By David J. Roberts, Senior Program Manager, IACP Technology Center
ontemporary advances in law enforcement information management and technology are laying an important foundation for recent efforts by the Bureau of Justice Statistics (BJS), Office of Justice Programs, U.S. Department of Justice, to create more detailed, accurate, and actionable measures of crime, criminality, and justice system performance. The BJS recently convened a Crime Indicators Working Group (CIWG) comprising law enforcement and justice leaders to provide guidance in the development of crime indicators using police administrative record information and other data sources to provide a better picture of the crime problem in local jurisdictions and throughout the nation.
As one component of this broader effort to build more comprehensive crime indicators, the BJS is in the early stages of initiating a project to assess the potential of harvesting incident-based crime data from the operational records management systems of a statistically representative sample of law enforcement agencies. The goal is to augment current data collection efforts to provide more detailed information on the elements of reported crimes, the attributes of such incidents, the victims, the offenders, and the social context of the crime.1 This project is called the National Crime Statistics Exchange (NCS-X) project. Both CIWG and NCS-X will build upon established crime statistical reporting programs and leverage recent advances in law enforcement information technology.
Uniform Crime Reporting
The systematic national collection, reporting, and analysis of crime and arrest data began 85 years ago when the International Association of Chiefs of Police (IACP) created its Committee on Uniform Crime Records in 1927, published a Uniform Crime Reporting (UCR) manual in 1929, and began data collection from 400 law enforcement agencies, resulting in the publication of Uniform Crime Reports for the United States and Its Possessions in January 1930. Following congressional legislation authorizing the attorney general to collect crime information, the Federal Bureau of Investigation (FBI) assumed responsibility for the UCR in September 1930, by which time the number of reporting agencies had grown to more than 800. The number of UCR reporting agencies would double to 1,658 agencies by 1933 and currently stands at 14,009 law enforcement agencies reporting in 2011.2
Since its creation, the UCR and the FBI’s annual publication of Crime in the United States has helped inform and shape public opinion on crime and public safety. Police chiefs and sheriffs, like chief executives of any organization, are measured on results, which often means public assessment on a variety of factors including the level of crime in communities, the extent to which citizens feel safe or safer than they have in the past, and the ability of agencies to “clear” crimes through the arrest of suspected violators.
The UCR program is based on monthly submissions of aggregate crime and arrest reports from state, tribal, and local law enforcement agencies. Detailed offense information is collected on eight index offenses in the summary UCR program, but even among these offenses, reporting provisions mask or obscure a potentially substantial volume of crime as a consequence of the “hierarchy rule” and the “hotel rule.”3 Complex classification and counting rules and reporting artifacts associated with the traditional summary UCR program are known to have produced what has been referred to as the “dark figure” of crime.4
Recognizing the limitations inherent in aggregate reporting in the summary UCR program, the BJS and the FBI funded a three-phased UCR redesign program in 1982.5 The original BJS-funded study recommended a two-tiered implementation strategy for unit-record (or incident-based) reporting nationally.6 Although the Blueprint recommended a two-tiered reporting structure for incident-based reporting, state UCR programs and the FBI determined to implement a single, nationwide, incident-based reporting program—the National Incident-Based Reporting System (NIBRS).
The National Incident-Based Reporting System (NIBRS)
The NIBRS expanded the range of offenses reported by the police and collects more detailed information on crime incidents. It collects offense and arrest data on 22 crime categories, spanning 46 offenses (compared to the 8 UCR index offenses), and additional offenses for which only arrest information is reported. The NIBRS collects 53 specific data elements from each reporting law enforcement agency, including multiple offenses within an incident; the location of the incident (for example, a bar, a motel, or a residence); characteristics of victims and offenders; relationships between victims and offenders; and the nature and scope of injury or dollar loss incurred in the incident. The system eliminates the need for the “hierarchy rule” because multiple types of crimes can be reported within a single incident, and it collects an expanded array of attributes involved in the commission of offenses, including whether the offender is suspected of using alcohol, drugs, narcotics, a computer in the commission of the offense, or a combination of these, and whether the arrestee was armed with a weapon.
While substantial progress has been made in the implementation of the NIBRS since the publication of the data reporting specifications by the FBI in 1988 and subsequently, only 43 percent of UCR reporting agencies, covering 28 percent of the U.S. population, are currently NIBRS certified.7 As a consequence, the NIBRS cannot be effectively used to provide accurate national estimates of crime and law enforcement response, which hinders federal efforts to assist state and local agencies in developing crime control policies and programs and impedes the production of national studies of the impact of crime, criminality, and law enforcement practices.
A joint BJS/FBI study of impediments to NIBRS implementation in 1997 found that costs associated with local agencies altering existing forms, training personnel in NIBRS data collection requirements, and revisions or updates to automated records management systems represented a significant barrier to NIBRS implementation. In addition, little attention had been focused on demonstrating the operational value of incident-based data for local crime analysis, research, resource deployment, or performance management.8
Advances in Law Enforcement Information Technology
Local police departments have substantially expanded their adoption of automation in the years since completion of the study of impediments to NIBRS implementation. Nearly two-thirds (65 percent) of even the smallest agencies (that is, those serving populations of less than 10,000) reported using automated records management systems in 2007.9 Local agencies also reported substantially greater use of in-field computers from 1997 to 2007, with 99 percent of large agencies (that is, those serving populations of 250,000 or more) reporting use of in-field computers, and even the smallest police departments usage climbed from 20 percent in 1997 to 50 percent by 2007.10 Recent advances in mobile technologies, tablet devices, and cloud-based computing portend even greater adoption of automation among agencies of all sizes.
The efforts of the BJS to expand the range and scope of crime indicators might well reflect the approach government has taken in developing leading economic indicators.11 Building leading crime and criminal justice indicators can help establish comprehensive measures assessing the health of our communities, the safety of our citizens, the operations of our justice agencies, and the performance of our justice system as a whole. As agencies implement automated records management and analytic systems to meet their local strategic planning and tactical deployment priorities, they are likely to build a crucial technical foundation that will enable them to substantively contribute to the development of new crime indicators and to participate in the National Crime Statistics Exchange program. ♦
1The National Crime Statistics Exchange (NCS-X) Project, Phase I, is described by the BJS in its solicitation, which was released on May 30, 2012. See http://www.bjs.gov/content/pub/pdf/ncse12_sol.pdf (accessed January 9, 2012). The project was awarded to a consortium of organizations, including Research Triangle Institute, the IACP, PERF, the IJIS Institute, and SEARCH.
2See Eugene C. Poggio et al., Blueprint for the Future of the Uniform Crime Reporting Program: Final Report of the UCR Study (Washington, D.C.: Bureau of Justice Statistics and Federal Bureau of Investigation, May 1985), 21, https://www.ncjrs.gov/pdffiles1/bjs/98348.pdf (accessed January 10, 2013); and Federal Bureau of Investigation, Uniform Crime Reporting Handbook (Clarksburg, W.Va.: FBI, 2004), 2, http://www2.fbi.gov/ucr/handbook/ucrhandbook04.pdf (accessed January 10, 2013). Preliminary 2011 UCR figures are available at http://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2011/preliminary-annual-ucr-jan-dec-2011 (accessed January 10, 2013).
3The initial crime index consisted of murder and non-negligent manslaughter, forcible rape, robbery, aggravated assault, burglary, larceny-theft of $50 or more, and motor vehicle theft. An eighth offense—arson—was added to the index by congressional mandate in 1979. The “hierarchy rule” of the UCR program limits the reporting of offenses to the single most serious offense in a series of offenses. For person offenses, however, one offense is scored for each person victim, regardless of the number of victims. Nevertheless, if multiple person offenses are committed against a single individual, only the most serious of those offenses will be reported in UCR (UCR Handbook, 10–12). Burglaries of multiple hotel rooms are scored as a single burglary (Ibid., 28–29).
4Albert D. Biderman, and Albert J. Reiss Jr., “On Exploring the ‘Dark Figure’ of Crime,” The ANNALS of the American Academy of Political and Social Science, 374, no. 1 (November 1967): 1–15, http://deepblue.lib.umich.edu/bitstream/2027.42/67517/2/10.1177_000271626737400102.pdf (accessed January 10, 2013). Biderman and Reiss refer to the “dark figure” of crime as “occurrences that by some criteria are called crime yet that are not registered in the statistics of whatever agency was the source of the data being used” (Ibid., 2). For a general discussion regarding the limitations of the UCR program, see Albert J. Reiss Jr., “Problems in the Documentation of Crime,” in A. L. Guenther, ed., Criminal Behavior and Social Systems (Chicago: Rand McNally, 1976), 111–130.
5In 1982, the BJS funded Abt Associates to examine the UCR program, its history, objectives, data elements, and relationships with other systems (see UCR Blueprint). In 1984, the FBI began the second phase of the project, the goal of which was to identify available options and recommend changes. In 1988, the FBI’s third phase produced specifications for data collection and submission and system implementation.
6See UCR Blueprint, 43–48. This strategy contemplated that only a sample of perhaps 3 percent to 7 percent of law enforcement agencies nationwide would report comprehensive incident-based data, consistent with the reporting requirements of today’s NIBRS, then called “Level II reporting.” The remaining 93 percent to 97 percent of law enforcement agencies would report incident-based data, but in a much more abbreviated format, focusing on Part I offenses with only a limited range of victim, offender, and more detailed incident data; this would be “Level I reporting.” Arrest data for both Part I and Part II crimes were to have been captured in both levels, with linkages to cleared offenses. In spite of these recommendations, the law enforcement community elected full NIBRS implementation, effectively endorsing Level II reporting for every agency.
7FBI, NIBRS Volume 1: Data Collection Guidelines (Washington, D.C.: FBI, August 2000); FBI, NIBRS Volume 2: Data Submission Specifications (Washington, D.C.: FBI, November 2001); FBI, NIBRS Volume 4: Error Message Manual (Washington, D.C.: FBI, December 1999); and FBI, NIBRS Conversion of NIBRS Data to Summary Data (Washington, D.C.: FBI, December 2009).
8David J. Roberts, Implementing the National Incident-Based Reporting System: A Project Status Report (Washington,
D.C.: Bureau of Justice Statistics, July 1997), http://www.bjs.gov/content/pub/pdf/INIBRS.PDF (accessed January 10, 2013). The study concluded
[A]gencies currently have automated, incident-based systems that effectively meet their operational needs, but fail to capture the necessary data in an appropriate format for NIBRS reporting. In these agencies, NIBRS represents changing their recordkeeping systems in order to generate statistical data for use at State and Federal levels, without any perceived benefit at the local level. For these agencies, the costs of implementing changes in reporting practices to make their systems NIBRS-compliant (for example, revising offense reporting forms, department-wide training, and software reprogramming), compounded by concerns over the impact NIBRS will have on the department's reported crime rate and a lack of understanding on how the data will be used at State and Federal levels, create formidable impediments to NIBRS implementation.
If NIBRS is to be implemented broadly, its purpose must be relevant to the operational records management systems of the local law enforcement agencies responsible for contributing the data, and it must demonstrate utility in State and Federal analyses of the data. In addition, it must not represent an undue burden on the law enforcement officers who capture the data at the street level. Technologies that assist in the collection and recording of incident data should be encouraged and funding assistance should be made available to law enforcement agencies to facilitate their adoption of NIBRS-compliant systems (Ibid., 13).
9Brian Reaves, Local Police Departments, 2007 (Washington, D.C.: BJS, December 2010), 22, http://bjs.ojp.usdoj.gov/content/pub/pdf/lpd07.pdf (accessed January 10, 2013).
11The Leading Economic Indicators (LEI) for the United States comprise the following 10 components: (1) average weekly hours, manufacturing; (2) average weekly initial claims for unemployment insurance; (3) manufacturers’ new orders, consumer goods and materials; (4) ISM Index of New Orders; (5) manufacturer’s new orders, nondefense capital goods excluding aircraft orders; (6) building permits, new private housing units; (7) stock prices, 500 common stocks; (8) Leading Credit Index, (9) interest rate spread, 10-year Treasury bonds less federal funds; and (10) average consumer expectations for business conditions. See “Global Business Cycle Indicators,” The Conference Board, http://www.conference-board.org/data/bcicountry.cfm?cid=1 (accessed January 10, 2013).
Please cite as:
David J. Roberts, "Advances in Law Enforcement Information Technology Will Enable More Accurate, Actionable Analysis," Technology Talk, The Police Chief 80 (February 2013): 58–59.