CSLLEA OMB_Part A_FINAL

SUPPORTING STATEMENT

Census of State and Local Law Enforcement Agencies, 2018

Overview

The Bureau of Justice Statistics (BJS) requests clearance to conduct the 2018 Census of State and Local Law Enforcement Agencies (CSLLEA). Historically, the CSLLEA generates an enumeration of all publically funded state, county, local and tribal law enforcement agencies operating in the United States and provides complete personnel counts for the approximately 20,000 law enforcement agencies operating nationally. For the purposes of the CSLLEA, a “law enforcement agency” is a publically-funded government entity responsible for enforcing laws, maintaining public order, and promoting public safety. To be within scope of the CSLLEA, a law enforcement agency must employ the equivalent of at least one full-time sworn officer with general arrest powers.¹ BJS has conducted the CSLLEA periodically since 1986, making the 2018 administration the seventh wave of data in the statistical series.

The CSLLEA collects information from local police departments, sheriffs’ offices, primary state law enforcement agencies, and special jurisdiction agencies. Local police departments include municipal, county, and regional police that are granted authority from the local governing body that created it. Sheriffs’ offices are generally empowered to enforce laws at the county level, while primary state agencies provide general law enforcement services state-wide with a focus on highway and rural area enforcement. Special jurisdiction agencies provide law enforcement services in areas within a specific jurisdiction; they are usually full-service departments that have been granted law enforcement authority in tribal lands, parks, transportation assets (e.g., airports, subways), housing authorities, schools, hospitals, or government buildings.

The CSLLEA serves as the core of the law enforcement collections at BJS. The primary purpose of the CSLLEA is to provide personnel counts and the functions of all law enforcement agencies in the U.S. and receives high response rates (98-99%). The CSLLEA is the only national data collection that provides personnel counts by full-time or part-time status, sworn status, and primary job function. These data are frequently requested by practitioners, researchers and other stakeholders to compare similar law enforcement agencies and are also used in conjunction with other datasets such as the Federal Bureau of Investigation’s (FBI) Uniform Crime Reporting (UCR) Program and the Census Bureau’s American Community Survey.

In addition to providing these key data, BJS also will use the CSLLEA as a frame for its law enforcement surveys, including the Law Enforcement Management and Administrative Statistics (LEMAS) core (OMB 1121-0240) and supplements, and the Survey of Law Enforcement Personnel in Schools (SLEPS; OMB 1121-0339). The data captured on the CSLLEA instrument serve to inform which agencies may or may not be included in other surveys. For example, only general purpose agencies are on the frame for LEMAS surveys. Additionally, data from the CSLLEA will allow BJS to identify law enforcement agencies that serve school districts and those that have personnel that provide school safety, the two agency requirements for SLEPS. Using CSLLEA as a frame for these and other surveys means BJS does not have to burden agencies later to determine if they are in scope.

The 2018 CSLLEA collection involves two phases. In the first phase, BJS will cognitively test the revised instrument with 48 agencies based on agency type (i.e., local and county police, sheriff’s office, or special purpose) and size (i.e., 100 or more full-time equivalent sworn officers or less than 100 full-time equivalent sworn officers). A maximum of 8 agencies of each type and size will be asked to participate in testing. BJS has modified questions pertaining to limited sworn officers on the 2018 CSLLEA to better measure the work of some law enforcement agencies and will use the cognitive interviews to determine the ability of respondents to differentiate between fully sworn and limited sworn personnel.

Pending minimal modifications to the instrument in the first phase, in the second phase, BJS will conduct the main data collection. The 2018 CSLLEA is designed to collect general information on state, county and local law enforcement agencies. The 7-item survey asks about the level of government that operates the agency, total operating budget, personnel counts and the functions the agency performs on a regular or primary basis.

A. Justification

1. Necessity of Information Collection

Under Title 34, United States Code, Section 10132, the Bureau of Justice Statistics (BJS) is directed to collect and analyze statistical information concerning the operation of the criminal justice system at the federal, state, tribal, and local levels. It disseminates high quality information and statistics to inform policy makers, researchers, criminal justice practitioners, and the general public. The Criminal Justice Statistics Program encompasses a wide range of criminal justice topics, including victimization, law enforcement, prosecution, courts and sentencing, and corrections. Law enforcement agencies are the primary point of entry into the criminal justice system and play a crucial gate keeping function in receiving reports of offenses, investigating crimes, making arrests, and detaining suspects. Subsequently, these agencies are a major provider of statistical data on crime.

The size, purposes, efficiency, fairness, and impact of law enforcement in the U.S. are ongoing national policy issues. Law enforcement agencies in the U.S. are numerous, diverse, and highly fragmented with substantial differences in size, role, and activities. Data collected during the 2008 Census of State and Local Law Enforcement Agencies (CSLLEA) administration indicate there were 17,985 agencies operated by general purpose agencies (i.e., any public agency with sworn officers whose patrol and enforcement responsibilities are primarily delimited by the boundaries of a municipal, county, or state government) and special purpose agencies (e.g., campus law enforcement, transportation, natural resources, etc.). These law enforcement agencies employed more than 1.1 million full-time personnel in 2008, including 765,000 sworn officers. About half (49%) of these agencies employed fewer than 10 full-time sworn personnel, while the largest 7% of state and local law enforcement agencies employed nearly two-thirds (64%) of all sworn personnel. Agencies serve a variety of functions from patrol and response, criminal investigation, traffic and vehicle-related functions, detention-related functions, court-related functions, special public safety functions (e.g., animal control), task force participation, and specialized functions (e.g., search and rescue).

Collecting data on issues related to law enforcement personnel and functions are of critical concern to BJS – since its mission to collect, analyze, publish, and disseminate information on the operation of justice systems at all levels of government. Developing and maintaining an accurate picture of the nation’s law enforcement workforce is paramount to understanding the current state of policing in the United States. BJS has conducted establishment surveys of law enforcement agencies regularly since 1987, with the core of BJS’ law enforcement statistics program being the CSLLEA.

The proposed 2018 CSLLEA is the only systematic, national-level data collection providing a complete enumeration of the Nation’s state and local publically-funded law enforcement agencies and counts of their personnel. In addition, the CSLLEA provides the basis for distinguishing among various types of law enforcement agencies by asking respondents to indicate the types of functions performed by the agency (e.g., law enforcement, investigative, jail management, court security, and process serving). Unlike other sources of law enforcement information, such as the Federal Bureau of Investigation’s (FBI) Police Employee Data or U.S. Census Bureau’s Equal Employment Opportunity (EEO) collection (see section 4), the CSLLEA provides additional data to supplement basic personnel counts, including breakdown of agency functions, and how sworn personnel are allocated to functions such as duties related to law enforcement, jail operations, and courts.

Information collected through the CSLLEA can provide local area estimates of personnel counts and functions; statistics that are essential for better understanding the landscape of American policing. These important statistics can be used to examine the correlates of crime and estimate the effect of law enforcement practices on crime rates. Agency-level data are needed to better understand law enforcement performance in terms of crime and victimization rates. In addition, an understanding of the number and functions of law enforcement personnel will assist in planning for public welfare response to national emergencies. During a disaster, law enforcement officers play a critical role in operations such as search and rescue, evacuations, door-to-door checks, and maintaining overall public safety. Information about the number of law enforcement personnel employed by agency and the functions these officers perform will aid in evaluating needs for national emergency preparedness. No other national data collection can provide comprehensive data on the services performed by law enforcement agencies.

Accurate counts of law enforcement personnel, as well as the type of agencies employing personnel, are necessary for implementing the majority of BJS’s law enforcement statistics program. Without these statistics, BJS could not produce national estimates on the organization and administration of police and sheriffs’ departments, including agency responsibilities, operating expenditures, job functions of sworn and civilian employees, demographic characteristics of sworn personnel, officer salaries, education and training requirements, equipment use, and polices.

The last administration of the CSLLEA was in 2014 (OMB approval obtained on 5/15/2014; 1121-0346). The 2014 CSLLEA was not publically released due to data collection issues and lower than expected response rate due to survey length. In regards to data collection issues, the sampling frame development was poor, with agencies classified incorrectly, and duplicate and out-of-scope agencies were both over- and under-identified. Furthermore, there was a mail merge error that scrambled the agency IDs within certain states that the data collection agent was unable to remedy. The 2014 CSLLEA instrument was expanded to 21-items from 6-items in the 2008 administration. While part of the expansion was due to disaggregating the one function question into 5 separate items, the 2014 instrument included 10 new items. As such the 2014 CSLLEA only achieved an estimated 84% unit response rate with an average 80% item response rate among valid responses. BJS determined the issues severely compromised the data quality and has not released the 2014 data.

To remedy the data collection issues and prevent them from affecting this administration of the CSLLEA, BJS has taken several steps. A new data collection agent, RTI International, has been selected to administer the 2018 CSLLEA and has already successfully administered the 2016 LEMAS Body-Worn Camera Supplement (BWCS; OMB 1121-0354) and 2016 LEMAS core. RTI also began work with BJS in 2015 on the Law Enforcement Agency Roster (LEAR) once it became apparent that the 2014 CSLLEA would not be useful as a frame. Using the 2008 and 2014 CSLLEA universe as the base, the LEAR integrated multiple datasets in an effort to compile a complete list of active general purpose law enforcement agencies.² The goal of the LEAR was to serve as the universe list from which the 2016 LEMAS core and 2016 BWCS samples could be pulled. The LEAR served this purpose and work has continued to integrate special purpose agencies into the frame for the 2018 CSLLEA administration. As such the original LEAR has evolved into the Agency Records Management System (ARMS). The ARMS improves the accuracy of agency contact information by incorporating enhanced review features, faster loading of updated point of contact information from survey administrations and better integration with the LECS website to allow agencies to update their contact information. The goal of the ARMS is to increase the speed, accuracy, and ease of keeping agency contact information up to date while reducing burden on staff. Additionally, the ARMS will allow BJS to continuously maintain a frame for the CSLLEA and other BJS law enforcement collections rather than having to create a new frame for each administration. One of the issues with the lag time between CSLLEA administrations is the lack of a current point of contact information for each agency, which reduces effective outreach. The ARMS allows BJS to continuously update contact information for each data administration. This will ultimately reduce follow-up time and data collection errors when merging in agency information with survey data. The current ARMS work has involved verifying agency operating status and current point of contact information for special purpose agencies through online web searches. Once clearance is obtained for the 2018 CSLLEA, RTI will conduct direct agency outreach if information cannot be obtained online.

To address issues related to the questionnaire length, BJS has reduced the number of items on the 2018 CSLLEA to better match the 2008 administration. The 2018 CSLLEA contains 7 items, including aggregating the functions from five items to one item. Even with the addition of a limited sworn category on the personnel items, BJS anticipates a reduction in burden as the additional category will only help clarify these items for agencies with both types of sworn personnel. Since BJS will also be testing the instrument as part of this clearance package, we will be able to assess if these assumptions are true. If they do not add clarity, the limited sworn category will be dropped and the items will mirror what was administered previously without issues.

The first page of the 2018 CSLLEA captures basic descriptive information about the name, address and Agency Record Identifier (ORI), which can link the responses to past and future law enforcement organizational surveys. Information about the person completing the survey is also captured. This information directly addresses a 2009 review of BJS programs by the National Research Council (NRC), which recommended that BJS law enforcement surveys should “collect more information about law enforcement agency behavior and performance and enhance the use of agency identifiers to encourage the linkage of agency-specific organizational characteristics with agency specific-crime statistics and with the demographic characteristics of the jurisdictions served by each agency”. In addition to capturing this on the survey, the ARMS contains multiple identifiers such as ORI (used by the FBI UCR) and FIPS codes used by the Census Bureau. The ARMS ID will be used on the CSLLEA instrument and the ORI reported by agencies will be verified against what we have obtained previously. These unique IDs will already be associated with each agency within the CSLLEA to allow for easy linkage to other datasets, including the LEMAS and past CSLLEA waves.

The proposed CSLLEA instrument (CJ-38; Attachment 1) includes 7-items that fall into four categories:

1. Agency Geographic Jurisdiction (Q1-2)

Previous administrations of the CSLLEA (1986, 1992, 2000, and 2014) included an item that described the type of government or geographic jurisdiction covered by the agency (Q1). The item allows BJS to classify agency type, which is used for stratification to sample for other collections. It also allowed BJS to examine the functions of law enforcement agencies across different types. New to the 2014 CSLLEA is an additional item asking if the agency serves as the primary law enforcement agency within that jurisdiction. Due to increasing coordination among law enforcement agencies, this will help us identify who takes the lead in enforcement versus who serves in a supplemental capacity. For certain survey collections, such as the LEMAS core, one of the respondent requirements is for the agency to have primary law enforcement function so including this item will allow BJS to better identify eligible agencies for the sampling frame.

2. Operating Budget (Q3)

We collect the total operating budget from each law enforcement agency. This information is often used in conjunction with personnel size (authorized vs actual) in order to assess budget constraints. Budget is also used to determine optimal staffing sizes.

3. Functions (Q4)

The proposed 2018 CSLLEA instrument will indicate the type of functions performed including patrol and response, criminal investigation, traffic, detention-related, court-related services, forensic services, task force participation and specialized functions. Detailed sub-functions in each of these sections will provide a descriptive overview of the various services and roles law enforcement agencies provide to the community and general criminal justice system.

4. Personnel (Q5-7)

The proposed 2018 CSLLEA instrument will indicate the number of authorized full-time sworn positions and collect detailed information on the number of full-time and part-time, fully sworn, limited sworn and non-sworn paid personnel in law enforcement agencies. These counts provide a general description of agency size and are used to identify changes in the number and type of personnel employed nationally. Of special interest are differences between sworn officers with general arrest powers and those with limited arrest powers. While BJS has measured the counts of general versus limited in LEMAS administrations and has found evidence that this is a useful distinction for sheriffs’ offices, little is known about the use of limited sworn personnel in special purpose agencies. Through cognitive testing we will be able to determine how often limited sworn officers are employed by agency types and also obtain further clarity on proper terminology (e.g., fully sworn versus limited sworn) on these items.

In addition to providing a national enumeration of full-time sworn personnel employed by law enforcement agencies, the instrument will collect data on the sex of officers. About 12% of sworn officers in general purpose agencies are female. We do not know the extent of female representation in special purpose agencies. There has been increased attention on the number of women in law enforcement and the CSLLEA will be able to provide this information for every law enforcement agency in the US.

Cognitive testing of the 2018 CSLLEA will be conducted in spring 2018 immediately after receiving OMB approval. Pending successful completion of cognitive testing (i.e., instrument requires minimal modifications that will not increase respondent burden), the 2018 CSLLEA full data collection will begin August 2018 and commence May 2019. During this 9-month data collection, the instrument will be administered to approximately 20,000 law enforcement agencies in the US.

2. Needs and Uses

BJS employs various methods to capture data to better understand the criminal justice system. For example, BJS captures data on crime from resident and inmate surveys and collects administrative data to supplement survey data where available. Data collections on agency characteristics are primarily conducted through establishment surveys, and this is the primary data collection vehicle for the law enforcement core collections. The CSLLEA is the only systematic establishment survey that produces local-level estimates of personnel and functions of almost every law enforcement agency in the U.S.

BJS Needs and Uses

BJS has conducted establishment surveys of law enforcement agencies regularly since 1987. The core of BJS’ law enforcement statistics program is the CSLLEA. The primary goals of the CSLLEA are to (1) develop a national roster of active publically-funded law enforcement agencies that employ the equivalent of one full-time sworn personnel with general arrest powers, and (2) generate accurate and reliable national statistics describing the characteristics and functions of these agencies.

The primary use of the national law enforcement roster is to ensure an accurate universe of agencies for all BJS law enforcement programs, including the Law Enforcement Management and Administrative Statistics (LEMAS) core and supplement surveys, Census of Publicly Funded Forensic Crime Laboratories (CPFFCL), Survey of Law Enforcement Personnel in Schools (SLEPS) and Survey of Campus Law Enforcement Agencies (SCLEA). In addition, the roster generated from the 2018 CSLLEA will be used to update information in BJS’s Law Enforcement Agency Identifiers Crosswalk (LEAIC). The LEAIC is designed to link other data resources, such as crime counts from the FBI’s UCR program or U.S. Census data, to each law enforcement agency in the US.

Since 1987, BJS has successfully implemented 10 waves of the LEMAS (OMB 1121-0240) core survey. The LEMAS core samples roughly 3,500 state and local agencies and provides national estimates about the characteristics of the 15,500 state and local general purpose law enforcement agencies; the functions they perform; the resources available to them; the number, type, and working conditions of their employees; the automation of agency functions and their information systems; the extent to which weapons are authorized and used; the formal policies that guide and restrict the behavior of sworn personnel; and the organizational responses utilized by these agencies to address contemporary law enforcement challenges. Agency type and number of full-time sworn officers collected through the CSLLEA are needed to draw a representative sample of agencies for the LEMAS core survey.

Similarly to LEMAS, both the CPFFCL and SLEPS collections rely on the CSLLEA for frame development. The CPFFCL is a survey administered to approximately 500 publically-funded forensic crime laboratories in the US, a majority of which are connected to a law enforcement agency. BJS has conducted the CPFFCL periodically since 2002. The CSLLEA provides necessary information for developing the CPFFCL frame by identifying which law enforcement agencies run a forensic crime laboratory. The SLEPS is a new collection through which BJS will gather national-level data on the extent of law enforcement involvement in the nation’s schools and the typical roles and responsibilities of officers working in schools. Since there are no other national collections that capture the number of school resource officers employed by every law enforcement agency, the CSLLEA is essential for building the SLEPS frame. The SLEPS was originally awarded in 2015 and the 2014 CSLLEA was to be the frame. Since the 2014 CSLLEA could not be used as the frame and there were no other viable alternatives, the SLEPS has been postponed until after the 2018 CSLLEA is completed.

In addition to the general purpose agencies captured in the LEMAS survey, the CSLLEA frame provides a means for identifying a variety of special purpose agencies. Special jurisdiction law enforcement agencies are responsible for providing police services in areas within another jurisdiction. These types of agencies are usually full-service departments granted law enforcement authority in parks, transportation assets (e.g., airports, subways), housing authorities, schools, hospitals, or government buildings. The single largest type of special purpose law enforcement agencies are campus police, which BJS surveyed in 1995, 2005, and 2012. The 2012 SCLEA (OMB 1121-0334) was administered to a sample of campus law enforcement agencies operated publically. BJS plans to administer the next SCLEA in 2020 and will use the 2018 CSLLEA for frame development.

The universe generated from the CSLLEA will also be used to update information in LEAIC. The LEAIC is designed to link other data resources to BJS law enforcement data and used to associate local governments to law enforcement agencies in order to allocate the Bureau of Justice Assistance’s Justice Assistance Grants (JAG) funding. In addition, the universe of state, local, and special purpose law enforcement agencies is used by other DOJ offices to generate sampling frames for data collections, such as OJJDP’s National Youth Gang survey.

In addition to providing a universe of agencies to ensure an accurate sampling frame, the CSLLEA is itself a valuable source of descriptive information on the characteristics and trends in law enforcement employment in the U.S. The administration of the 2018 CSLLEA will produce national statistics about the number of publically funded law enforcement agencies, the number of sworn and non-sworn personnel, and the range of functions performed by those agencies during reference year 2018 (2017 for cognitive testing). These data will be used to produce multi-year trends regarding characteristics of state, local, and special purpose law enforcement agencies.

Without the CSLLEA data, BJS will be unable to describe the functions performed by law enforcement agencies to the nation. For example, the CSLLEA provides concrete measures of the extent to which publically funded law enforcement agencies are capable of providing specialized services, such as crime analysis, bomb/explosives disposal, underwater recovery, and direct victim assistance or programs. In addition, the CSLLEA provides a measurement of the number of full-time sworn personnel assigned to patrol, investigations, school safety, jails or courts. The CSLLEA is also able to capture changing patterns of law enforcement personnel and functions that are then used to develop more detail collections or investigations through other vehicles. For example, the National Institute of Justice has expressed interest to BJS in administering a survey on the capabilities of law enforcement agencies to handle digital forensic evidence. The CSLLEA will identify those agencies that process digital forensic evidence, which can serve as the basis of a frame for a LEMAS supplement on digital forensic evidence.

The list below details the type of information that will be available through the 2018 CSLLEA data:

• Number/percentages of law enforcement agencies by government entity

• Rates/percent of law enforcement agencies that are primarily responsible for law enforcement

• Average total operating budget

• Rates/percentages of agencies that regularly engage in the functions of patrol and response, criminal investigation, traffic and vehicle-related functions, detention, court, forensic services, special safety, task force, and specialized tasks

• Number of full-time and part-time fully sworn officers, limited sworn officers and non-sworn employees

• Sex of full-time sworn personnel with general arrest powers and with limited arrest powers

• Number/percentages of officers whose primary assignment is to patrol, investigations, school safety, jail duties or court duties.

BJS will use the data gathered through the administration of the 2018 CSLLEA to disseminate information about law enforcement to the public. Past reports using the CSLLEA (http://www.bjs.gov/index.cfm?ty=dcdetail&iid=249):

National Sources of Law Enforcement Employee Data

Tribal Crime Data Collection Activities, 2012

Compendium of Tribal Crime Data, 2011

Hiring and Retention of State and Local Law Enforcement Officers, 2008

Census of State and Local Law Enforcement Agencies, 2008

Tribal Law Enforcement, 2008

Census of State and Local Law Enforcement Agencies, 2004

Census of State and Local Law Enforcement Agencies, 2000

Census of State and Local Law Enforcement Agencies, 1996

Census of State and Local Law Enforcement Agencies, 1992

Uses of the CSLLEA by Others

The information generated from the CSLLEA is widely used and cited by the law enforcement professional and research communities. The statistics generated by the CSLLEA are requested and used by police chiefs, sheriffs, legislators, planners, researchers, and others to identify personnel and budgetary needs, trends, and priorities in law enforcement. The CSLLEA has been used to track employment trends in state and local law enforcement in the U.S. since 1992, and will continue to inform policy making, planning, and budgeting at all levels of government.

According to the National Archive of Criminal Justice Data at the University of Michigan (NACJD), the CSLLEA data have been used in 34 academic publications.³ A majority of these publications have focused on the number of police departments in the U.S. and their size. For example, research in this area has examined the relationship between police agency size and crime clearance rates. The CSLLEA has also been used to characterize law enforcement agencies by the functions they perform, such as forensic services and serving tribal lands.

In addition to academic publications, BJS receives a number of information requests that can be fulfilled by the CSLLEA data. Uses of information include policy decisions, budget hearings, research and planning, market research, benchmark comparisons, grant applications, and journalistic purposes. BJS tracks the types of information requested that can be provided through its law enforcement data collections, as well as those requests that the data cannot fulfill. Since 2014, BJS responded to approximately 100 requests in which the CSLLEA data was referenced. A number of these inquiries asked about the status of the 2014 CSLLEA and if we had data newer than 2008 available. Examples of entities that have request data that could be fulfilled by the CSLLEA in the past year include:

BuzzFeed News: Sought data to calculate the number of sworn officers on patrol duty for every local/state law enforcement agency in the U.S.

Vice News: Asked for a list of the 50 largest law enforcement agencies in the U.S.

CBS Evening News: Sought information on the total number of police officers in the U.S.

Leverage Assessments: Asked to obtain a dataset with each agency in the U.S. and their sworn officer count.

360Armor, Inc.: Asked to obtain a list of all law enforcement agencies in the U.S.

NBC Nightly News: Sought information on the percent of female officers nationwide.

Pennsylvania State Police: Asked for a list of law enforcement agencies ranked by size.

Chesterfield Police Department: Asked for a list of law enforcement agencies within Virginia and the number of sworn law enforcement officers.

DataNerds.com: Sought a list of law enforcement agencies in the U.S. along with their jurisdictional level (i.e., local, county, state, etc.).

Quicket Solutions: Sought a database containing the number of sworn officers by state, county and city in U.S.

Pegasus Program: Asked for a list of sheriffs’ offices and their sworn officer size.

Tuscaloosa Police Department: Sought a list of law enforcement agencies with 237-327 (287 +/- 250) sworn officers grouped by state.

Genetec: Asked for a list of law enforcement agencies in the U.S.

The Marshall Project: Sought a list of 1-officer law enforcement agencies in the U.S.

ABC7 Chicago: Asked for a list of law enforcement agencies in Illinois.

POLICE Magazine, PoliceOne and Ph.D. candidate: Sought information on the availability of the 2014 CSLLEA data.

Anticipated Products

BJS anticipates producing two primary reports from the 2018 CSLLEA. Detailed information on the reports to be produced is discussed under Section 16, Project Schedule.

At the time of the initial publication from the 2018 CSLLEA, BJS will release fully-documented data files for public use through the NACJD.

3. Use of Information Technology

The 2018 CSLLEA uses a multi-mode design in which respondents are directed to a web survey through mailed and emailed instructions. The web survey is hosted by BJS’s data collection agent, RTI International (RTI).⁴

The instrument has been designed using commercially available survey software that will allow RTI to send an email to respondents explaining the CSLLEA survey and containing a hyperlink to the questionnaire. Respondents will access the survey website using a unique Personal Identification Number (PIN) and password provided by RTI. Attachment 2 shows screen-shots from the 2018 CSLLEA questionnaire and the page layouts that web respondents will encounter as they complete the survey.

The web survey application will incorporate consistency checks to validate data entries and machine edits that check for inconsistent, out-of-range, or missing responses. These automated processes will help ensure data quality and minimize respondent burden resulting from follow-up contact to resolve data discrepancies or other issues. Respondents will be able to start the survey, break-off, and later resume from the point in the survey where they last entered data. The survey software will allow for real-time online tracking of respondents thereby allowing BJS to monitor the completion of each agency’s responses. In addition, the web system supports the export of survey data and paradata in various formats specified by BJS.

Agencies may have various reasons why they do not respond via the internet, for example some might not have reliable internet access and others might find it difficult to complete online because of the need to involve multiple people in preparing the response. Agencies that require paper access will have multiple methods of receiving paper versions of the instrument. Hard copies will be sent via mail during routine non-response follow-up. Agencies will be able to download a PDF version of the survey from the survey site that can be printed or e-mailed to agency staff. Respondents can then gather data in hard copy and enter it into the online survey instrument or scan and return the completed survey form via mail or e-mail.

To process completed hard copy surveys, RTI will use a software package that employs Optical Character Recognition (OCR) to electronically convert scanned images of handwritten, typewritten or printed text into machine-encoded text. Data captured via OCR will be manually reviewed to ensure accuracy. Use of this technology will minimize paper-handling, reduce processing time, increase reliability, and enhance retention of written survey responses.

Upon completion of the project, the final dataset and supporting documentation will be made available to the public without restriction in an online archive in multiple statistical platform formats. Access to these data permits analysts to identify the specific responses of individual agencies and to conduct statistical analyses about law enforcement agencies. These data will have agency- and jurisdiction-specific identifiers that will permit public use in combination with other data files with similar identifiers.

The BJS-produced findings from the 2018 CSLLEA will be provided to the public in electronic format. These reports will be available on the BJS website as PDF files. BJS may also produce a web-based, data analysis tool for the 2018 CSLLEA to increase the ease with which the public can access information about specific agencies or types of agencies.

3. Efforts to Identify Duplication

Two of the personnel questions on the CSLLEA are also included on the LEMAS. Item 5 is considered an essential item for both collections—it serves as the measure of agency size on the CSLLEA for sampling purposes and for reporting out results on both collections. The number of full-time sworn equivalent is calculated from this item and allows BJS to pull a stratified sample from the CSLLEA for our various law enforcement surveys. For the LEMAS, we use FTE on our report tables. Due to fluctuations in staffing size year to year, our estimates rely on having an up-to-date FTE count. The other item that overlaps on the CSLLEA and LEMAS, is Item 6, which is being added to the CSLLEA for the first time in 2018. The LEMAS can only provide gender estimates for general purpose agencies. With the addition of sex on the CSLLEA, we can determine the proportion of female officers in special purpose agencies. No other data collection exists that can provide this statistic.

Based on our knowledge of the federal statistical system, in general, and law enforcement surveys in particular, BJS has determined that the 2018 CSLLEA includes measures of the number of law enforcement personnel that are also included in two ongoing surveys by other Federal agencies:

1. The FBI annually collects information from LEAs about the number and sex of sworn and nonsworn personnel as part of the “Number of Full-Time Law Enforcement Employees” (OMB No. 1110-0004) through the Police Employee Data collection.

2. The U.S. Census Bureau tabulates and publishes Equal Employment Opportunity (EEO) information on the sex, race and ethnicity of sworn and non-sworn employees of state and local governments who work in a protective service. This information is available for geographies that represent worksite and residence. This information has been based on the decennial census and more recently on the American Community Survey (OMB No. 0607-0810 & 0607-0936).

CSLLEA vs. FBI Police Employee Data

BJS identified four variables—the number of male (fully) sworn, male nonsworn, female (fully) sworn, and female nonsworn personnel—that are collected and reported by the FBI survey and by BJS in the CSLLEA and LEMAS surveys. The items about personnel in the FBI survey are collected in conjunction with annual data collections of hundreds of items about reported offenses and about assaults on law enforcement officers. The FBI uses these data to report on offense, arrest, and assault rates per sworn personnel.

In the five years (1992, 1996, 2000, 2004 and 2008) for which both the FBI survey and the BJS CSLLEA were conducted, the FBI collected data from 3,600 to 5,200 fewer agencies (24.9%) and reported about 100,000 fewer total personnel (10.0%). These differences are due in part to the different criteria for inclusion of agencies and personnel in the two surveys. The FBI survey is limited to personnel paid “with law enforcement funds” while the BJS surveys include all personnel regardless of what funds pay their salaries. In addition, the BJS survey captures all agencies that employ the equivalent (i.e., two part-time staff) of at least one full-time sworn personnel; while the FBI requires at least one full-time sworn staff member. Lastly, the FBI survey is limited to agencies that report to the FBI’s UCR program during a particular year (Reaves, 2011).

Personnel items included in the CSLLEA are used to provide the basis for computing the percentages of sworn personnel by law enforcement function. The CSLLEA also collects information about limited sworn personnel and part-time employees of law enforcement agencies, whereas the FBI does not.

To summarize, BJS and FBI data collections differ on several key measures:

• Definition of law enforcement officer that varies depending upon how the officer is funded at the agency

• Scope of agencies considered for inclusion in data collection efforts

• Data collection goals

• BJS includes additional demographic variables for sworn personnel

These design elements lead to differences in the estimated number of total sworn officers, which persist over time across various waves of data collection. The number of duplicate data collection items in the BJS and FBI data collection is small, and the information collected is necessary to meet the goals of each survey.

CSLLEA vs. EEO

The EEO tabulations provide national estimates on the number of sworn and non-sworn personnel involved in protective services for state and local governments. With the EEO tabulations it is not possible to disaggregate the law enforcement related job codes that may be subsumed under the “protective service” heading. This dataset also provides geographic rather than agency staffing estimates. Estimates are provided for location of employment or residence rather than the law enforcement agency. CSLLEA data reflect place of work rather than location of work or place of residence. The EEO data are not granular enough to disaggregate the number of sworn officers working in local law enforcement versus sheriff’s offices and they also do not provide a breakout by gender.

BJS provides national estimates of the number of law enforcement agencies and the numbers of fully sworn, limited sworn and non-sworn personnel. The CSLLEA is the only data source that provides gender of fully and limited sworn officers based on the employing agency rather than residence for general and special purpose agencies. This allows for national estimates at all jurisdiction levels: local, county, and state.

3. Efforts to Minimize Burden

Efforts to minimize burden have focused on three areas: extensive frame cleaning, instrument design and support services.

RTI has conducted extensive frame cleaning efforts to ensure that only agencies within the scope of the CSLLEA are included in the frame. This effort has involved reviewing publically accessible data, including agency and city websites, budgets, and meeting minutes of city councils, to determine which agencies are in-service, publically funded, and have at least one full-time equivalent sworn officers. This work minimizes burden on out-of-scope agencies and other government agencies that would need to respond for agencies that are closed or out of service.

BJS has reviewed each CSLLEA instrument and the historical use of these items to ensure that the survey content addresses only those issues needed to achieve the core goals described in Section 2. Furthermore, as described in Part B, Section 4, BJS is requesting clearance to conduct a series of cognitive interviews to ascertain whether respondents can provide the requested data (e.g., do the agencies track the necessary information) and to determine the most appropriate question strategies (e.g., instructions, terminology, question wording) to maximize comprehension and standardized interpretation across agencies.

Additionally, the 2018 CSLLEA survey has cut a significant number of items from the 2014 CSLLEA administration (7 items vs. 21 items). Almost 50% of law enforcement agencies in the U.S. have fewer than 10 personnel. Lack of time is a typical refusal reason given when an agency does not participate and this is more prominent in the smaller departments. The 2014 CSLLEA had a 1-hour burden and we estimate the 2018 CSLLEA to have a 30-minute burden since it more closely aligns with the 2008 instruments, which achieved 99% response rate.

We expect that most respondents will make use of the online version to complete the survey. Several web-based system functions will be in place to ease the burden of survey completion. RTI will utilize an intelligent log-in program for data collection, which will store agency information and responses, allowing for multi-session, non-sequential completion of the survey instrument. Since agencies, particularly the larger ones, may need to send the survey to multiple people within the organization or use different databases to find data, this will reduce the burden on them by facilitating data entry from different people. It will also reduce the burden by allowing them to pause data entry pending confirmation of information from others in the agency.

A help desk will be staffed during normal business hours (east coast time) and will be available to respondents through a toll-free number. Respondents may also receive a hard copy questionnaire, along with directions, by mail upon request. Additionally, respondents will be able to access a PDF version of the survey online, which can be printed. Once completed, this paper version of the survey can be used to enter data into the web-based survey instrument or can be returned via email, fax, or mail.

Since some respondents will complete the CSLLEA using hard copy forms, the questionnaire has undergone a structured, standardized instrument appraisal to evaluate the questions and form layout relative to the tasks required of respondents, specifically how respondents understand and respond to survey questions. One component of the review is to assess burden imposed by confusing terminology, unclear instructions, and difficulty accessing definitions and other information needed to answer questions. Results from this appraisal have helped refine the current hard copy questionnaire and the cognitive interview protocol.

3. Consequences of Less Frequent Collection

Based in part on recommendations from the National Research Council (NRC), BJS has determined that it is necessary to establish a more regular schedule of future surveys of law enforcement agencies. To this end, a significant portion of BJS’s law enforcement data collection efforts have been combined into the Law Enforcement Core Statistics (LECS), which is comprised of the CSLLEA, LEMAS core, and LEMAS topical supplements. These data collection efforts will now share a common alternating schedule that will serve to reduce burden and increase the timeliness of data collection. Other BJS law enforcement surveys that use the CSLLEA frame have been incorporated into the schedule in order to minimize burden on law enforcement agencies that are invited to participate in those surveys. Table 1 shows the data collection schedule for these projects.

Table 1. Data collection schedule for the Law Enforcement Statistics Unit, 2018-2024

Collection	Start of Data Collection
2018 CSLLEA	August 2018
SLEPS	June 2019
2020 LEMAS Core	August 2020
2020 SCLEA	August 2020
2020 CPFFCL	August 2020
2021 LEMAS Supplement (topic TBD)	August 2021
2022 CSLLEA	August 2022
2023 LEMAS Supplement (TBD)	August 2023
2024 LEMAS core	August 2024

Conducting multiple surveys in a single year may lead to lower response rates and result in less precise and biased estimates for key survey items. Under the LECS model and taking into consideration other key BJS collections, only one survey administration using the same sample will be administered per year. For example, the 2020 LEMAS Core and 2020 SCLEA use different sampling frames, therefore, the same law enforcement agencies will not be in these collections. While there are large agencies in the 2020 LEMAS Core that have crime labs that will be in the 2020 CPFFCL, the point of contact for the two surveys do not overlap. The CPFFCL survey is sent to lab administrators while the LEMAS Core survey is sent to the chief executive. The personnel filling in the surveys will not overlap. The goal is to reduce burden on agencies that will be selected with certainty for each collection. Furthermore, the proposed data collection schedule will allow for reliable indicators of officer staffing, and changes in staffing, every two years.

7. Special Circumstances Influencing Collection

No special circumstances have been identified for this project.

7. Adherence to 5 CFR 1320.6 and Outside Consultation

The research under this clearance is consistent with the guidelines in 5 CFR 1320.6. The 60-day notice for public commentary was published in the Federal Register, Volume 82, Number 220, pages 53527-53528 on Thursday, November 16, 2017 (Attachment 3). The 30-day notice for public commentary was published in the Federal Register, Volume 83, Number 33, pages 7079-7080, on Friday, February 16, 2018 (Attachment 4). In response to the 60-day notice, we received three requests for a copy of the instrument and one communication containing general comments on the importance of the collection.

7. Paying Respondents

Neither BJS nor RTI will provide any payment or gift of any type to respondents. Respondents will participate on a voluntary basis.

7. 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 CSLLEA represent institutional characteristics of publicly-administered law enforcement agencies. Information collected from these organizations is considered within the public domain. The fact that participation in this survey is voluntary and that information about individual agency responses will be available to the public is included on the first page of the survey instrument. However, BJS will not release the names, phone numbers or email of the actual persons responsible for completing the 2018 CSLLEA.

7. Justification for Sensitive Questions

No questions of a sensitive nature are proposed for the 2018 CSLLEA.

7. Estimate of Hour Burden

BJS has estimated the respondent burden for the proposed CSLLEA Cognitive Interview Task at 72 hours and for the proposed 2018 CSLLEA at 15,072 hours (Table 2). The estimate for the cognitive interviews is based on a 30-minute burden for the completion of the draft CSLLEA hard copy questionnaire and a 60-minute burden for participating in the cognitive interview.⁵ Forty-eight respondents will be cognitively interviewed; the total burden for this effort is therefore estimated at 72 hours.

The burden hours for full study implementation includes the average time required per respondent to complete the survey form. Also included is the average time devoted to contacting agencies beyond the survey collection either through verifying contact information during frame cleaning (if not available online) and/or direct follow-up contact after survey completion to resolve discrepancies or missing data. The estimated burden hours for the 2018 CSLLEA is 15,072.

Table 2. Estimated Burden Hours for the CSLLEA

Collection	Sample size	Average time to complete form	Average time pre- and post-survey	Total average time	Reporting hours
Cognitive interview	48	0.50	1.00	1.50	72
2018 CSLLEA	20,000	0.50	0.25	0.75	15,000
Total					15,072

7. Estimate of Respondent Cost

For the cognitive interviews, BJS anticipates that one or more persons per surveyed agency will spend time reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. However, only one person per agency is expected to participate in the cognitive interview. Whether the response is provided by one or by more than one person, the weighted average for the total burden for each agency is estimated to be 1 hour. Assuming a pay rate approximately equivalent to the GS-12 / 01 level ($72,168 per year), the estimated agency cost of employee time would be approximately $34.58 per hour. Forty-eight agencies will be asked to participate in this activity for 72 hours, therefore the total cost is estimated at $2,490.

Approximately 20,000 agencies will be asked to participate in the 2018 CSLLEA. Based on the estimated time burden per response and employee pay rate, the total respondent employee time cost burden is estimated at $518,700.

There are no anticipated costs to respondents beyond the employee time expended during completion of the survey instrument and addressed in above. Therefore, the total cost burden to respondents associated with this clearance request is $521,190.

7. Cost to the Federal Government

The 2018 CSLLEA is being developed and conducted under a multi-year cooperative agreement under the LECS program. Table 3 reflects the cost to administer the 2018 CSLLEA.

Table 3. Estimated costs for the 2018 CSLLEA

Category	Cost
BJS costs
Staff salaries
GS-13 Statistician (25%)	$25,000
GS-15 Supervisory Statistician (3%)	$4,000
GS-13 Editor (10%)	$10,000
Other Editorial Staff	$5,000
Front-Office Staff (GS-15 & Directors)	$2,000
Subtotal salaries	$46,000
Fringe benefits (28% of salaries)	$12,880
Subtotal: Salary & fringe	$58,880
Other administrative costs of salary & fringe (15%)	$8,832
Subtotal: BJS costs	$67,712
Data Collection Agent (RTI)
Personnel (including fringe)	$295,352
Travel	$0
Supplies	$0
Consultants/Contracts	$291,257
Other	$141,667
Total indirect	$270,274
Subtotal: Data Collection Agent Costs	$998,550
TOTAL COSTS	$1,066,262

7. Reason for Change in Burden

The CSLLEA was last conducted in 2014 with a total estimated burden of 18,075 hours. The total estimated time burden of the 2018 CSLLEA has decreased by 3,003 hours from the estimated burden for the 2014 CSLLEA. The reduction in burden is due to four specific changes to the collection and one change in the components included in the calculation.

• The 2014 estimate included 75 hours for federal law enforcement agencies to complete a 1 hour survey. There will be no data collected from federal agencies in the 2018 CSLLEA.

• The 2018 CSLLEA estimated burden includes 72 hours for cognitive interviewing of the draft survey instrument. There was no burden included for equivalent activities in the 2014 CSLLEA.

• The estimated time burden per respondent for the 2018 CSLLEA has been reduced by the elimination of several questions that had been asked on the 2014 CSLLEA. For initial survey administration, the reduction is 30 minutes per case.

• The population estimates for the 2014 and 2018 censuses are different: 18,000 and 20,000, respectively.

• To calculate the 2018 CSLLEA burden, 15 minutes per case was included for follow-up with respondents to clarify data discrepancies and obtain missing data. No burden for this contact had been included in the 2014 estimate.

7. Project Schedule and Publication/Analysis Plans

The cognitive interviewing for the 2018 CSLLEA is scheduled to begin immediately upon receipt of OMB approval, approximately the end of March 2018. Cognitive interviewing will last approximately 3 months. RTI will submit a report of the results to BJS in July 2018. At that time, BJS will make any necessary modifications to the instrument and advise OMB of the cognitive interviewing results in order to launch the full data collection in August 2018. It should be noted that any changes to the instrument will not increase respondent burden and will likely reduce the overall burden if items are removed.

The data collection for 2018 CSLLEA is scheduled to begin in August 2018. The data collection period is 9 months. The survey design calls for the initiation of data analyses including the assessment of nonresponse biases when the response rate hits 50%. This program anticipates a final response rate in excess of 95%. However, BJS would report national estimates based on response rates greater than 80%, if necessary, to be able to release published findings by early 2020. Table 4 contains the project schedule for the full data collection.

Table 4. Project Schedule

Stage	Type of contact	Date	Attachment number
CSLLEA informational website	All	Day 1	--
Survey invitation letter (with URL and login instructions), CSLLEA flyer, letter of support, POC form	All	Day 7	5-8
Email invitation (with URL and login instructions)	All	Day 14	9
Completion thank-you	All	Variable	16
First reminder email	Non-respondents	Day 44	10
Second reminder letter	Non-respondents	Day 65	11
Third reminder email	Non-respondents	Day 86	12
Telephone non-response contact	Non-respondents	Day 117	13
Fourth reminder email	Non-respondents	Day 138	14
End-of-study reminder	Non-respondents	Day 187	15
Close data collection	Non-respondents	Day 285	--
Analysis	N/A	Months 9-12	--
Reports	N/A	Months 12-15	--

7. Expiration Date Approval

The expiration date will be shown on the survey form.

7. Exceptions to the Certification Statement

BJS is not requesting an exception to the certification of this information collection.

1 A sworn officer with general arrest powers is a person formally authorized to make arrests for all crimes without geographic jurisdiction limitation while acting with the scope of explicit legal authority.

2 The LEAR development work was done under the 2014 CSLLEA clearance (OMB 1121-0346). The 2014 CSLLEA cover page was used to verify agency and point of contact information when agencies needed to be contacted directly. Most agency information was obtained through online searches.

3 See http://www.icpsr.umich.edu/icpsrweb/NACJD/biblio/series/00169/resources?collection=DATA&archive=ICPSR&sortBy=1

4 BJS’s cooperative agreement with RTI for the 2018 CSLLEA was the result of a competition (FY2015 Law Enforcement Core Statistics Program, Solicitation, BJS-2015-4248; see https://www.bjs.gov/content/pub/pdf/lecs15sol.pdf).

5 The burden estimate for the 2008 CSLLEA (OMB Number 1121-0240), which was comparable in length and content to the current 2018 CSLLEA, was also 30 minutes.