CCSC_PartA_final

SUPPORTING STATEMENT (PART A)

Criminal Cases in State Courts (CCSC)

The Bureau of Justice Statistics (BJS) proposes a new data collection from state courts, the Criminal Cases in State Courts (CCSC). The CCSC project will collect criminal data to, at a minimum, replicate the Felony Sentences in State Courts report series, which was based on data collected from the National Judicial Reporting Program (NJRP; OMB control number 1121-0130, expired 4/30/2008).¹ NJRP was conducted every 2 years from 1983-2006. It was a nationally representative sample of state courts of general jurisdiction in 300 counties, and collected detailed information on cases that ended in felony convictions, including the demographic characteristics of felons, conviction offenses, types of sentences, sentence lengths, and time from arrest to conviction and sentencing. The Census Bureau, the data collection agent for NJRP, would send staff to the sampled courts and enter data from sampled cases by hand. Recognizing that most courts were implementing computerized case management systems, BJS solicited a redesign of the data collection methodology for NJRP and a second courts data collection, State Court Processing Statistics (SCPS; OMB control number 1121-0306, expired 7/31/2013) from 2009 to 2011, to determine the feasibility of collecting the data via computer extracts requested directly from state courts.

While nationally representative, the NJRP only collected data about cases that resulted in felony convictions. From 1988 to 2006, BJS maintained SCPS, a parallel data collection on felony case filings, although that data collection used a sample of courts of general jurisdiction in 40 of the 75 largest counties.² SCPS followed felony case filings for 24 months (homicides) or 12 months (all other offenses), so it was able to include data about cases that did not end in conviction and included cases that were filed as felonies but disposed as misdemeanors. SCPS also collected contextual case information, including past criminal history and pretrial release and misconduct. SCPS had authorization to collect these records from courts and other state agencies (like pretrial services agencies and state sentencing commissions) as needed. Because the data were collected from multiple sources, SCPS cost about as much as NJRP, but with a smaller sample. BJS discontinued SCPS in 2009 to investigate whether a nationally representative data collection could be funded and supported under the same solicitation referenced above.

Since 2011, many state courts have increased their ability to generate annual reports about some aspect of their caseloads, indicating that state courts have increased their capacity to extract at least some aggregated data from those systems. BJS tested its ability to utilize that capacity with a smaller data collection regarding juveniles charged as adults in the Survey of Juveniles Charged in Adult Criminal Courts (SJCACC, OMB control number 1121-0349, expired 11/30/2017). Data were collected in 2015-2016 with a reference year of 2014. Twenty-four states and the District of Columbia provided complete data extracts. BJS also collected partial data from four other states. In all, BJS obtained at least partial data from 28 states and the District of Columbia, and four additional states indicated that they would normally be able to provide statewide data, but could not at the time of the data request. Overall, 32 states and the District of Columbia were able to provide extracts in 2016.

BJS has also completed initial work for the CCSC to determine how many states allow bulk data requests from researchers, whether states have centralized data for the whole state, how many states have some centralized data but are missing counties from the system, and how many states have no centralized data.³ BJS has determined that 31 states and the District of Columbia have complete centralized data, 5 states have data systems that cover multiple counties but not the entire state, and 14 states have no centralized data.⁴

Ideally, BJS would conduct a census of state court systems and collect the needed data directly from the states. However, as state court data systems do not provide sufficient coverage of the country, and because states with centralized data may be unable or unwilling to participate in the data collection, BJS anticipates that for some portion of the country, counties will need to be sampled to produce nationwide estimates. BJS anticipates at least some counties would be selected with certainty in any national sample. This request for clearance includes a detailed description of BJS’s plan to collect centralized state data, the counties BJS expects to sample with certainty, and an outline of how BJS expects to proceed in decentralized states. BJS will submit a material change of circumstances memo with a detailed sampling plan once the census of state court systems has been completed.

BJS is unsure of the availability of data in Puerto Rico. BJS intends to assess the structure and the data reporting capacity of the Puerto Rico court system with the intent of assessing the availability and quality of data. At this time, BJS does not intend to include Puerto Rico in full data collection and reporting for the CCSC.

A. Justification

1. Necessity of Information Collection

Under 34 USC § 10132 (Attachment 1), BJS is directed to collect and analyze statistical information concerning the operation of the criminal justice system at the federal, state, and local levels. State courts are an essential component of the criminal justice system.

Since BJS discontinued the NJRP and SCPS programs, there have been no nationally representative reports on felony case filings or terminations in state courts. BJS has no basic data like how long a felony case takes from case filing to sentencing, or how many cases are disposed in a single year by offense type. As a result, states cannot assess whether their court case processing statistics are higher or lower than some national average, and the nation as a whole has no reliable indicator of the number of people charged with or convicted of felony offenses.

CCSC will provide an understanding of the criminal cases processed by state courts by collecting data on criminal case dispositions for a single year. Data collected include defendant demographics, method of disposition (e.g., dismissal, enrollment in diversion program, plea, court or jury trial, or other dispositions), the most serious charges at arrest and disposition, type of disposition, sentencing, and sentencing information (when available). BJS will examine the time between key case events, such as the time between case filing and case disposition and case disposition to sentencing. BJS also expects to report on the distribution of offenses, how often cases are terminated by trial as opposed to non-trial outcomes like taking a plea, and key demographic patterns among those convicted of felony offenses in the U.S.

Neither BJS nor any other organization is reporting case-level data from state criminal courts; at least, not without transforming the case-level information into indicators. The last nationally representative state court data collection was BJS’s NJRP in 2006, more than a decade ago. CCSC will provide a more complete assessment of how criminal cases are processed in state courts. The flow of cases through state courts affects trends in the correctional systems, and is a wide gap in national crime data.

BJS intends to ask state courts to provide an extract of all felony and serious misdemeanor criminal cases where a disposition was entered in 2019.⁵ After BJS receives the file, BJS and its data collection agents, Urban Institute and NCSC, will determine how best to easily create the analytical files. It is expected that a single data file will be too unwieldly to analyze. BJS will likely create a felony file, a serious misdemeanor file, a general jurisdiction courts file, and a limited jurisdiction courts file. File structure determinations will be made after the extracts are cleaned and standardized, mapped to BJS offense codes, and merged with the other state files.

The research questions the CCSC data will be able to address include -

• How many felony cases did state criminal courts process in 2019? How many misdemeanor cases?

• What types of offenses were charged?

• What were the demographic characteristics of defendants?

• How were the cases disposed?

• What were the sentences by case or by type of offense?

Universe of Cases

States vary as to how individuals are charged with crimes. The most common methods utilized are–

• Arrest—not all arrests are filed as cases, as prosecutors can decline to prosecute. There are other types of arrests that are made after a judge signs an arrest warrant or a bench warrant. Arrest warrants are typically signed after a judge reviews the investigation of a crime done by law enforcement. The arrest warrant signed by the judge and executed by law enforcement acts as the charging document in criminal court, but the court case is typically filed as an information or indictment.

• Summons/citations/promise to appear—law enforcement may not arrest an individual for less serious offenses but provide them with a summons and a court date. Summonses are often used for traffic cases, but can also be used for some misdemeanor offenses, such as shoplifting. The law enforcement agency and not the prosecutor’s office files the case with the court.

• Direct file by law enforcement—some jurisdictions eliminated the need for a case to be referred to prosecutors, and allow law enforcement to directly file the case with the court. This may be limited to certain types of offenses, depending on state statute.

• Sworn complaint—some types of offenses, such as domestic violence cases, can be filed through a sworn complaint from the victim or a witness. In this circumstance, a victim or witness will report to either a court or police station and take an oath, and then provide a written statement. A judicial officer, often a magistrate, will set a court date to hear the victim’s sworn testimony. At that hearing, the judge will either dismiss the case or file a temporary order of protection for a short time period in order to schedule a hearing where both parties will be present.

• Information or indictment—prosecutors typically file the case as an information after case review or as an indictment after a review by grand jury. In an information, a prosecutor reviews the arresting officers’ information and decides there is sufficient evidence to charge the individual with a crime. Cases filed by information do not require the review of the grand jury, and can be filed quickly by the prosecutor. A case filed as an information is often considered to be less serious than a case filed after a grand jury indictment. More serious cases with more serious offense types, such as felony violent crime cases, are typically filed after a grand jury indictment.

The method of filing affects where a particular case is filed. An indictment is typically filed in a court of general jurisdiction, while an information may be filed in a court of limited jurisdiction or one of general jurisdiction. Summonses, sworn complaints, and direct filings by law enforcement could be filed in a municipal court or courts of limited or general jurisdiction. Courts of general jurisdiction typically hear more serious criminal cases, such as felonies. Courts of general jurisdiction often act as intermediate appellate courts, hearing appeals from decisions made in limited jurisdiction courts.

Courts of limited jurisdiction are restricted to certain types of cases or to certain hearing types, and this varies by state statute. For example, limited jurisdiction courts may hear criminal and traffic cases where the punishment is less than a year in local jail or punishable by fine only, or handle bail hearings and first appearances or arraignments for felonies that will otherwise be handled in a court of general jurisdiction. Limited jurisdiction courts handle higher volume than general jurisdiction courts, because minor criminal cases are more common. Municipal courts are even more limited, hearing perhaps routine traffic, ordinance infractions and very minor misdemeanor cases.

Courts of general jurisdiction normally have better data reporting capabilities than courts of limited jurisdiction, and courts of limited jurisdiction usually have better data reporting capabilities than municipal courts. In the past, BJS only included courts of general jurisdiction in data collection efforts, because the cases were more serious and the data were better recorded.

Limited jurisdiction courts may be able to generate better quality data than in the past due to improvements in their information systems. BJS expects that statewide systems will be better positioned to provide limited jurisdiction court data, while county systems may not be. For that reason, BJS will pursue limited jurisdiction court data from statewide systems, and limit county system requests to general jurisdiction work. BJS will use the data from the limited jurisdiction court to provide data on the statewide systems only.

BJS will exclude municipal courts at this time. There are lists of limited and general jurisdictions in each state; however, no such lists exist for municipal courts. BJS would require significant time and additional funds to develop a list of municipal courts, which would divert the focus away from collecting data on the most serious cases.

Data Elements Requested

The core data of interest covered by the CCSC include case identifiers, defendant information, charge information, disposition of charges, and sentencing. The information will allow BJS to produce a national description of many of the felony and serious misdemeanor criminal cases processed in state courts.⁶

• Case identifiers. The information requested includes the county or locality reporting the case, the level of court (single-tiered court, limited jurisdiction court, or general jurisdiction court), and an anonymized case identifier.

• Defendant information. The information requested for the defendant includes an anonymized defendant identifier (to protect identity), date of birth, sex, race and ethnicity, indicator of indigent status at any point during the case, and state identification number. BJS will use demographic information to allow for comparisons by type of offense and case outcomes. The anonymized defendant identifier is to identify persons in the data in the event that the state defines a case at the charge-level (i.e., a single defendant with four separate cases, that are actually four charges arising from a single incident).

• Charge information. The information requested includes date of offense or date of arrest (courts reported having one or the other date), filing charge information (as available in the case management system. BJS expects courts will be able to provide statute number, description, whether the charge was a felony or misdemeanor, and degree of severity (e.g., felony 1 and misdemeanor a). This same information will be requested at disposition and sentencing (date, statute, felony/misdemeanor, degree of severity, and whether the charge was modified between filing and disposition). The request will include charge disposition, specifically conviction, transfer to another court, acquittal, mistrial, entry into problem-solving court, diversion to program other than problem-solving court, continued without a finding, dismissed, prosecution abandoned (nolle prosequi), or other outcome. The date of offense or arrest will be used to measure the time between offense or arrest and case filing. The dates of disposition and sentencing will be used to measure the time between filing and disposition, and between disposition and sentencing. The charge data will be used by BJS to report on the types and seriousness of the offenses. These will be compared with conviction information, when applicable.

• Sentencing information. The type of sentence ordered for convicted defendants includes type of sentence imposed (e.g., death penalty, life in prison, state prison sentence, jail sentence, lifetime supervision, extended supervision, split sentence, probation, house arrest, fine, restitution, community service, time served, post-sentence diversion program, or other sentence) and length of the sentence. Sentencing information will be reported using type of convicted offenses. BJS also requests any conditions of the sentence (e.g., sex offender registration, firearm restrictions, and motor vehicle consequences), whether the sentence is concurrent with or consecutive to other charges or cases, any credit for time served the defendant earned being incarcerated pre-conviction, and the amounts ordered for any fines, court costs, or restitutions.

2. Needs and Uses

BJS will use these data to produce national estimates of the processing of defendants charged in criminal court with a felony, and potentially a misdemeanor. With these data, BJS will report on demographics of the defendants, the charge types, and the case outcomes. BJS needs this information to better understand the nature and flow of these cases through the justice system. In addition, quantifying the volume and nature of criminal cases will help BJS to report overall variations and trends in justice system processing.

BJS also needs to conduct the CCSC to assess the feasibility of using administrative data from courts across the country to support its broader needs for detailed criminal court processing data. As mentioned, BJS tried to implement the National Judicial Reporting Program (NJRP) using a similar model of data collection in 2011. Early on in the NJRP work state courts raised concerns about the capacity of their administrative data to support such a national data collection effort. Working with the National Center for State Courts (NCSC), the sub-awardee for the CCSC, BJS has concluded that this approach is now not only possible, but also practical and cost-efficient.

Many constituencies will use the data collected by the CCSC. Among those who are expected to use the data are other Department of Justice and federal agencies, state policymakers, corrections officials, and researchers:

• National Institute of Justice could use the data to design court research programs.

• NCSC could use data collected by the CCSC to better target training and technical assistance for specialty dockets, criminal case processing times, or to assess outcomes of training and technical assistance (TTA).

• State policymakers and researchers can use the CCSC data to examine and predict the impact of adopting new policies, such as diversion programs.

• Finally, the nature of the proposed data collection will support the comparison of state statistical data to national statistical data.

BJS will make the CCSC data set available to the public through restricted use files located at the National Archive of Criminal Justice Data Archive (NACJD) at the University of Michigan, Ann Arbor, MI (https://www.icpsr.umich.edu/icpsrweb/content/NACJD/index.html). Researchers will be required to write a short justification describing their use of the data; provide a data security plan; and obtain approval or waiver from an official institutional review board (IRB).

BJS may consider providing a public use file in the future, but will initially offer the file under restricted access.

3. Use of Information Technology

CCSC will largely consist of the collection of electronic data files from states and counties. For the Survey of Juveniles Charged in Adult Criminal Courts (SJCACC), BJS received complete statewide data from 24 states and the District of Columbia. BJS also received partial data from four states, for a total of 28 states and the District of Columbia submitting at least partial statewide data. Four additional states indicated they could provide statewide data, but were unable to at the time of data collection (e.g., because they were transitioning to new statewide case management system or they considered the request to be too burdensome during the time period).

BJS obtained a generic clearance from OMB to interview state court leaders regarding the data elements available for extraction from their case management systems. BJS has concluded that the ability of states to provide court data on a statewide basis varies considerably –

• 31 states and the District of Columbia have centralized data systems (i.e., a single data file covers the whole state). The 31 states and D.C. cover about 47% of the total U.S. population aged 18 and over (the target population of interest),

• 5 states (Arizona, Colorado, Illinois, Virginia, and Washington) are mostly centralized; a few counties in each state, typically the most populous, are not on the centralized system. The five states cover about 10% of the total U.S. population aged 18 and over,

• The remaining 14 states have no centralized statewide data system. These 14 states cover about 43% of the total U.S. population aged 18 and over,

BJS will use a staged data collection strategy. In the first stage, BJS plans to request data from the 31 states and the District of Columbia with statewide data systems, and the data available from the five states with some centralized data (BJS will request data from the statewide systems and the counties in those states not on the statewide system).

Most sampling strategies proportional to population have a certainty stratum.⁷ Because of this, BJS plans to contact the 18 counties in the 14 decentralized states that have more than one million residents aged 18 and over as part of the first stage. BJS anticipates these counties will be selected with certainty in the final sampling strategy.

BJS does not have a final sampling plan for the counties in the remaining states that are not covered by in the first stage. In part, this is because BJS expects some states with statewide or near-statewide coverage to decline to participate. If that happens, BJS expects to sample counties from states that decline to provide statewide data, so the complete sampling frame for stage two depends on contact with the 36 statewide and District of Columbia systems and the 5 states that are mostly centralized.⁸

In the second stage of the data collection, BJS will stratify by population size the counties not covered by data collected in the first stage and sample from those counties. The frame for the second stage will be all counties with less than one million residents aged eighteen and over in the 14 states lacking statewide data systems, and counties in the 36 states or District of Columbia where the statewide data provider declines to cooperate with the BJS request.

If states or counties are willing but not technically capable of providing data, BJS will offer to assist the state by helping to write the programs to allow those extracts, or to scrape an online data website. BJS will provide the respondents with technical assistance as needed to minimize respondents’ efforts in data collection and to improve data quality control. Respondents will have the option to 1) provide a formatted or unformatted extract of only the relevant data or 2) provide a larger data file that the Urban Institute (the BJS data collection agent) will format into a file suitable for analysis (i.e., a data dump). Urban Institute will provide the respondents with a secure method for data transfer and will work with the respondent to determine the data format and completeness of the data. BJS’s experience with its other collections show that electronic data submission greatly reduces burden relative to other methods of collection. BJS and Urban Institute will evaluate submitted data using logic checks to uncover high rates of missing and out-of-range values.

4. Efforts to Identify Duplication

The CCSC is not duplicated by any other federal agency or program. BJS is the only government agency to collect nationally representative case-level data from state adult criminal courts.⁹

Other organizations collect data from state courts for various projects. The National Center for State Courts (NCSC, the subawardee for the CCSC) collects data from courts, most notably for the State Court Organization and Court Statistics Project.

State Court Organization describes the organization of state courts, including how many levels of trial and appeals courts in the state, the number of judges, requirements for juries and verdicts, and the number and types of court staff.

Court Statistics Project collects data from state courts about the numbers of cases filed by case type categories (e.g., criminal, civil, traffic, domestic relations, and juvenile). In the most recent collection, 43 states provided aggregate data about their criminal case filings in 2017. Data collected for the Court Statistics Project does not identify the number or types of charges in criminal cases, but does provide case filing and clearance measures for felony and misdemeanor cases. Neither collection includes defendant demographic information, case-level data, or sentencing information, as proposed in the CCSC.

Measures for Justice (MFJ) was founded in 2011 with the goal of developing a performance measurement system to assess local courts across different jurisdictions within and across states. MFJ wants to create sets of indicators that allow courts to be compared in terms of public safety, fair process, and fiscal responsibility. Currently, MFJ has approximately 50 measures, of which a total of 32 are available to various extents across different states and counties. MFJ collects case-level data from state courts; if data are not available from state courts, MFJ supplements with data from other justice agencies (e.g., prosecutors, public defenders, jails, and largest arresting agency). MFJ has also employed data-scraping of court websites.

BJS will provide nationally representative data on criminal state court felony and serious misdemeanor case processing across the U.S., while MFJ is developing a series of indicators to compare county or state court systems along various dimensions. While both the CCSC and MFJ projects could collect defendant demographic information, CCSC will report defendant race, while MFJ may create a ratio indicator of defendant race to the overall county race. MFJ could then report what percentage of the county’s total white population is involved in the criminal justice system, and compare that percentage to other counties in the state or across states.

A search of the National Criminal Justice Reference Service repository and other internet search engines did not reveal any other similar projects. The information is not available in any current BJS data collections.

5. Efforts to Minimize Burden

To minimize respondents’ burden, the CCSC offers multiple methods by which they can submit data. If feasible, they can provide electronic data extracts. States can submit their extract in any format to ease respondent burden (Attachment 2).¹⁰ BJS also expects that some states may prefer to provide a “data dump” of their case management system, where the state provides all of the cases handled by the state. If the state or county is not able to provide a data extract, BJS will offer to assist in writing the technical programs needed to extract the data, or to scrape a public website with the court data. The Urban Institute or a subcontractor will process formatted files, unformatted files, and data dumps to obtain the BJS data elements, as needed.

BJS and the Urban Institute developed a data extraction protocol based on preliminary research done in this project, which was tested with 1 state (Minnesota) and 8 counties listed in table 1 below:

Table 1. Pilot counties

Pilot Test Site No	County	State	Region	Population 18+
1	Orange County	California	West	2,487,180
2	Bexar County	Texas	South	1,478,380
3	Palm Beach County	Florida	South	1,202,485
4	New Castle County	Delaware	Northeast	439,108
5	Franklin County	Ohio	Midwest	1,005,657
6	Washoe County	Nevada	West	364,959
7	Penobscot County	Maine	Northeast	123,702
8	Kosciusko County	Indiana	Midwest	60,427
9	Minnesota State	Minnesota	Midwest	4,308,564

Feedback from this pilot test of the data collection procedures and from the thirty-eight interviews with state court data experts revealed that even in decentralized states, requests for data should still be made to the state court administrator. The pilot counties also requested that the data request clearly specify the data elements. The data extract guide was then revised based on the information gathered about available data elements.

6. Consequences of Less Frequent Collection

BJS has not collected nationally representative data on criminal courts using data extracts as the primary method of collection. BJS conducted the last nationally representative court data collection in 2006 with the NJRP, and the last subnational collection was in 2009 with SCPS. BJS also conducted the Survey of Juveniles Charged in Adult Criminal Courts in 2014 using statewide data extracts from available states, and from counties apportioned into Primary Sampling Units to represent the remaining states. BJS plans to collect the CCSC data every three years so changes in filing practices, sentencing, and other trends in court processing can be measured.

7. Special Circumstances

No special circumstances have been identified.

8. Adherence to 5 CFR 1320.8(d) and Outside Consultations

In the early phases of the CCSC project, BJS and Urban Institute consulted with substantive experts, NCSC, and court personnel, including court administrators and data specialists under a previous generic clearance request. Between July and November 2019, BJS, Urban Institute and NCSC contacted the state court administrators in all 50 states, the District of Columbia, and Puerto Rico. Of these, the project team held data information calls with 37 states and the District of Columbia.¹¹ These discussions resulted in suggestions for core data elements, the coordination necessary to request the data, and the estimated time and cost necessary to extract the requested data.

The research under this clearance is consistent with the guidelines in 5 CFR 1320.6. The first 60-day notice for public commentary was published in the Federal Register, Volume 84, Number 205, pages 56,838-56,839 on October 23, 2019 (Attachment 3). Seven comments were received: three requested copies of the data extract guide (with no subsequent comments), and one comment described the effort as being too broad. The commenter recommended that BJS focus more specifically on obtaining data from people who were denied access to court records, and to describe their experiences.

The other three comments (Attachments 13, 14, 15) requested that information on interpreters be added. Specifically, the requesters wanted to know: 1) for whom the interpreter was ordered (defendant, witness, victim) and by whom (requested by party or ordered by court), 2) whether the interpreter was present for all hearings, and 3) the qualifications of the interpreter. BJS did not ask about interpreters during the state interviews, but expects the tracking of interpreters to vary widely by jurisdiction. As a result, BJS did not make any adjustment to the data extraction guide because BJS did not want to add more burden to the request. BJS summarized the response to comments in Attachment 16.

The 30-day notice for public comment was published in the Federal Register, Volume 85, Number 30, pages 8325-8326, on February 13, 2020 (Attachment 4).

9. Paying Respondents

States routinely charge money to access their data systems. BJS will pay the states as required by the states.

10. Assurance of Confidentiality

According to 34 U.S.C. 10134, the information gathered in this data collection shall be used only for statistical or research purposes, and shall be gathered in a manner that precludes their use for law enforcement or any purpose relating to a particular individual other than statistical or research purposes. The data collected through the CCSC represent individuals charged publicly in state courts, and the information is included in the public domain. Expunged, sealed, and otherwise restricted cases will not be included.

BJS’s confidentiality statute (34 U.S.C. 10231) is being invoked for this collection, only in the event that state courts submit actual defendant identifiers, rather than anonymized identifiers. In the event that courts submit defendant state identification numbers, BJS will not create anonymized identifiers. BJS will not release any personally identifiable information in any public dataset.

11. Justification for Sensitive Questions

No questions or data elements of a sensitive nature are included in the data extraction guide.

12. Estimate of Respondent Burden

The CCSC data collection will employ various methods to obtain data from states and localities. To determine the burden imposed on respondents, estimates were obtained from criminal court data providers from the pilot counties and from interviews with the state court administrators. Additionally, estimates were reviewed by NCSC since they have Court Statistics Project that requires the processing of electronic court records. The CSP entails the submission of formatted electronic records from states. Based on the pilot work, BJS expects the following burden estimates:

Table 2. Burden estimates

	Source (state or pilot county)	Hours to prepare state or county for extract	Hours to verify data extract
Orange County, CA	Pilot county	40
Palm Beach County, FL	Pilot county	17
Alaska	State	40
Idaho	State	60
Indiana	State	11.5
Kentucky	State	60
Michigan	State	8
Nebraska	State	25
New Hampshire	State	80
New Mexico	State	24
Wyoming	State	40
Average burden	All	30	10

BJS averaged the reported burden from the 38 interviews conducted under a previous generic clearance and pilot counties that responded prior to the posting of the 60-day notice. BJS also estimates that an additional 10 hours of follow-up time will be needed to review codebooks, data fields, and potential data issues with each respondent. BJS did not receive any comments on the burden estimate from the 60-day public comment period.

The entire burden estimate for all 150 respondents (states and counties) is estimated at 6,000 hours. This includes 4,500 hours for respondents to prepare and submit the extracts, and 1,500 hours of follow-up contacts to validate the data submitted.

13. Estimate of Respondent’s Cost Burden

Respondents will incur costs for their time to respond. BJS expects that one information technology professional would be responsible for preparing and validating the extract, with pay approximately equivalent to the GS-15/01 level ($126,810). Based on this assumption, the office cost of employee time would be approximately $63.41 per hour. For each jurisdiction (state or county), the total respondent cost burden is estimated at $2,536.40 for the 40 burden hours, including time to discuss the data extraction guide, prepare the extract, and time to confirm and validate the extract. Fringe benefits are estimated to average 46% of the base cost, or $1,166.74, resulting in total salary and benefits of $3,703.14. Indirect costs are estimated to average 37% of the salary and benefits total, or $1,370.16, for an overall total respondent burden of $5,073.30 per respondent. For 150 respondents, the total burden is estimated at $760,995.

14. Costs to Federal Government

The total expected cost to the federal government for this data collection is estimated to be up to $1,446,990 over a two-year period. This work consists of planning, developing the questionnaires, preparation of materials, collecting the data, evaluating the results, and generating the reports. A BJS GS-level 14 statistician will be responsible for overseeing the project.

Table 3. Estimated Costs for the Criminal Cases in State Courts Project

BJS costs (for a 2 year period)
	Staff salaries		Year 1	Year 2	Total
		GS-14 Statistician (25%)	$30,329	$31,239	$59,573
		GS-15 Supervisory Statistician (10%)	$14,270	$14,698	$28,030
		GS-13 Editor (10%)	$10,266	$10,574	$20,170
		Senior BJS review of report	$5,000	$5,000	$10,000
		Total BJS Salaries	$59,865	$61,511	$121,376
	Fringe benefits (28% of salaries)		$16,762	$17,223	$33,985
	Subtotal salaries and fringe		$76,627	$78,734	$155,361
	Other administrative costs of salary and fringe (15%)		$11,494	$11,810	$23,304
	Subtotal BJS costs		$88,121	$90,544	$178,665
Data Collection Agent (Urban Institute/NCSC)
	Direct salaries		$416,500	$375,980	$792,480
	Indirect costs		$241,100	$164,975	$406,075
	Reimbursement to states and localities for data extract preparation		$25,000	$10,000	$35,000
	Computing and supplies		$20,530	$14,240	$34,770
Subtotal: Data collection Agents			$703,130	$565,195	$1,268,325
Total estimated costs			$791,251	$655,739	$1,446,990

15. Reason for Change in Burden

Not applicable to this project. New data collection.

16. Project Schedule and Publication Plan

Pending OMB approval, the CCSC data collection is slated to begin in Spring 2020 and end approximately one year later. Once all data are collected, they must be standardized to match the BJS data elements. The data can then be merged and assessed for patterns of nonresponse. State statutes will be mapped to BJS criminal offense definitions while the data collection is underway. During this time, Urban Institute or NCSC may contact the states to resolve any data quality issues.

After the data are cleaned and merged, the data will be analyzed and reported in a BJS report titled Criminal Case Processing in State Courts (Fall 2021). Since this is BJS’s first time collecting case-level criminal court data since 2009, there will be a technical report titled Standardizing State Criminal Court Data (Fall 2021).

Based on the Survey of Juveniles Charged in Adult Criminal Courts (SJCACC), BJS expects that states will indicate early on whether they plan to participate in the project. The first stage of the project may be shorter reflecting the fact that the second stage will likely take much longer. BJS expects the first stage to take approximately 5 months to obtain the data and identify any nonresponders, and the second stage to take approximately 12 months. Data standardization will occur during the entirety of the project, and is expected to finish in 24 months.

The proposed data collection schedule is outlined below:

Table 4. Data collection schedule

Activity/contact	Stage 1: Centralized states, partial states, and known certainty stratum counties	Stage 2: County sample (14 states, remaining counties not covered in FL, IN, MS), and any refusals from centralized states	Attachment
Develop frame for sample; draw sample		Weeks 1 - 20
BJS introduction letter	Week 1 - mail	Week 20-21 (end of month 5)	5
Contact data provider to arrange extract	Week 2 – mail, email or phone	Week 22	6
Initial follow-up (discuss data agreements, payment, etc)	Week 6 - phone	Week 28 (Month 7)	7,8
Final follow-up contact for data not yet received	Months 4-5 – phone, email	Months 11-15	9,10
Begin to determine non-responders from refusals	Months 4-5 (internal)	Months 11-15 (internal)	n/a
Confirm refusals: Stage 1: request permission to contact counties within the state Stage 2: replace counties and start cycle again	Week 20-21 (end of month 5)	Week 68 (end of month 16)
Data collection closes	End of month 5	Month 18	n/a
Outreach to confirm and clarify data	Week 12 – month 12	Months 13-19	11
Clean and map data	Week 12 – month 12	Months 13-20
Thank you email, once all data are validated and incorporated	Week 12-month 12	Months 13-20	12
Map state offense statutes to BJS NCRP elements	Week 12-month 12	Months 13-20	n/a
Merge state and county files	Months 15-20		n/a
Data documentation and delivery	Month 24		n/a
Reports and data delivered to archives	Month 26		n/a

As part of verification of the data, preliminary analyses will begin as soon as each data submission is received. This will allow for callbacks with the respondents to clarify the data.

17. Display of Expiration Date

The expiration date will be shown on the data extraction guides and on any mailed or emailed correspondence.

18. Exception to the Certificate Statement

Not applicable to this project. New data collection.

Attachments

1. 34 USC § 10132

2. Data extraction guide

3. 60 day notice

4. 30 day notice

5. BJS introduction letter

6. Request for data

7. Initial follow-up script

8. Sample generic MOU

9. Second follow-up

10. BJS final follow-up

11. Confirm data script

12. Thank you email

13. Civil Rights Division comments

14. NCAJ comments

15. LAFLA comments

16. BJS response to public comment CCSC

1 This document is limited to criminal cases filed in state courts. It does not include information on civil or juvenile cases filed in state courts.

2 SCPS final data collection occurred in 2009, and also attempted to collect data via data system extracts. When extracts were not possible, SCPS data collection agents collected the data in the field.

3 This work was done by examining state laws, state court websites, and state court guidelines or rules available publicly. There may be exceptions to state policies that were not listed, including exceptions for government data requests. Additionally, the quality of the data has not been assessed as courts can restrict case-level data access to certain persons (e.g., attorneys, defendants, probation agents) or by requiring a paid subscription to access the data.

4 Seven states (Colorado, Illinois, Idaho, Kansas, Maine, New Hampshire, and Oklahoma) expressly prohibit or restrict bulk data extracts, but BJS is unsure about how restrictive the bulk data prohibitions are. BJS will work with the state to develop a Memorandum of Understanding (MOU) regarding how the data are to be protected, used, and archived.

5 BJS will request all felonies and any serious misdemeanor cases where the potential punishment includes a potential punishment of a year or more incarceration. If states or counties find it easier to provide all criminal cases, BJS will identify the eligible cases after submission.

6 Municipal courts are excluded. Traffic cases that do not have a potential penalty of time in jail are excluded.

7 Urban Institute conducted correlations of county population with criminal case filings from NCSC’s Court Statistics Project. The correlations were typically 0.9 or higher, indicating the population size is highly correlated with criminal case volume.

8 Statewide systems may refuse to participate and prohibit BJS from contacting counties within the state. In that event, BJS will move the state’s counties to the sampling strata, calculate the number of counties needed proportional to the size of the strata, and then remove the state’s counties from the strata. BJS believes the likelihood of this occurring to be low, based on the participation in the SJCACC.

9 As noted previously, NCSC and MFJ do collect data from state courts. NCSC collects aggregate counts of cases filed by case type, but does not collect case-level information. MFJ collects case-level data from state courts, but transforms the data into measures to allow comparison of the fairness of different justice systems.

10 BJS will map state offense codes to BJS offense codes using a series of crosswalks developed in the National Corrections Reporting Program (NCRP). NCRP completed extensive mapping of state codes over the past 37 years of data collection.

11 The project team had planned to call Puerto Rico, but needed additional time to translate the state data interviews into Spanish. BJS decided not to call after the January 2020 earthquakes. BJS plans to conduct data scoping with Puerto Rico under this clearance. No additional burden is expected.

15