcomes of a criminal case are influenced by the decisions of judges, prosecutors, defense attorneys, and jail administrators. Therefore, understanding pretrial decision-making and case processing is relevant to several criminal justice actors and agencies, thereby providing a clear justification for this project.

Given these ramifications for the management of jail populations, the importance on case outcomes, and the rising cost to taxpayers, the growth of the pretrial population has become a major area of concern for policymakers, practitioners, academics, and the general public. However, there is no nationally representative data available that could shed light on the processing of pretrial cases and cost-effective ways of managing pretrial populations. The only systematic data collection effort that captures information on pretrial processing from multiple jurisdictions is BJS’s State Court Processing Statistics (SCPS) program, which collects data on felony case processing in 40 of the 75 largest counties in the country. However, SCPS does not provide detailed data that are relevant to recent developments in the pretrial field, such as pretrial diversion/supervision programs and evidence-based practice in pretrial case processing.

Furthermore, SCPS does not include small or midsized jurisdictions, lacking the capability of producing national or regional estimates about pretrial detention or case processing. It also focuses exclusively on felony defendants, thereby providing no information about misdemeanants, who constitute the vast majority of arrestees each year. NPRP will address these limitations and provide crucial information to identify current trends and issues in the processing of pretrial cases.

Request for Developmental Work

BJS plans to engage in development research for the NPRP project under the generic clearance (OMB number 1121-0339). Given that BJS cannot fully predict how the national data collection program on pretrial case processing should be structured and implemented to achieve the accurate and timely collection of data, it is important to evaluate the feasibility of the NPRP project by canvassing jurisdictional capacities to report on detailed pretrial data. The results will inform the development of an effective design for a future iteration of NPRP data collection.

BJS has developed a preliminary study to assess the feasibility of gathering pretrial data from jurisdictions throughout the U.S. This feasibility study is called the Jurisdictional Capacity Survey (JCS). The study began by conferring with leading experts and practitioners in the field of pretrial justice, including the National Institute of Correction (NIC)’s Pretrial Executives Network and the National Association of Pretrial Services Agency. The input from this group led to the development of the JCS design.

This document is a request to OMB for developmental research under BJS’s generic clearance and seeks permission to field test the JCS (attachment A) in 75 jurisdictions to (a) assess the feasibility of collecting nationally representative data on pretrial case processing and (b) identify strategies to encourage survey participation from local criminal justice agencies. The generic clearance will provide BJS with a means to meet the obligations of the Paperwork Reduction Act of 1995 (PRA). The following sections describe proposed plans and methods to field the JCS.

Task 1: Identify the points of contact

The JCS will be administered through several steps. First, it will be necessary to identify the points of contact (POCs) with responsibility for completing the JCS. Wide variation exists across jurisdictions in terms of how pretrial functions are carried out. Some jurisdictions have a single, dedicated entity responsible for the processing and management of pretrial cases. In other jurisdictions, pretrial functions may be shared among a few local criminal justice agencies. Therefore, this task involves collecting information about the local structure of the criminal justice system and contacting local criminal justice agencies, including court personnel, sheriff and jail officials, and/or pretrial agencies, to identify the most appropriate person(s) to complete the JCS.

To this end, the project team has sampled 75 jurisdictions to participate in the JCS through purposive sampling that captures heterogeneity on key features, such as the size of jurisdictions and a dedicated pretrial program, related to jurisdictional ability to provide detailed pretrial data. Jurisdictions are generally counties or cities, and respondents can include multiple agencies. Table 1 shows the number of sampled jurisdictions for each of 18 strata, reflecting the heterogeneity of jurisdictional sizes and pretrial features characterized by the availability of pretrial programs and the use of commercial bails.

Table 1. Proposed Sampling Strata

Jurisdiction Size	Pretrial Program	Commercial Bail Use	Number of Jurisdictions
Small	Unavailable	Allowed	3
		Allowed but Rarely Used	3
		Not Allowed	3
	Available	Allowed	3
		Allowed but Rarely Used	3
		Not Allowed	3
Medium	Unavailable	Allowed	4
		Allowed but Rarely Used	4
		Not Allowed	4
	Available	Allowed	4
		Allowed but Rarely Used	4
		Not Allowed	4
Large	Unavailable	Allowed	5
		Allowed but Rarely Used	5
		Not Allowed	5
	Available	Allowed	5
		Allowed but Rarely Used	5
		Not Allowed	5
Total			72

In addition to the jurisdictions sampled through this stratification (n=72), the project team will include a few additional jurisdictions that are uniquely distinctive (e.g., the State of Kentucky – a state-level pretrial justice system; the District of Columbia – a federal independent pretrial agency; and the City of New York – a private corporation responsible for providing pretrial services in the most populous city in the United States). On behalf of BJS, the Urban Institute and its partners, the Pretrial Justice Institute and the National Association of Pretrial Services Agency (NAPSA), will reach out to local agencies in those jurisdictions via telephone and identify the POC in each jurisdiction. Many of these contacts are already known by NAPSA. Those contacts will be asked for contacts at other agencies within their jurisdiction that have pretrial data.

Once the POCs are identified, an invitation letter (attachment B) will be mailed to them. The letter will explain the purpose of the project and will be accompanied by a letter of endorsement from the National Association of Pretrial Services Agency and the National Association of Counties. The invitation letter will provide the URL to the web survey and instructions on how to complete the survey. Furthermore, it will contain the toll-free number and e-mail address for the project so that respondents may contact BJS or its data collection agent (Urban Institute) with questions or for assistance.

Task 2: Conduct online survey data collection

Task 2 involves the deployment of the survey. The online survey² consists of 36 questions across seven sections (A-G). Sections A and B ask for background information about the agency for which the point of contact works and how the pretrial population is defined by this agency. Section C gathers information about pretrial services provided and release mechanisms offered to the pretrial population through this agency. Sections D-F are designed to measure each jurisdiction’s ability to extract and manipulate data on the pretrial population and may require inter-agency collaboration to complete. Section D assesses the caseload of pretrial defendants by the type of release and offense severity.

Section E focuses on pretrial misconduct and asks agencies to provide information on re-arrest, violation, and court appearance rates. Section F tests jurisdictions’ ability to provide data on pretrial subpopulations by asking for the number of defendants who were detained, failed to appear in court, and were arrested for a new crime by age, race, gender, offense charged, release type, and prior criminal history. Finally, Section G provides a space for any comments, concerns, or clarifications. Table 2 provides a quick summary of each section.

Table 2. Summary of NPRP JCS

Section	Domain	N of items	Description
A	Pretrial Organization	2	Background information about agency
B	Pretrial Population	4	Background information about pretrial population
C	Pretrial Functions	4	Information about assessments and types of release used by jurisdiction
D	Pretrial Release Decisions	9	Data on detention and release, including information on type of release and detention mechanisms used
E	Pretrial Misconduct Section	3	Data on failure to appear, re-arrest, and violations
F	Pretrial Subpopulation Breakdown	7	Data on detention and misconduct by demographics, criminal history, and types of release
G	General Comments	1	Space for comments, suggestions, and clarifications

The design of the JCS reflects its aim to gather detailed information on the feasibility of collecting aggregate pretrial data for a random sample of jurisdictions in the future. Each of the questions in Sections D-F allows agencies to provide estimates when precise figures are unavailable. Each of the questions in these sections also asks agencies to report on the time required to retrieve the data and the challenges experienced while doing so. These features of the survey will help BJS identify the breadth of the data elements that jurisdictions can and cannot collect, the effort required to report on data, and the reasons why data for certain measures are not available.

Once the POCs submit the JCS, the project team will review survey responses and follow up to resolve open issues that may arise (see attachment C for follow up protocol). The project staff will also contact jurisdictions via e-mail and telephone that do not complete the survey to prompt them to do so. Thank you notes (attachment D) will be sent to jurisdictions upon survey completion.

Task 3: Summarize Survey Data and Debrief

The objective of Task 3 is to compile and assess the NPRP JCS responses. Findings generated from the review of the data and process will be used to assist in the refinement of data collection instruments and sampling methodologies for producing nationally representative pretrial statistics. The project team will also consult members of the NIC’s Pretrial Executive Network who have participated in the NPRP JCS to solicit additional feedback, which will help develop strategies for encouraging survey participation.

The findings from the survey and the feedback will be compiled into the NPRP Design and Development Report, which will highlight the survey’s findings and include different sampling design recommendations. The report will discuss the feasibility of developing national statistics on various aspects of pretrial case processing and the burden associated with each survey domain, such as the caseload of pretrial defendants, pretrial misconduct outcomes, the probability of detention given initial hearing, the length of detention, type of release, and the characteristics of persons detained. This report will also provide an action plan with details on how to encourage participation as well as an implementation plan. There is no intention to estimate population parameters with any level of precision. Instead, the data will be assessed to determine how to tailor the field approach (in the sense of Dillman’s total design method) to various “segments” of jurisdictions. This in turn will facilitate the development of efficient field protocols as well as refined cost and burden estimates for use in the design of the actual NPRP effort. As jurisdictions were strategically chosen to represent many different types of pretrial release systems, the results from these different types can be used to estimate the feasibility of a nationally representative collection. The level of effort and ability to respond to the survey for each type of jurisdiction can be applied to similarly situated jurisdictions not in this survey. This information can also be applied to the total cost to field a nationally representative survey.

Expected Burden to Complete the NPRP JCS Survey

The NPRP JCS is designed to be completed online. We expect that most respondents will make use of the online survey application. The online survey application will allow for multi-session, non-sequential completion of the survey instrument. As we anticipate that most respondents will need to seek out multiple information sources within their organizations, such flexibility in completing the NPRP JCS will facilitate data entry. It will also reduce the burden by allowing survey respondents to collect and provide information at their convenience. In addition, the online survey application will have tracking functions to effectively manage survey responses, and we will provide a toll-free number and e-mail address so that respondents may contact BJS or its data collection agent (Urban Institute) with questions or for assistance.

The draft NPRP JCS was pilot tested with several jurisdictions as listed below:

Jacksonville Sheriff’s Office, Florida

La Crosse County Chemical Health and Justice Sanctions, Wisconsin

Pretrial Services, San Diego Superior Court, California

Santa Barbara Superior Court Pretrial Services Division, California

Davidson County Sheriff's Office, Tennessee

These agencies range in size, region, and organizational structure. Respondents were asked to complete the survey and provide additional comments or suggestions regarding the clarity of survey questions. As part of the NPRP JCS, they were also asked to provide time burden estimates for completing the survey. We conducted a follow-up interview with them via phone, as needed, to clarify any issues or misinformation.

The estimated maximum time to complete the JCS (Tasks 1 and 2) and survey debriefing (Task 3) across all respondents in all jurisdictions participating the survey is 1,050 hours. The burden hour estimates are divided across the three tasks, including identification of points of contact (75 hours), survey data collection (900 hours), and survey debriefing (75 hours). The burden hour estimates are based on the pilot-testing results, feedback from the NIC’s Pretrial Executives Network, and prior experience collecting similar nationwide data. Table 3 summarizes the expected burden estimates.

Table 3. Expected Burden to Provide NPRP JCS Data

Tasks	Average burden per respondents	Total estimated burden hours
Identifying points of contact to complete survey instrument	1 hour	1 hour x 75 respondents = 75
Survey data collection	4 hours	4 hours x 225 respondents = 900
Survey debriefing	1 hour	1 hour x 75 respondents = 75
Total Respondent Burden = 1,050 hours

Human Subjects Research applicability

§45 CFR 46.102 defines a human subject as a living individual about whom an investigator conducting research obtains data through intervention or interaction with an individual or with his/her identifiable private information. Since the JCS collects information about the operation of criminal justice agencies, not about criminal justice professionals or pretrial defendants as individuals, the Urban Institute’s Institutional Review Board determined that the activities associated with data collection for this project are not considered human subjects research.

CONTACT INFORMATION

The contact people for questions regarding data collection and aspects of the design of this research are listed below:

Howard Snyder, Ph.D.

Deputy Director

Bureau of Justice Statistics

U.S. Department of Justice

810 7^th Street NW, Room 2326

Washington, DC 20531

Office Phone: (202) 616-8305

E-Mail: [email protected]

Tracey Kyckelhahn, Ph.D.
Statistician
Bureau of Justice Statistics
U.S. Department of Justice
810 7th Street NW, Room 2401
Washington, DC 20531

Office Phone: (202) 353-7381

E-Mail: [email protected]

APPENDICES

1. National Pretrial Reporting Program Jurisdictional Capacity Survey (JCS)

2. Initial Contact Protocol

3. Official Invitation Letter to Participate in JCS

4. Follow-Up Phone Call/Email

5. Thank You Postcard/Email

6. Debriefing Protocol

1 Minton, T. D. 2011. Jail Inmates at Midyear2011- Statistical Tables. Washington, D.C.: Bureau of Justice Statistics, Office of Justice Programs, U.S. Department of Justice.

2 A paper version of the survey will be available upon request.