National Academies Press: OpenBook

Modernizing Crime Statistics: Report 1: Defining and Classifying Crime (2016)

Chapter: 5 Proposed Classification of Crime for Statistical Purposes

« Previous: 4 Historical and Extant Classifications of Crime
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

– 5 –

image

Proposed Classification of Crime for Statistical Purposes

BUILDING FROM AND WEIGHING ALL of the preceding material—the range of crime-related information currently gathered (or that could be gathered) in existing data collections, the user and stakeholder input obtained at the panel’s workshop-style sessions, and the examples of past and current crime classification schemes—we arrive at the main purpose and sole formal recommendation in this first report. Below, we summarize a set of design principles and objectives (Section 5.1) before outlining our suggested classification in Section 5.2 and the way it differs from current classifications in Section 5.2.3. Due to its size, we present a “short” form in-line with Section 5.2.1 along with a corresponding set of attributes (Section 5.2.2); the full “long” form of the classification, with definitions and example of specific inclusions for each category is in one of the appendixes to the entire report, Appendix D. We close in Section 5.3 with a short preview of “next steps” and issues awaiting consideration in our second, final report.

5.1 OBJECTIVES FOR A MODERN CRIME CLASSIFICATION

5.1.1 Design Principles

We identified four basic principles to guide our development of a specific, modern classification of crime in the United States. Our panel’s charge (Appendix A) directly suggests some of these, including that (1) the suggested classification should not be limited to current crime statistics’ traditional focus on

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

violent or street crime, and should encompass new and emerging crime types—certainly, new crime types that have developed since the onset of the Uniform Crime Reporting (UCR) Program in 1929. Moreover, our charge compels us to consider topics that are not currently in the task set of any data collection maintained by the Bureau of Justice Statistics (BJS), the Federal Bureau of Investigation (FBI), or perhaps any agency. At the same time, we do face practical and logical limits in deeming various socially unacceptable behaviors as “crime”—which is to say that it would be inappropriate for the classification to include behaviors or phenomena that are nowhere deemed “criminal” acts in the nation.

Two additional basic design principles—on which we settled early in considering the problem—give structure and shape to our proposed classification:

  • (2) The suggested classification should satisfy all the properties of a fully realized classification for statistical purposes: A statistical classification of crime would be one meant to provide information on the structure and extent of crime, rather than just be an amalgam of topics related to crime. Summarizing for a United Nations-sponsored conference, Hancock (2013:4) articulates a set of “essential components of a statistical classification”1:
    • Maintenance of a consistent conceptual basis throughout;
    • Adoption of categories that are mutually exclusive and exhaustive, which is to say that a specific element/crime should correspond to only one category and that set of categories should span the whole terrain of “crime”;
    • Adoption of either a flat (basic listing) or hierarchic (subcategories nested within each other) structure; and
    • Definitions that are clear, unambiguous, and measurable, and which define the content of each category.

    For too long, UCR-based crime statistics have followed a “classification” only in the sense that events or incidents are labeled as crimes; classification becomes the cognitive exercise of determining which label, from a loosely structured list, is most applicable.

    At this point, two related points should be made very clear. First, as described in Section 1.2, we take the criminal offense as the basic unit for classification purposes. Accordingly, in the parlance of a statistical classification, we seek to partition “crime” into offense categories, such that each individual offense corresponds to one and only one category. Second, by preferring that the classification have a hierarchical structure, we mean only that the finest-grained offense categories can be “rolled

__________________

1In this work, Hancock (2013) summarizes and extends earlier comments on the task of classification by Hoffman and Chamie (1999).

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  • up” to meaningful, higher-level aggregates (e.g., detailed variants of trafficking in persons for specific purposes being nested within a higher-level trafficking heading), and vice versa. It is not to suggest anything hard-set about the ordering of offenses within the classification, such as their severity or importance. The relationship between these two points is driven by the assumption that the operational unit of analysis for eventual data collection will be crime incidents that may consist of more than one offenses—and we oppose the imposition of a UCR-type “Hierarchy Rule” that would try to compress a broader criminal incident into something where only one offense is permitted/collected. Certainly, one could argue for an NCVS-type algorithm that may flag one of several offenses within a given incident as the “most serious” in some sense—but it is inherently wasteful to discard valuable information through imposition of an arbitrary Hierarchy Rule.

  • (3) The suggested classification should follow—to the greatest extent possible—an attribute-based approach, yet should also be a hybrid with a code- or definition-based approach due to the nature of the topic: This is a complicated principle that arises from reconciling two fundamentally opposite impulses. The first—and possibly the chief aim of a proposed new classification of crime—is a reflection of the near-century span it has taken to embark on such a reassessment of national crime statistics: The primary objective is to accomplish flexibility in content and coverage. Central to achieving that aim is relaxing strict adherence to the precise wording of state criminal or penal codes in favor of more generalizable, behavior-based definitions. We think it useful to express this sentiment as a formal finding, for clarity:

    Conclusion 5.1: The definitions and concepts in the current U.S. crime statistics system were developed primarily from categorization of statutory language, which varies by jurisdiction. Reliance on statutory language is inflexible and not comprehensive, and it is unduly focused on limited input sources (reports from police/law enforcement or individual victims).

    Further, as a sign of a clear “break” from the strictures of past categorizations (and a preventative of another 90 years passing before U.S. crime statistics are reassessed in a comprehensive way), we also suggest the need for the flexibility that comes from continual monitoring of the field and from regular, periodic review:

    Conclusion 5.2: “Crime” continues to evolve and take different shapes. Accordingly, there is a need for an expansive

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

framework for crime classification that is amenable to periodic revision.

To ensure this flexibility, we find much to admire in attribute-based classification schemes, akin to that described in Section 4.1.2—plans that focus on the behavior/action of a criminal offense while simultaneously gathering critical contextual information that could support reanalysis (and eventual reclassification using revised standards, if need be).

Yet the second, contradictory impulse is also very strong, and derives from the specific topic matter at hand: In working with “crime,” a break from the meaning (if not the specific language) of criminal law can only be taken so far. Behaviors may be very “bad” or socially undesirable but, as noted in beginning Section 1.2, the simplest distinction between “crime” and general “bad” behavior is that “crime” is that which is unlawful. Accordingly, federal and state criminal codes remain an essential reference and anchor in defining and classifying crime. For at least baseline concordance with the actions for which law enforcement officers can make arrests and that the justice system can pursue charges, it would be useful for the main thrust of the classification to resemble a list of known, identifiable offenses.

Hence, reconciling these two concepts, we arrive at the statement of the design principle above and the suggestion of a hybrid approach akin to that used in the International Classification of Crime for Statistical Purposes (Section 4.1.5). This hybrid approach combines, as coequal companions, a classical listing of criminal offense categories (based on behavioral definitions, and invoking their applicability to unlawful behaviors as appropriate to account for state/local variation in underlying law) and a list of attributes (or contextual variables) to be collected about each offense or incident. While it can certainly be argued that this hybrid structure adds unnecessary complexity, we think that its benefits in providing flexibility and range outweigh such added complexity. (Indeed, the direct counterargument has weight—recording values of objective, simply measurable attributes or contextual variables is arguably less burdensome and complex than the cognitive task of absorbing an entire set of definitions to find the right individual offense category.) Flexibility in building out a classification comes from negotiating what kinds of phenomena are best handled as specific, defined offenses in the classification table or left as more general offenses (but analyzed with reference to associated attribute values). Two wide-ranging examples that we will discuss below, but that are useful to illustrate the concept here, include near-fatal shooting of another person (on one hand) and shoplifting (on the other). Both of those offenses could be entered, and specifically defined, as individual specific categories in the classification

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

table; likewise, they could be logged as assault + [weapon use attribute = firearm] or theft + [incident location attribute = retail store]. In any event, collection of the attribute information along with the offense classification permits more detailed analysis—and future revisions could either fold specific offense-attribute combinations into the classification table or dissolve them, as judged appropriate.

To these design principles, we think it useful to add one related to purpose—one that certainly hearkens back to the spirit of the nation’s earliest crime statistics, and one that was touched upon by virtually every user or constituency we consulted. This principle is that (4) the suggested classification should be designed in order to enable and promote comparisons between jurisdictions, between periods of time, and across state and national boundaries. For years, volumes of Crime in the United States have directly admonished UCR data users against making direct comparison of crime levels across jurisdictions. Yet facility for comparability of information on crime is certainly one of the highest—if not the highest—desire for crime statistics, not to belittle or condemn jurisdictions that are worse off crime-wise than one’s own but to be able to sort out which anti-crime approaches or interventions may work (and which may not). The desire to assess relative crime risk or offending levels between different neighborhoods, precincts, or places is real, and should be a major consideration. But, while high-level aggregates like states or the nation as a whole may be of little direct value for many domestic crime statistics users, easing cross-national comparison of types of crimes is also a useful goal. This is particularly the case when attention shifts toward white-collar offenses rather than violent “street” crime—and especially when starting to get a sense of the levels and extent of cybercrime, given the way that computer involvement can supersede traditional physical geography. There is certainly a mass of complicated analytical issues involved with making cross-national comparison that would require detailed studies in their own right, but such comparison certainly is not enabled at all if definitions do not have some modicum of standardization.

5.1.2 Objectives of Crime Classification

These four basic principles, together and separately, raise a number of corollaries that merit separate explication. These related, objectives include the following:

  • Balance the desire to add and focus on “new” crimes with statistical series continuity, retaining categories and definitions that still “work:” Suggesting a complete upending of all concepts for measuring crime—just for the sake of change—would be a disservice. The task has to be approached with some care to avoid hubris—after all, some of the definitions and
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  • concepts laid down with the UCR program in 1929 have endured in current crime statistics because they are still useful, and because they are salient in the public eye today just as they were then. We also recognize that some change in crime measurement concepts has taken place over the decades—with great difficulty—and this work should not be summarily dismissed. Perhaps most notably, years of debate and deliberation went into the 2011 update and revision of the definition of rape, and the current UCR program is still in the process of “bridging” the old and new definitions. Surely, as a predecessor National Research Council (2014) panel explored in greater detail, the measurement of rape and sexual assault can be improved, but it would not be wise to summarily overhaul a bitterly fought-for definition just to make all things “new.” (As we discuss below, the definition—and all others—should not be invariant for another several decades, and should be changed if it is demonstrably in need of refinement.) Another force for maintaining some continuity in categories and definitions is consistent with the constraint we noted earlier (and will again), that—in the field of crime—the language of federal and state criminal law is a necessary reference and anchor. Specifically, we have to recognize that some of the topics and definitions in current U.S. crime statistics are explicitly spelled out by federal law—law that can and should be revised over time, too, but that is nonetheless a working constraint.

  • Establish a classification that covers, or at least anticipates collection of, crime incidences involving a range of different actors/units and modes of collection: Reviewing BJS’s ambitious data collection portfolio, our predecessor National Research Council (2009a) panel noted that the parallel-track collection of (some) data about the juvenile justice system and the dearth of general information on white-collar and business-involved crime were both major gaps in an otherwise strong portfolio. Both gaps still loom large, even when expanding view outside of BJS’s own holdings, and so a suggested classification for “crime in the U.S.” should be applicable to juveniles as well as adults, and to businesses or establishments as victims (or offenders) as well as persons. In terms of the mode of collection, we think it safe to assume that both local law enforcement reports of crimes known to the police and household survey measures of victimization will continue to be a major part of crime measurement going forward, but not the only sources. We are obliged by our charge to consider “nontraditional” types of crime, and correspondingly must envision a role for nontraditional actors in the crime measurement system—private companies, credit agencies, nongovernmental agencies, private police/security firms, and other federal statistical agencies—for which the “standard” data collection modes might not be most apt.
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  • Reflect and balance user needs by resolving known issues with existing crime measurement systems: Perhaps most notably, as we will discuss in more detail below, the distinction between “aggravated assault” and “simple assault” has always been—and remains—murky. Over the decades, though, this definitional ambiguity has become more than just a source of measurement inaccuracy—it has become a recurring controversy, with misclassification (deliberate or not) of aggravated assaults (more visible and scrutinized, Part I offenses) as simple assaults (less visible, Part II) commonly being the root problem when police departments are accused of swaying crime statistics to look better. A new measurement system, and classification system, that actually permits construction of statistics on assaults where injury is inflicted—or where a firearm is actually discharged—would be beneficial to a broad swath of users. Similarly, in the absence of fuller participation in the National Incident-Based Reporting System (NIBRS), it is also the case that users and practitioners of crime statistics lack the capacity to drill down broader offense types by key, policy-relevant strata (e.g., within-household or domestic assaults). A new classification, and resulting system, would ideally resolve that problem and facilitate more extensive analyses of crime statistics.
  • Establish a classification that is current, relevant, and capable of lasting for several years—but one that can and should be revisited and updated on a regular basis: This point may occur low in this listing, but that placement should not be construed to minimize its importance. The basic charge to our panel to step back and consider the crime statistics infrastructure as a whole is, now, effectively a once-in-a-lifetime opportunity (not having been done extensively since 1929)—but it should not be so, going forward. It is critical to point out that our suggested classification is intended as a start but not an end, and a regular feedback and refinement/revision routine should be agreed upon in setting up a new crime measurement system. Update mechanisms can be intricate—for the International Classification of Diseases (ICD), World Health Organization (2011) member states reached agreement in the late 1990s for a multilayer updating scheme enabling minor updates annually, more substantive updates (as needed) every three years, and (the goal of a) thorough revision every 10 years—and can also evolve over time.
  • Anticipate that the raw incident count may not be the only, or even the most ideal, metric that may apply: Classification is fundamentally an exercise in grouping things, and certainly the basic count of events that fall into a particular category is the most natural product when data are collected according to a classification. But raw counts alone do not adequately describe the impact or effect of crime. For some crime types, an estimate of financial harm/damage may be a much more salient (and important, to various data constituencies) measure than a basic tally.
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  • Anticipate data integration in the future. We already alluded to this in summarizing our charge’s mandate that we look beyond the current data collections of BJS and the FBI, but the point can be made more expansively still. For example, the potential availability of more richly geocoded data on crime—and with them the capacity for linkages to other sources, not just to demographic and social data but to feeds from automated sensors—may be of great use in deriving risk assessments for small local areas. Likewise, information culled from online sources has already been used by police to determine on-street activity, and may ultimately prove useful in refined statistical collection as well.

Finally, it might not be a principle of construction, but it is nonetheless important to state clearly one thing that the proposed classification is not. The classification we outline below is not a new list of codes to replace in full—immediately—the crimes measured by any of the current U.S. crime statistics programs, whether the UCR (including NIBRS), the National Crime Victimization Survey (NCVS), or anything else. It is not yet ready for direct implementation, and is a preparatory and partial step in that direction, but much work remains in our final report to work with this classification, suggest which categories must (and which should not) be implemented immediately, and which would be good to know about (but perhaps not immediately available or essential).

5.2 RECOMMENDED CLASSIFICATION OF CRIME

Weighing these principles and objectives, reviewing the demands of crime statistics users and stakeholders, and considering the existing examples of crime classification plans, we find one uniquely promising alternative:

Conclusion 5.3: The International Classification of Crime for Statistical Purposes (ICCS) framework, proposed and maintained by the United National Office on Drugs and Crime (UNODC), meets the desired criteria for a modern crime classification, and the use of shared, international frameworks enables studies of transjurisdictional and locationless crime.

To be clear, we argue that the ICCS provides a strong base on which to construct a modern classification of criminal offenses, which would in turn be used to develop a modern set of crime indicators for the United States. The ICCS is not a classification that can or should be applied directly, exactly as is, without any customization based on the user/stakeholder needs described in Chapter 3—but, by the same token, extensive departure from the ICCS structure would undermine the classification’s value for comparative purposes.

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

There are highly admirable elements of other exemplar classification systems: For instance, the state of Victoria’s fresh-eyes approach to building from an already-strong base in the Australia and New Zealand Standard Offense Classification turned up some desirable alternative structuring, such as the handling of drug offenses. The pure attribute-based classification prototyped by the SEARCH Group in 1975 remains a striking, ahead-of-its-time work, and gives hope that very flexible classifications can be given workable, operational forms in today’s computing environment. And the Irish Crime Classification System’s “condensed” formulation plays loosely with pure hierarchical layering but, doing so in the interest of distilling information of peak public interest, is an elegant solution to conveying a large quantity of information in approachable form. In working through revisions to the ICCS in stating our recommended classification, we also seek to borrow useful elements from these other classifications.

That said, we found the ICCS an ideal base to work with because it corresponds closely with our desired principles. In particular, its explicit pairing of a fairly detailed classification tree or listing of categories with an extensive set of attributes is the closest fit to a hybrid approach (i.e., the appearance of a list, but with vast added flexibility through the attributes). The ICCS was also an ideal base because of its emergence as the product of several years of expert collaboration by numerous national statistical offices and crime statistics producers—solid work that warrants consideration and adaptation, rather than starting from scratch. It is also to the ICCS’s credit that it has won the approval of the various United Nations commissions to which it has been submitted—not because full compliance with international standards is a paramount goal, but because the capacity for international comparison of different crime types is ultimately a good thing, in our assessment. It is also to the ICCS’ credit that it was constructed in concert and with involvement from Australia, Ireland, and other nations from whose crime classifications we have drawn ideas and inspiration.

Accordingly, much of what follows and which is spelled out in more detail in Appendix D is directly derived from and comports with Version 1.0 of the ICCS as promulgated by the UNODC. We acknowledge the UNODC’s role and its coordination of expert work groups to develop the ICCS, and properly give them credit for much hard work in structuring a great mass of material. To be clear, though, we have made some revisions to both the base classification and to the attribute list; we will return to these differences in Section 5.2.3 after stating our suggested classification (in brief form) and attribute list.

Recommendation 5.1: The attribute-based classification of offenses described in brief in Sections 5.2.15.2.2 and in detail in Appendix D should be used as an initial framework for

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

developing modern statistical measures of crime in the United States.

5.2.1 Suggested Classification of Crime for Statistical Purposes (Short Version)

Our suggested classification of crime is premised upon 11 first-level categories, which provide a general structure to the classification:

  1. Acts leading to death or intending to cause death
  2. Acts causing harm or intending to cause harm to the person
  3. Injurious acts of a sexual nature
  4. Acts of violence or threatened violence against a person that involve property
  5. Acts against property only
  6. Acts involving controlled substances
  7. Acts involving fraud, deception, or corruption
  8. Acts against public order and authority
  9. Acts against public safety and national security
  10. Acts against the natural environment or against animals
  11. Other criminal acts not elsewhere classified

These are essentially identical to the 11 first-level offenses defined in the ICCS, save for the substantive difference that we reverse the order of the elements in the label for category 4. The ICCS’s version, “Acts against property involving violence or threat against a person,” casts robbery and other crimes under the heading as property-focused crimes involving an element of violence. However, particularly given its placement in the list (following three broad headings of violent crime and before the purely property-based crime category), we tend to think of the category as violent crimes that involve property.

Our classification sets forth 71 offenses at the second level of the hierarchy (denoted X.X in the listings), relative to 62 such second-level categories in the ICCS. We subdivide 38 of these second-level crimes into 131 third-level entries (denoted X.X.X); hence, fully expanding our classification to the third level (treating unsplit second-level offenses comparably to defined X.X.X categories) would yield a list of 164 third-level crime categories. (The comparable total for the ICCS is 165 third-level categories.) Similarly, we split 13 third-level offenses into 38 specific fourth-level (X.X.X.X) categories, meaning that a fully expanded fourth-level listing from our classification would have 189 listings. (The ICCS would yield 230 fourth-level offenses, fully extended.)

  1. Acts leading to death or intending to cause death

    1.1. Murder and intentional homicide

    1.2. Nonintentional homicide

    1.2.1. Nonnegligent manslaughter

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  1. 1.2.2. Negligent manslaughter

    1.2.2.1 Vehicular manslaughter;

    1.2.2.2 Nonvehicular manslaughter

    1.3. Assisting or instigating suicide

    1.3.1. Unlawful2 assisted suicide

    1.3.2. Other acts leading to death by suicide

    1.4. Unlawful euthanasia

    1.5. Unlawful feticide

    1.6. Unlawful killing associated with armed conflict

    1.7. Other unlawful acts leading to death

  2. Acts causing harm or intending to cause harm to the person

    2.1. Assault

    2.1.1. Serious assault involving shooting or discharge of a firearm

    2.1.2. Serious assault by means other than discharge of a firearm

    2.1.3. Minor assault

    2.2. Threat

    2.2.1. Serious threat through shooting or discharge of a firearm

    2.2.2. Serious threat through the display or pointing of a firearm

    2.2.3. Serious threat by means other than firearm

    2.2.4. Minor threat

    2.2.5. Other acts causing or threatening injury or harm

    2.3. Acts against liberty

    2.3.1. Abduction of a minor

    2.3.1.1 Parental abduction; 2.3.1.2 Abduction by a family member; 2.3.1.3 Abduction by a legal guardian; 2.3.1.4 Abduction by another person

    2.3.2. Kidnapping for ransom

    2.3.3. Illegal restraint

    2.3.4. Hijacking

    2.3.5. Illegal adoption

    2.3.6. Forced marriage

    2.3.7. Other deprivation of liberty or acts against liberty

    2.4. Slavery and exploitation

    2.4.1. Slavery and involuntary servitude

    2.4.2. Forced labor

    2.4.2.1 Forced labor for domestic services; 2.4.2.2 Forced labor for industry services; 2.4.2.3 Other forced labor

__________________

2In most cases in this classification, we omit the term “unlawful” in the name of the offense, though we commonly use it in the detailed definitions in Appendix D in order to explicitly acknowledge that the underlying behavior may not be deemed criminal in all states and jurisdictions. However, we think it appropriate to include “unlawful” in the title of those offenses involving death but that vary by statute and legal authority.

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  1. 2.4.3. Other acts of slavery and exploitation

    2.5. Trafficking in persons

    2.5.1. Trafficking in persons for sexual exploitation

    2.5.2. Trafficking in persons for forced labor or services

    2.5.3. Trafficking in persons for organ removal

    2.5.4. Trafficking in persons for other purposes

    2.6. Coercion

    2.6.1. Extortion or blackmail

    2.6.2. Other acts of coercion

    2.7. Negligent acts

    2.7.1. Negligence in situations of persons under care

    2.7.1.1 Negligence in situations of children under care; 2.7.1.2 Negligence in situations of other dependent persons under care; 2.7.1.3 Other negligence in situations of persons under care

    2.7.2. Professional negligence

    2.7.3. Negligence related to driving a vehicle

    2.7.4. Other acts of negligence

    2.8. Dangerous acts

    2.8.1. Acts that endanger health of another person

    2.8.2. Operation of a vehicle under the influence of alcohol or other psychoactive substances

    2.8.3. Other dangerous acts leading to injury

    2.9. Acts intended to induce fear or emotional distress

    2.9.1. Harassment

    2.9.2. Stalking

    2.9.3. Other acts intended to induce fear or emotional distress

    2.10. Defamation

    2.11. Discrimination

    2.12. Acts that trespass against the person

    2.12.1. Invasion of privacy

    2.12.2. Other acts that trespass against the person

    2.13. Other acts causing harm or intending to cause harm to the person

  2. Injurious acts of a sexual nature

    3.1. Rape

    3.1.1. Rape with force

    3.1.2. Rape without force

    3.1.3. Rape involving inability to express consent or nonconsent

    3.1.4. Threat of rape

    3.2. Sexual assault

    3.2.1. Physical sexual assault

    3.2.2. Threat of a sexual nature

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  1. 3.3. Sexual violations of a nonphysical nature

    3.4. Sexual exploitation of adults

    3.5. Sexual exploitation of children

    3.5.1. Child pornography

    3.5.2. Child prostitution, production and provision

    3.5.3. Child prostitution, procurement

    3.5.4. Other sexual exploitation of children

    3.6. Other injurious acts of a sexual nature

  2. Acts of violence or threatened violence against a person that involve property

    4.1. Robbery

    4.1.1. Robbery from the person

    4.1.2. Carjacking/robbery of a car or vehicle

    4.1.3. Robbery of valuables or goods in transit

    4.1.4. Robbery of an establishment or institution

    4.1.5. Robbery of livestock

    4.1.6. Other acts of robbery

    4.2. Terroristic or disruptive threats to buildings or critical infrastructure

    4.3. Other acts against property involving violence against a person

  3. Acts against property only

    5.1. Burglary

    5.1.1. Burglary of business premises

    5.1.2. Burglary of residential/private premises

    5.1.3. Burglary of public premises

    5.1.4. Other acts of burglary

    5.2. Theft

    5.2.1. Theft of a motorized vehicle or parts thereof 5.2.1.1 Theft of a motor vehicle; 5.2.1.2 Illegal use of a motor vehicle; 5.2.1.3 Theft of parts of a motor vehicle; 5.2.1.4 Other theft of a motorized vehicle or parts thereof

    5.2.2. Theft of personal property

    5.2.2.1 Theft of personal property from a person; 5.2.2.2 Theft of personal property from a vehicle; 5.2.2.3 Other theft of personal property

    5.2.3. Theft from business or other nonpublic organization

    5.2.4. Theft of public property

    5.2.5. Theft of livestock

    5.2.6. Theft of services

    5.2.7. Other theft

    5.3. Acts against computer systems

    5.3.1. Unlawful access to a computer system

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  1. 5.3.2. Unlawful interference with a computer system or computer data

    5.3.2.1 Unlawful interference with a computer system;

    5.3.2.2 Unlawful interference with computer data

    5.3.3. Unlawful interception or access of computer data

    5.3.4. Other acts against computer systems

    5.4. Intellectual property offenses

    5.5. Property damage

    5.5.1. Arson

    5.5.1.1 Arson of personal/residential property; 5.5.1.2 Arson of business or other nonpublic establishment property;

    5.5.1.3 Arson of public property

    5.5.2. Reckless burning

    5.5.3. Other damage of property

    5.6. Other acts against property only

  2. Acts involving controlled substances

    6.1. Unlawful possession or use of controlled drugs for personal consumption

    6.2. Unlawful cultivation or production of controlled drugs

    6.3. Unlawful trafficking or distribution of controlled drugs

    6.3.1. Street-level selling of quantities of controlled drugs suitable for personal consumption

    6.3.2. Wholesale distribution/trading/possession of controlled drugs

    6.4. Unlawful acts involving drug equipment or paraphernalia

    6.5. Other unlawful acts involving controlled drugs, psychoactive substances or precursors

  3. Acts involving fraud, deception, or corruption

    7.1. Fraud

    7.1.1. Consumer financial and products/services fraud

    7.1.2. Identity theft

    7.1.3. Fraud against businesses or establishments, including nonprofit organizations

    7.1.4. Fraud against government agencies

    7.1.5. Other types of fraud

    7.2. Forgery/counterfeiting

    7.2.1. Counterfeiting means of payment

    7.2.1.1 Counterfeiting means of cash payment; 7.2.1.2 Counterfeiting means of noncash payment

    7.2.2. Counterfeit product offenses

    7.2.3. Acts of forgery/counterfeiting documents

    7.2.4. Other acts of forgery/counterfeiting

    7.3. Corruption

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  1. 7.3.1. Bribery

    7.3.2. Embezzlement

    7.3.3. Abuse of functions

    7.3.4. Trading in influence

    7.3.5. Other acts of corruption

    7.4. Acts involving proceeds of crime

    7.4.1. Money laundering

    7.4.2. Illicit trafficking in cultural property

    7.4.3. Fencing stolen goods

    7.4.4. Other acts involving proceeds of crime

  2. Acts against public order and authority

    8.1. Acts against public order behavioral standards

    8.1.1. Violent public disorder offenses

    8.1.2. Acts related to social public order norms and standards

    8.1.3. Other acts against public order behavioral standards

    8.2. Acts against public order sexual standards

    8.2.1. Prostitution offenses

    8.2.2. Pornography offenses

    8.2.3. Other acts against public order sexual standards

    8.3. Acts related to freedom of expression or control of expression

    8.3.1. Acts against freedom of expression

    8.3.2. Acts related to expressions of controlled social beliefs and norms

    8.4. Acts contrary to public revenue or regulatory provisions

    8.4.1. Tax evasion, and other acts against taxation provisions

    8.4.2. Market manipulation, insider trading, and other acts against market or financial regulations

    8.4.3. Acts against regulations on alcohol, tobacco, or gambling

    8.4.3.1 Acts against regulations on alcohol or tobacco;

    8.4.3.2 Acts against regulations on gambling

    8.4.4. Customs violations

    8.4.5. Other violations of public revenue and regulatory provisions

    8.5. Acts related to migration

    8.5.1. Offenses related to smuggling of migrants

    8.5.2. Unlawful entry/border crossing

    8.5.3. Unlawful employment or housing of an undocumented migrant

    8.5.4. Other unlawful acts related to migration

    8.6. Acts against the justice system

    8.6.1. Obstruction of justice

    8.6.2. Breach of justice system authority

    8.6.3. Preparatory or enabling crimes

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  1. 8.6.4. Other acts against the justice system

    8.7. Acts related to democratic elections

    8.7.1. Acts intended to unduly influence voters at elections

    8.7.2. Other acts related to democratic elections

    8.8. Acts contrary to labor law

    8.8.1. Collective labor law violations

    8.8.2. Individual labor law violations

    8.9. Acts contrary to juvenile justice regulations or involving juveniles/minors

    8.9.1. Status offenses

    8.9.1.1 Status offenses committed by juveniles; 8.9.1.2 Status offenses committed upon juveniles

    8.9.2. Other acts contrary to juvenile justice regulations

    8.10. Other acts against public order and authority

  2. Acts against public safety and national security

    9.1. Acts involving weapons, explosives, and other destructive materials

    9.1.1. Unlawful possession or use of weapons and explosives

    9.1.1.1 Unlawful possession or use of firearms; 9.1.1.2 Unlawful possession or use of other weapons or explosives; 9.1.1.3 Unlawful possession or use of chemical, biological, or radioactive materials; 9.1.1.4 Other acts related to possession or use of weapons and explosives

    9.1.2. Trafficking of weapons and explosives

    9.1.2.1 Trafficking of firearms; 9.1.2.2 Trafficking of other weapons or explosives; 9.1.2.3 Trafficking of chemical, biological or radioactive materials; 9.1.2.4 Other acts related to trafficking of weapons and explosives

    9.1.3. Other acts relating to weapons and explosives

    9.2. Acts against national security

    9.3. Acts related to organized criminal groups

    9.3.1. Racketeering, and violations of the Racketeer Influenced and Corrupt Organizations (RICO) Act

    9.3.2. Other acts related to an organized criminal group

    9.4. Terrorism

    9.4.1. Participation in a terrorist group

    9.4.2. Financing of terrorism

    9.4.3. Other acts related to the activities of a terrorist group

  3. Acts against the natural environment or against animals

    10.1. Acts that cause environmental pollution or degradation

    10.2. Acts involving the movement or dumping of waste

    10.3. Trade or possession of protected or prohibited species of fauna and flora

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  1. 10.4. Acts that result in the depletion or degradation of natural resources

    10.4.1. Illegal logging or mining

    10.4.2. Illegal hunting, fishing, or gathering of wild fauna and flora

    10.5. Acts against animals

    10.6. Other acts against the natural environment or against animals

  2. Other criminal acts not elsewhere classified

    11.1. Violations of military law

    11.2. Violations of tribal law

    11.3. Torture

    11.4. Piracy

    11.5. Genocide

    11.6. War crimes

    11.7. Other criminal acts not elsewhere classified

    5.2.2 Provisional Set of Attributes or Tags to Accompany Proposed Classification of Crime for Statistical Purposes

    The following set of attributes—some covering the complete crime incident (and so, potentially, multiple offenses); others applying to each offense within an incident; and the remainder describing the victim, (suspected) offender/perpetrator, and victim/offender relationship—is of coequal importance in specifying this suggested classification. We use the term “attributes” to describe these data items; the UNODC’s ICCS describes them as “tags.” Thinking ahead to eventual implementation of a crime data system building from this classification, it is also reasonable to describe them based on their major function: These attributes are disaggregating variables that should allow eventual users of the crime data to zero in on substantive subsets of crime of interest (and to reclassify offenses and incidents according in different ways).

    In working with the ICCS “tags” as a base, and reflecting comments made during the panel’s user/stakeholder workshop-style sessions, the intent of this suggested attribute table is to focus on contextual variables that are objectively (and relatively easily) measurable. Certainly, one model for eventual collection of crime statistics is resolving the task down to relatively simple questions—a data collection instrument that could be completed with relative ease by either a law enforcement officer on the street or survey respondent, or perhaps in an automated fashion through harvesting text strings from electronic written incident reports. The focus on straightforward-to-define attributes enables that strategy and is meant to improve end data quality, but it does have consequences in excluding some points of information. We built notions of the quantity/value of drugs involved into the statements of offenses, but have omitted a general attribute for value of property involved (e.g., stolen

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

or damaged)—the value of property variable being commonly described as fraught with error or difficulty by current NIBRS-reporting departments. Likewise, reliable information about the mental health status (and history) of offenders, or the drug or alcohol involvement/usage by both offender and victim, would be greatly beneficial for policy development—but so difficult to measure precisely and objectively in a routine crime statistics collection system that we omit them. We have also tried to be relatively sparing in the number of attributes, again in the interest of easing eventual implementation.

Incident Attributes

  • Incident date and time: Date and time when incident occurred (or began); time, ideally, at least to precision of nearest hour of the day (00–23) as in NIBRS
  • Incident geographic location: Ideally an appropriately anonymized latitude/longitude pair; otherwise, a geocode to some small-area geography such as census block or tract
  • Incident location type: Based, initially, on a partitioning of location codes currently defined in NIBRS
    • – Residential location: Residence or home; school or college residential facility; nursing care or assisted living facility; other
    • – Store or retail: Grocery store or supermarket; convenience store; liquor store; pharmacy or drugstore; department or discount store; specialty store; shopping mall; auto dealership; other
    • – Financial institution: Bank or savings and loan; automated teller machine (ATM) separate from bank; other
    • – Commercial establishment: Commercial or office building; hotel, motel, or other lodging; service or gas station; rental storage facility; farm or agricultural facility; industrial site; other
    • – Entertainment venue: Bar or nightclub; restaurant; theater; stadium/arena; gambling facility (casino/racetrack); other
    • – School grounds and academic buildings: College/university; secondary; middle; elementary
    • – Civic or justice system establishment: Hospital, medical office, or clinic; daycare or child care facility; public safety (police or fire station or substation) facility; religious facility; military installation; community center; government or public building (including courthouses); correctional facility (jail, prison, penitentiary); mission or homeless shelter; other
    • – Transportation or related facility: Private or commercial motor vehicle (car, taxi, truck, etc.); public transportation or rail vehicle (train, subway, bus, etc.); water-borne vehicle; airport or bus/train terminal or station; parking lot or garage; dock, wharf, or modal terminal; other
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
    • – Outdoors: Street, sidewalk, road, alley, or highway; park or playground; beach, lake, or waterway; camp or campground; field or woods; other – Other physical location: Construction site; abandoned/condemned structure; tribal lands; other/unspecified physical location
    • – Multiple physical locations
    • – Not location-based (e.g., cybercrime)
  • Number of victims, all offenses in incident: Numeric, with provision for coding “victimless” crimes and unknown/many victims
  • Number of offenders, all offenses in incident: Numeric
  • Extraterritoriality: Yes/No depending on whether the incident includes offenses that may have been committed, executed, or initiated in another country or state, but federal or state law holds that such offense may be considered as though committed within local jurisdiction
  • Multiple jurisdiction: Yes/No if the incident includes offenses that could be legitimately counted or reported in more than one law enforcement agency’s operational jurisdiction

Per-Offense Attributes

  • Offense attempt or completion:
    • – Attempted
    • – Threatened (neither attempted nor completed)
    • – Completed
    • – Not applicable
    • – Not known
  • Victim/offender situation:
    • – Single victim/single offender
    • – Single victim/unknown offender(s)
    • – Single victim/multiple offenders
    • – Multiple victims/single offender
    • – Multiple victims/multiple offenders
    • – Multiple victims/unknown offender(s)
  • Victim/offender relationship: Perpetrator is, to the victim. . .
    • – Stranger
    • – Relative: Parent; child; sibling; grandparent; grandchild; other relative
    • – Known to the victim: Current spouse or intimate partner; former spouse or intimate partner; colleague (e.g., at workplace); friend; other known person/acquaintance
  • Type of weapon or force involved, for attack-type crimes:
    • – No weapon or force involved
    • – Weapon or force involved
      • Attack with firearm: Handgun/pistol; rifle/long gun; other/unknown
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
      • Attack with non-firearm weapon: Knife, cutting instrument, or sharp object; blunt object; motor vehicle; poison; fire or incendiary device; explosives; drugs or narcotics; other external weapon/force
      • Attack without external weapon: Bodily attacks (hands/fists, feet, etc.); asphyxiation or strangulation; drowning or submersion; pushing into harm’s way (e.g., from high place)
    • – Unknown
  • Group support/motivation:
    • – Organized: Gang-related; organized crime-related; terrorism-related; unknown organized involvement
    • – Unorganized (e.g., protest/demonstration)
    • – Not applicable
    • – Not known
  • Apparent/suspected bias motivation: Yes/no indicators for any of:
    • – Race/ethnicity
    • – Religion
    • – Sexual orientation
    • – Gender
    • – Disability
    • – Professional affiliation (including justice system/law enforcement)
    • – Other bias
    • – No apparent bias motivation
    • – Not known
  • Cybercrime-related: Yes/No depending on whether the use of computer data or computer systems was an integral part of the modus operandi of the offense
  • Type of property involved (stolen, damaged, etc.) in offense, if applicable: Should include some revision of the roughly 80 property types defined in NIBRS, and include intangibles such as intellectual property and personally identifiable information
  • Type, and quantity, of drug/psychoactive substance involved (for Group 6 controlled drug offenses): Should involve some variant on the roughly 20 drug types/classes currently coded in NIBRS, up to 3 of which can be specified and including a category to indicate involvement of more than 3 drug types

Victim Attributes

  • Type of victim: Should include, at minimum:
    • – Person/individual
    • – Business
    • – Financial institution
    • – Government
    • – Religious organization
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
    • – Law enforcement officer
    • – Other
    • – Unknown
  • Race/ethnicity, victim: Consistent with race/ethnicity categories used by U.S. Census Bureau and/or required by U.S. Office of Management and Budget
  • Gender, victim:
    • – Male
    • – Female
    • – Not applicable
    • – Not known
  • Age, victim: Numeric
  • Resident status, victim: (with respect to operational jurisdiction of reporting law enforcement agency)
    • – Nonresident
    • – Resident
    • – Unknown
  • Citizenship, victim:
    • – U.S. citizen (includes dual citizens with U.S.)
    • – Foreign citizen
    • – Refugee/no citizenship
    • – Not known

Offender Attributes

  • Type of offender: Same values as type of victim
  • Race/ethnicity, offender: Same values as race/ethnicity, victim
  • Gender, offender: Same values as gender, victim
  • Age, offender: Same values as age, victim
  • Resident status, offender: Same values as resident status, victim
  • Citizenship, offender: Same values as citizenship, victim

5.2.3 Changes and Deviations from the ICCS and from Current U.S. Crime Measurement Norms

To make clear again, our suggested classification draws directly from the UNODC’s ICCS, with and implying all due credit to the UNODC task force and expert work group for their work in drawing up the structure. Some changes that we make to the ICCS are essentially cosmetic in nature, Americanizing spellings and removing from the list of exclusions in the long-form presentation some specific offenses that seem clearly to be features of European law rather than U.S. standards. We also lightly revise some category titles for clarity or use of common U.S. terminology, as in adding “for ransom” to 2.3.2 Kidnapping for ransom (to more clearly signal the distinction with

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

illegal restraint or abduction, namely the demand for money or action in return for release of a victim) or using the more commonly understood term “carjacking” in 4.1.2 as a label rather than generic reference to robbery of a vehicle (in the presence of, and at least threatening violence toward, the operator).

Structural and Stylistic Differences

We have already alluded to one divergence from the UNODC’s ICCS model, which we settled on early in the process: We chose to be substantially more sparing in working the classification down to the fourth level of categorization. The net numbers of second- and third-level categories we define are very similar to corresponding totals in the ICCS, but the totals mask some purposeful rearrangements. In some cases, we collapse fourth-level breaks suggested by the UNODC; in others, we “promote” some lower-level categories up by one or two levels, where it seemed natural to do so. The Irish “condensed” crime classification was one inspiration for this stylistic choice, but more generally we sought to make the second- and third-level category lists as meaningful as possible, and so felt that working down to that level (particularly with the added flexibility for reclassification that would come with the companion collection of attribute data) would be sufficiently precise for an array of uses.

Revisiting the point we made above in arguing for a hybrid approach between purely attribute-based and purely code-/definition-based classification systems, we had to continually deal with a necessary tension: When is it appropriate to carve out an explicit, new crime category (having corresponding weight with all the others), rather than to rely on attribute collection to modify and facilitate deeper analysis/reclassification by downstream users? There are no easy answers—hence, again, the importance of an ongoing feedback and review process to refine the classification and related measures. In the next subsection, we describe in detail the most prominent example in which we opted to fold an attribute into the categories themselves: Where the ICCS would collect information about assaults and threats and rely on a weapon-use attribute to go into more detail, we judged it a priority to create firearm-specific subcategories, so that a U.S. crime statistics system would be equipped to assess the number of shootings. Similarly, 2.3.1 abduction of a minor is an offense in which both victim age and the victim-offender relationship combine to create such materially different offenses that separate categories seem justified, relative to just general “abduction” or illegal restraint with the accompanying attributes. But the decisions go the opposite direction, too. Shoplifting is one such example—a high-volume offense, and hence one specific offense that could dwarf the counts of other specifically included crimes under category 5.2.3 theft from business or other nonpublic organization. But volume of incidents alone

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

did not seem to us to be compelling argument for carving it out as a numbered offense. Some gradation by the type of property stolen (and, accordingly, a rough sense of its value) might be more salient, but that information is already called for in our attribute table, so it would seem unduly redundant to assign categories that way.

Stating the problem more generally, we recognize that achieving categories that are mutually exclusive and exhaustive requires the drawing of sharp lines between events and circumstances that may be ambiguous and would appear different to different observers. Moreover, the ostensibly objective attributes, such as age, race, sex, and education may incorporate some ambiguity as well: a 20-year-old single mother may not be directly comparable in important respects to a 20-year-old college student; the measurement of information on race and ethnicity, and self-identification of the same, has always been fraught with definitional peril; and sex and gender identification are beginning to raise similar concerns. Nonetheless, we feel that these categories may provide some measure of comparability, and the specific values or choices offered in collecting the attribute data can be revised as needed in the future.

Disentangling “(Aggravated) Assault” and Prioritizing Information on Shootings

The biggest changes that we suggest in our classification occur early on, and directly result from the shortcomings in current data—and most-desired improvements in measurement quality, going forward—that we heard from stakeholder groups in our conversations. These changes are the deliberate avoidance of the long-standing concept of “aggravated assault,” instead focusing on components that were previously fused under that general heading, and the incorporation of one particular type of weaponry—firearms—into category headings for assault and threat, so as to be able to systematically count events that involve the shooting of a firearm.

On the first of these points, it is important to note that assault has always been a difficult concept in crime measurement, lacking as it does the clearly identifiable extreme-point outcomes of either ultimate harm (death) or zero harm. The drafters of the original UCR definitions grappled with the problem that generic definitions of assault can be overly broad (International Association of Chiefs of Police, 1929:197–199):

Assault is generally defined as any unlawful physical force, partly or fully put in motion, creating in the victim a reasonable apprehension of immediate physical injury. This definition is sometimes extended by statute to include any unlawful attempt, coupled with the present ability to inflict serious bodily injury on the person of another. [However,] it is clear that if either of the foregoing definitions were adhered to, a large number of unimportant and petty offenses would be included in this class.

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

Offenses like touching another in a rude, insolent, and angry manner, and hazing, would fall in the same category and score just as heavily as assault with a deadly weapon with intent to kill, and maiming. Obviously, there is no comparison in the relative gravity of these two types of assaults and the inclusion of both in the same class would diminish its reliability as an index of crime.

Accordingly, the UCR drafters sought to “confine” assault in its short list of Part I crimes to “serious or aggravated,” borrowing the term “aggravated” from its occurrence in some state laws “where it is employed to indicate a special kind of assault, such as assault while hooded or disguised, assault upon an officer in the discharge of his duty, assault in a court of justice, and the like.” In staking out their own definition of “aggravated assault,” the drafters noted their intent to have the category cover “only those serious assaults most likely to result in severe bodily injury or death.” This was a reasonable objective, but one somewhat undercut over the next several paragraphs of their own description:

  • The UCR drafters observed seven “natural classes” or subcategories within their new “aggravated assault” category—including a central one that presumes good knowledge of intent (“assault with intent to kill or murder”), one laden with unclear legal terminology (“maiming, mayhem, and assault with intent to maim or commit mayhem”), and one at considerable odds with the interpersonal violence focus of the other subcategories (“willful obstruction of railroads”).
  • In one sentence, the 1929 UCR manual sets the precedent for the still-current practice of firearm involvement being a trigger for an assault being deemed aggravated—these aggravated assaults are “most likely to be reported to the police” precisely “because of the gravity of their nature and because in each case the overt act is accompanied by the use of a weapon or means likely to produce death or great bodily harm.” However, a few sentences later, “offenses such as shooting, or throwing at or into railroad trains[,] have been omitted from the aggravated assault group”—not because they lack the potential to produce death or great bodily harm, but because “they are usually offenses of malicious mischief by children.”

Ambiguity as to whether the use of an external weapon automatically makes an assault “aggravated” rather than “simple”—and so counted among the more serious UCR Part I offenses, rather than the less visible Part II—continues in current UCR usage. The UCR program currently uses a two-sentence definition that stipulates that weaponry “usually” triggers counting as aggravated assault (Federal Bureau of Investigation, 2013b:37):

[Aggravated assault is] an unlawful attack by one person upon another for the purpose of inflicting severe or aggravated bodily injury. This type of

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

assault usually is accompanied by the use of a weapon or by means likely to produce death or great bodily harm.

Arguably, this definition creates the necessary opening for assaults involving only hands/fists, feet, and such, but that do lead to serious, life-threatening injury, to be deemed aggravated assault, and the “automatic” inclusion as aggravated assault would come about through the use of an external weapon. This last point seems to be the point raised shortly following the definition in the current UCR manual (Federal Bureau of Investigation, 2013b:38):

It is the practice of local jurisdictions to charge assailants in assault cases with assault and battery, disorderly conduct, domestic violence, or simple assault even though a knife, gun, or other weapon was used in the incident. This type of offense is reported to the UCR Program as Aggravated Assault.

What has transpired over the decades with “aggravated assault” is not necessarily a purely naive misapplication of an unclear and overly broad definition—but it certainly could be. Participants in the panel’s workshop-style sessions critiqued the overinclusive nature of the current UCR definition, frustrated at the stereotypical barroom dispute that—in the mere presence of a single firearm—leads to the recording of potentially dozens of aggravated assaults—with no sense whatsoever from the data whether the firearm was actually used, how real the threat may have been, or whether any actual shooting or wounding took place. But just as there is no solid reason to believe that misclassification between aggravated and simple assault is purely naive, neither can the possibility of willful “cooking the books” to make headline-grabbing violent crime totals appear lower be dismissed. When police departments have been accused of distorting crime statistics—for example, in Milwaukee (Poston, 2012b,a,d,c), Chicago (Bernstein and Isackson, 2014a,b, 2015), and most recently (at this writing) Los Angeles (Poston and Rubin, 2015)—one common cause is the blurred line on counting assaults. In the Los Angeles case, reanalysis of a sample of nearly 4,000 case files across the seven years 2008–2014 suggested that, in each of those years, aggravated assault had been undercounted by roughly 36 percent (3,700 cases) due to misclassification of events as simple assault (Bustamante, 2015:9).3

The Los Angeles Police Department (LAPD) case is illustrative of other sources of error in distinguishing assault types, raising topics we will return to in our second report as they are more the province of implementation and methodology than fundamental classification. But it is worthy of brief mention here because there is a definitional and conceptual issue at root: the perpetual lack of concordance between state criminal codes, national (UCR)

__________________

3The audit, by the Inspector General for the Los Angeles Police Commission, also found that events that should have been counted as aggravated assault for UCR purposes were logged as other Part II offenses as well, including kidnapping or sexual assault (Bustamante, 2015:9).

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

reporting standards, and the code lists used in information systems throughout the process. In brief: Individual officers file paper investigative or incident reports, and are directed to “title” them with the California Penal Code offenses involved. Information from the paper report is then entered into the Los Angeles Police Department’s records management systems by records clerks, who can assign up to four Crime Class Codes to the report’s entry. The records management system, in turn, translates the Crime Class Codes into UCR categories for reporting to the California Department of Justice (and then to the FBI). That said, California is among the states whose criminal codes distinguish between assault (“an unlawful attempt, coupled with a present ability, to commit a violent injury on the person of another”) and battery (“any willful and unlawful use of force or violence upon the person of another”)—with battery, and the actual explicit use of force, being the more severe offense (California Penal Code §§ 240 and 245, respectively). The problem for crime statistics reporting is that the definitions do not strictly align, but is exacerbated by flaws in the code lists used to populate the local records management system. An officer could properly “title” an incident as being “felony battery”—battery through use of hands/fists or feet, without weapons, that causes serious bodily injury—but “felony battery” does not exist in the Crime Class Code list. Hence, offenses labeled by the reporting officer as “felony battery” or just “battery” are prone to being coded by the records clerks as just “battery”—which the records management system treats as simple assault for UCR purposes rather than aggravated assault (Bustamante, 2015).4

The problem is further exacerbated by discordance between the records system’s routine for ranking the Crime Class Codes and the UCR Program’s hierarchy rule for Summary Reporting System data; to wit, in an incident in which a person is kidnapped by gunpoint/threatened force, the kidnapping code would outrank the element of (UCR aggravated) assault (and logically so, given that the circumstances of the kidnapping would be of primary investigative importance), yet in the UCR the aggravated assault would be the ranking offense for summary purposes. Some errors, however, are more clearly due to inadequate or ineffective training. For example, the only circumstances in which reporting LAPD officers are instructed to “title” their reports as to whether an assaultive incident is “aggravated” or “simple” in nature are domestic violence or child abuse incidents. Yet an Inspector General audit found that over 75 percent of the original incident reports failed to include such a distinction, meaning that they were likely to be coded as simple assault by clerks (Bustamante, 2015).

__________________

4Similar mismatches occur for the (admittedly rarer) Penal Code-defined offense of mayhem, serious assault resulting in permanent disfigurement/mutilation. Mayhem is explicitly considered aggravated assault in the UCR definitions—but, again, the Los Angeles Police Department system lacks a Crime Class Code for mayhem, leading to those (few) cases being classified as Part II “other miscellaneous crimes” (Bustamante, 2015).

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

Current nationally compiled statistics on aggravated assault tell something about the levels of serious but nonlethal violence in the United States, but it is not clear exactly what that “something” is. One thing that it clearly is not, however, is a clean measure of the number of shootings (or, more generally, assaults or threats in which firearms are used, fired or not fired) for any given time period. Participants in our workshop-style meetings repeatedly described frustration that such a measure is not easily (or at all) derivable from current national sources for comparative purposes—particularly because the number of shootings last night, or last week, or last month are summary statistics that every police chief, sheriff, or other law enforcement official is said to want to know (and is expected to know) instantly. It is certainly important to know the weaponry (or not) involved in every homicide, every assault, and every serious threat—hence its prominence as an attribute/tag in our listing—but firearm involvement is of such high public import and interest that we judge shootings (or at least the necessary elements to easily calculate such a total) to merit a role, built in to our suggested classification’s categories rather than relying solely on reference to the attribute listing.

Based on these two objectives, we structure pieces of our categories 1 and 2 to try to decompose the common elements of current “aggravated assault”—and deliberately do not use that term in new headings—and fold information that would otherwise be covered by the weapon-used attribute into relevant category headings:

  • Under category 2, we “promote” assault and threat into two separate second-level headings rather than combine the two, as the ICCS does. In our definitions, we attempt to distinguish between assault as behavior that results in actual injury/harm and threat as behavior that does not.
  • Our assault category distinguishes between “severe” and “minor” assault based on the level of injury inflicted, and further distinguishes serious assault by whether it involves a shooting or not.
  • Our threat category includes the proviso that it involves intentional behavior against a person that is “not part of the attempt or completion of some other defined crime,” by so doing trying to isolate the element of threat from other crimes such as harassment, threat of rape, or the like. In subcategories for threat, we seek to disentangle the type of gun brandishing/display-but-no-firing occurrences that would currently be counted as aggravated assault from those where a shot is fired (but does not cause injury).
  • To avoid problems with mutual exclusivity of categories—in particular, confusion or overlap with our serious assault subcategory—we differ from the UNODC ICCS by striking their “attempted intentional homicide” as a subcategory under category 1. The specific legally defined crimes that are named in the ICCS under “attempted intentional homicide”—
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

conspiracy to murder and attempted (but failed/incomplete) killings in terrorist attacks—are addressed elsewhere in the classification.

Delineating Places in Classification for Fraud, Arson, Harassment/Stalking, and Other Offenses

Areas where we fundamentally disagree with the ICCS are very rare, and so instances where we differ from the ICCS are generally motivated by other factors. In the case of fraud (7.1), the ICCS invokes a very general definition (“obtaining money or other benefit, or evading a liability through deceit or dishonest conduct”), and differentiates only between “financial fraud” and “other acts of fraud” at the second level of classification. Accordingly, ICCS’s “other acts for fraud” swells to include things from identity theft to insurance fraud to medical quackery (not amounting to malpractice). We take as nearly certain that fraud represents sufficiently unfamiliar measurement terrain for nationally compiled statistics that further refinement of the category is inevitable, eventually, based on what data are or are not available for collection. But, even with that, we thought it best to aim somewhat more broadly in this initial classification:

  • A task force wholly separate from our panel, assembled by the Stanford Center for Longevity’s Financial Fraud Research Center (FFRC) and including membership from BJS, devoted time to constructing a full classification for financial fraud alone; their resulting classification is stated by Beals et al. (2015) and summarized in brief in Box 5.1. From the FFRC’s solid work, we adapt a more detailed definition of fraud (and the related definition of deceit), and seek to impart some of the key features of the FFRC classification in our fraud subcategories and included offenses, including recasting the first subcategory as “consumer financial and products/services fraud” to connote the broad array of behaviors included.
  • Given its emphasis on classes of financial fraud, the offense of identity theft does not fit naturally into the FFRC classification. However, we retain it as a subtype of fraud (as 7.1.2 in our hierarchy) for several reasons. It is one of the explicit inclusions as an “other type of fraud” in the ICCS, so we are “promoting” it in our classification, and it has generally been classed as a type of fraud or deception in U.S. applications (e.g., topic supplements to the National Crime Victimization Survey and the surveys of consumer fraud conducted by the Federal Trade Commission).
  • As does the FFRC, we add a subcategory for fraud against businesses or establishments (including nonprofit organizations); the ICCS recognizes
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

only a distinction between finance fraud against persons and finance fraud against the state/government.

One crime type of relatively long standing in U.S. crime statistics that is not well handled or delineated in the ICCS is arson. Though it can be a particularly problematic crime to measure, and though we are not bound to strictly conform to definitions in current federal or state law, we think it appropriate to define arson (and the related crime of reckless burning, which does not include the insurance collection or similar motive that commonly drives arson) as a standalone category of 5.5 property damage. The ICCS treats arson solely as a named example of an inclusion under general property damage. The ICCS suggests integrating information about the property/location in question into the definition of categories under the core property-related offenses of burglary and theft, namely suggesting that there are meaningful differences between those crimes when the target is residential, commercial/business, or public premises. We concur, and apply the same subcategory breaks to arson.

As a final example of a major change we suggest relative to the ICCS, we modify the definitions of harassment and (particularly) stalking based on what we heard from advocacy groups and practitioners in our workshop-style meetings. We agree with the ICCS about grouping these two crime types under the heading (2.9, in our classification) of acts intended to induce fear or emotional distress. But, in our assessment, the ICCS definitions of both crimes fundamentally misstate how the crimes are considered in American procedure. In particular, in U.S. usage, both are considered “course of conduct” offenses in which the harassment or stalking “incident” is not a single action at a single point of time, but rather a pattern of behavior over a spell of time (however short). To further clarify:

  • The ICCS builds from other United Nations documents to define harassment as, at minimum, “improper behaviour directed at and which is offensive to a person by another person who reasonably knew the behaviour was offensive. This includes objectionable or unacceptable conduct that demeans, belittles or causes personal humiliation or embarrassment to an individual.” Again, we find that this definition is incomplete, rather than strictly inaccurate, because the behavior in question is not specified as a pattern of repeated conduct. The ICCS subcategorizes “harassment in the workplace” as distinct from “other harassment,” but we decline to follow that split, given that workplace harassment (including sexual harassment in the workplace) is not commonly included in federal or state criminal code, but may be subject to civil sanctions and (particularly) penalties under employers’ policies.
  • The ICCS definition of stalking—“unwanted communication, following or watching a person”—is, again, ill-specified in our estimation. It misses
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

Box 5.1 Financial Fraud Resource Center Suggested Taxonomy of Criminal Fraud

  1. Individual financial fraud

1.1 Consumer investment fraud

1.1.1 Securities fraud

1.1.1.1 Equity investment fraud — 7 subcategories by type of equity (e.g., Real Estate Investment Trust, oil and gas exploration)

1.1.1.2 Debt investment fraud — 4 subcategories by type of investment (e.g., promissory note)

1.1.1.3 Other securities fraud

1.1.2 Commodities trading fraud

1.1.2.1 Forex (foreign exchange) fraud

1.1.2.2 Commodity pool fraud

1.1.2.3 Precious metals fraud

1.1.2.4 Other commodities fraud

1.1.3 Other investment opportunities fraud

1.1.3.1 Hollywood film scam

1.1.3.2 Property/real estate scam

1.1.3.3 Rare object scam

1.2 Consumer products and services fraud

1.2.1 Worthless or nonexistent products (intentionally entered agreement)

1.2.1.1 Worthless products — 9 subcategories by product type (e.g., weight loss products, fake gemstones)

1.2.1.2 Paid never received — Subcategories for online marketplace fraud and other

1.2.1.3 Other worthless/nonexistent products

1.2.2 Worthless, unnecessary, or nonexistent services (intentionally entered agreement)

1.2.2.1 Phony insurance

1.2.2.2 Immigration services/Notario fraud

1.2.2.3 Invention fraud

1.2.2.4 Fraud loss recovery

1.2.2.5 Debt relief scam — 5 subcategories (e.g., student debt relief, mortgage relief)

1.2.2.6 Credit repair scam

1.2.2.7 Fake credit lines and loans — Subcategories for fake loans, fake credit lines/credit cards, and other

1.2.2.8 Fortune-telling fraud

1.2.2.9 Phishing websites/emails/calls — Subcategories for tech support scam, spoofing websites, and other

1.2.2.10 Timeshare resale fraud

1.2.2.11 Adoption scam

1.2.2.12 Internet gambling fraud

1.2.2.13 Fake buyers scam

1.2.2.14 Unnecessary or overpriced repairs, or repairs never performed — Subcategories for auto repair, home repair, and other

1.2.2.15 Travel booking scam

1.2.2.16 Website hosting/design scam

1.2.2.17 Domain name scam

1.2.2.18 Other

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

1.2.3 Unauthorized billing for products or services

1.2.3.1 Buyer’s clubs

1.2.3.2 Unauthorized billing, Internet services — Subcategories for online Yellow Pages and other

1.2.3.3 Unauthorized billing, phone services — Subcategories for cramming, slamming, and other

1.2.3.4 Unauthorized billing, magazines

1.2.3.5 Unauthorized billing, credit monitoring services

1.2.3.6 Other unauthorized billing fraud

1.2.4 Other consumer products and services fraud

1.3 Employment fraud

1.3.1 Business opportunities fraud

1.3.1.1 Multilevel marketing scheme

1.3.1.2 Vending machines/ATM leasing scam

1.3.1.3 House flipping courses

1.3.1.4 Business coaching scam

1.3.1.5 Other

1.3.2 Work-at-home scam

1.3.2.1 Home assembly

1.3.2.2 Envelope stuffing

1.3.2.3 Mystery Shopper

1.3.2.4 Reshipping

1.3.2.5 Other

1.3.3 Government job placement scam

1.3.4 Other employment scam

1.3.4.1 Nanny scam

1.3.4.2 Modeling fraud

1.4 Prize and grant fraud

1.4.1 Prize promotion/sweepstakes scam

1.4.1.1 Free product

1.4.1.2 Free vacation

1.4.1.3 Cash prize

1.4.1.4 Sweepstakes scam

1.4.1.5 Other

1.4.2 Bogus lottery scam

1.4.2.1 Foreign lottery scam

1.4.2.2 Other

1.4.3 Nigerian letter fraud

1.4.4 Government grant scam

1.4.5 Inheritance scam

1.4.6 IRS tax refund opportunity

1.4.7 Other prize and grant fraud

1.5 Phantom debt collection fraud

1.5.1 Government debt collection scam

1.5.1.1 Court impersonation scam

1.5.1.2 IRS back taxes scheme

1.5.1.3 Other

1.5.2 Lender debt collection scam

1.5.2.1 Obituary scam

1.5.2.2 Loan debt scam

1.5.2.3 Other

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

1.5 Phantom debt collection fraud (continued)

1.5.3 Business debt collection scam

1.5.3.1 Fake health and medical debt

1.5.3.2 Other

1.5.4 Other phantom debt fraud

1.6. Charity fraud

1.6.1 Bogus charitable organization

1.6.1.1 Bogus natural disaster-related charity

1.6.1.2 Bogus disease-related charity

1.6.1.3 Bogus law enforcement charity

1.6.1.4 Bogus veteran charity

1.6.1.5 Bogus church/religious group charity

1.6.1.6 Bogus animal shelter

1.6.1.7 Bogus alumni charitable giving

1.6.1.8 Bogus children’s charity

1.6.1.9 Bogus political group

1.6.1.10 Bogus youth organization

1.6.1.11 Other

1.6.2 Crowdfunding for bogus cause

1.6.2.1 Fake personal medical expenses

1.6.2.2 False identity as natural disaster or national tragedy survivor

1.6.2.3 Other

1.6.3 Other charity fraud

1.7 Relationship and trust fraud (wherein expected outcome is fostering a relationship)

1.7.1 Romance or sweetheart scam

1.7.2 Friends or relatives imposter scam

1.7.2.1 Grandparent scam

1.7.2.2 Other

1.7.3 Other relationship and trust fraud

  1. Fraud against organizations

2.1 Fraud against government agencies, programs, regulations, and society

2.1.1 Government programs (includes welfare fraud; disability fraud; Medicare/Medicaid fraud)

2.1.2 Government regulations (includes immigration fraud; voting fraud; tax fraud; stamp fraud)

2.1.3 Other (includes insider trading; environmental fraud)

2.2 Fraud against nongovernmental businesses or organizations

2.2.1 Occupational fraud (committed by internal perpetrator) (includes corruption; asset misappropriation; financial statement fraud)

2.2.2 Fraud committed by external perpetrator (includes insurance fraud; bank fraud; fraudulent suppliers)

NOTES: Taxonomy accompanied by suggestion for 6 incident-related attributes/tags—general incident description (9 values), method of advertising (6 values), purchase setting (5 values), method of money transfer (10 values), dollar loss categories, and duration; 7 values of a victim descriptor tag (e.g., veteran victim; victim reported fraud to authorities); and 5 values of a perpetrator descriptor tag. Full development of category 2, fraud against an organization, was deemed “beyond the scope” of initial activity.

SOURCE: Beals et al. (2015:Sec. VI).

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

the course-of-conduct, repeated-acts nature of the offense and, worse, too strongly connotes the notion of stalking almost necessarily involving surreptitious physical surveillance of a victim. We redefine stalking for our purposes based on the language used in some state criminal codes and, in particular, the Model Stalking Code suggested and revised by the National Center for Victims of Crime (2007).

Other Major Changes from International Classification or from Current U.S. Practice

Other fairly major restructuring of categories or revision of labels or definitions, relative to the ICCS or to current U.S. practice, embedded in our suggested classification include the following:

  • In the text of the definition for our 1.1 murder and intentional homicide, we add the phrasing “committed with reckless indifference to life” as one standard for defining intentional homicide. While we are not beholden to the extant text of federal or state criminal code, “reckless indifference to human life” is prominent in the wording of the American Law Institute’s Model Penal Code definition of murder/intentional homicide, and so has made its way into various state statutes. In this nod to current widespread usage, we also recognize the need for care with the term “reckless,” because it might be inappropriately deemed to be equal with “negligent.”
  • The ICCS crafts a second-level category for sexual violence, with rape and sexual assault as third-level categories within. We elevate both those offenses to second-level status (3.1 and 3.2, respectively), to increase their visibility as crime categories. We also make substantive revisions to the subcategories of both. We revise the ICCS’s mention of “statutory rape”—which can be a problematic term in U.S. state statute—to more directly get at what we think is intended as a subcategory, namely rape involving inability to express consent or nonconsent (3.1.3). In terms of sexual assault, the ICCS differentiates between physical sexual assault, non-physical sexual assault, and other sexual assault. However, the distinction between the last two subcategories is not readily clear from the ICCS’s stated definitions and legal inclusions; we think that the more general (single) subcategory “threat of a sexual nature” is an effective complement to the physical sexual assault entry, once we add a new second-level category 3.3 for sexual violations of a nonphysical nature, which would cover such behaviors as voyeurism or recording a person without consent.
  • We restructure the categories under 3.5 sexual exploitation of children, for two principal reasons. First, the ICCS’s single subcategory for child prostitution blends two very different types of unlawful and unacceptable
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  • behaviors—put most bluntly, between the pimp/provider side and the procurement/“customer” side—that we think merit separate delineation. Second, the ICCS includes a separate subcategory for sexual grooming of children, a term that may have picked up usage in international law discussions but that we think has not acquired sufficient standing in U.S. usage (or federal/state law) to warrant a separate partition in our classification; we fold it into our category 3.5.4 for other offenses related to sexual exploitation of children.

  • In the ICCS, category 4—using our revised title, acts of violence or threatened violence against a person that involve property—is effectively synonymous with robbery. One crime type that has made its way into the criminal codes of some U.S. states, that is not directly referenced in the ICCS, and that we add to our classification to fill a perceived gap in category 4, is sometimes simply referred to as “terroristic threats.” We expand that label, and definition, to refer more fully to terroristic or disruptive threats to buildings or critical infrastructure—covering bomb, biohazard, and other threats to buildings, transportation facilities, power infrastructure, and the like, done to cause serious harm to a large number of people (where that harm might be severe disruption of daily routines as well as actual physical injury). An actual completed act of terrorism, beyond the threat, would fall under 9.4.1’s participation in a terrorist group (along with whatever homicide or assault offenses might apply), but the communicated threat to cause disruption seems sufficiently serious as to warrant separate categorization.
  • The ICCS’s category 6 heading, acts involving controlled drugs or other psychoactive substances, includes the second clause so as to include alcohol, tobacco, or other substances—things that may be more tightly regulated in other countries than they are in the American experience. For the most part, the specific items listed as inclusion in the alcohol and tobacco section of the ICCS read to us as being more about revenue or customs policies than about the substances themselves. Accordingly, we took particular note of how Victoria handles drugs, alcohol, tobacco, and related substances in its crime classification and borrowed the approach for our classification. We retitle our category 6 to refer only to “controlled substances” and, short of specifying specific weights or measures (which would certainly vary by drug), differentiate between “street-level” quantities and “wholesale” quantities under 6.3’s unlawful trafficking or distribution of controlled drugs heading. We take care to define 6.1 as unlawful possession or use of controlled drugs for personal consumption, adding in the explicit “unlawful” nomenclature to reflect state-by-state variation in permissions; currently, marijuana would fall under this heading because it is a federally defined controlled substance, but states are increasingly decriminalizing
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  • possession of personal-consumption amounts (and the individual state definition would determine whether the event is “unlawful” or not). We do leave possession of “wholesale” levels of controlled substances as a specific offense under category 6.3.2 because U.S. law treats possession of controlled substances as prima facie evidence of distribution (or intent to distribute) if the quantity in question is above that legally deemed suitable for personal consumption. Meanwhile, inasmuch as the content of the ICCS’s alcohol and tobacco plank seemed to deal with taxation/revenue policy, we combined the two with another “sin tax” category with variation by state—gambling—as a subcategory under 8.4 acts contrary to public revenue or regulatory provisions.

  • For sake of parsimony, and recognizing the likely difficulty that lies ahead in collecting data on the topic, we opted to collapse the ICC’s distinction between “active bribery” and “passive bribery” under our category 7.3.1. Colloquially, the distinction between active and passive bribery is simply between offering/paying a bribe and receiving a bribe. As a first cut, we think it best to add a note that our notion of bribery includes both sides but not to define separate categories.
  • We make numerous changes throughout category 8, acts against public order and authority—none of which are large in scope but that, collectively, are meant to preserve consistency with the ICCS (to facilitate comparison to the greatest extent possible). However, elements of categories 8.1–8.3 are instances where there is ground for greater discrepancy between other nations’ “public order” behavioral or sexual standards and the U.S. standards—and the staunch American value of freedom of expression certainly complicates (if not fully negates) definition of offenses under the ICCS’s acts related to freedom of expression or control of expression (our category 8.3). We attempt to disentangle tax, customs, and other revenue-related provisions in defining subcategories under heading 8.4, and we also delineate two new subcategories under category 8.5 acts related to migration: unlawful entry/border crossing (into the U.S.) and unlawful employment or housing of an undocumented migrant, elevating two important concepts that were previously compressed into one of the ICCS’s catch-all categories. Under 8.6.2, the ICCS uses the phraseology “breach of justice order,” which corresponds too closely with the notion of a formal “court order” in American usage. From the inclusions under this category, it is evident that the ICCS intends the subcategory to be substantially broader, including the specific offenses of resisting arrest and violating procedures (as an inmate) internal to correctional facilities. Accordingly, we use the substitute wording “breach of justice system authority.” Finally, and arguably the biggest change in this specific category, we expand the ICCS’s mention of status offenses—behaviors that are criminal solely
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
  • because of the age of either the victim or the perpetrator—and shift it from the catch-all category 11 to category 8, where it seems a more natural fit (and which avoids truancy/absence from school from falling in the same category with treason and genocide).

  • Under category 9, acts against public safety and national security, we eliminate a second-level heading suggested by the ICCS—removing acts against health and safety as a standalone category, and adding mention of occupational safety and health violations to 2.8.1’s acts that endanger health of another person. Category 9.3’s handling of organized crime was and remains a point of debate with the UNODC work group, and is certainly worthy of inclusion in an international crime framework. However, we struggled with how best to cast this in our suggested classification, largely due to the inherent difficulty of fully documenting organized crime involvement without (and even sometimes with) lengthy investigation. The ICCS’s offense category “participation in an organized criminal group” seemed overly broad, and it was difficult to identify specific crimes related to organized crime—other than racketeering offenses—in its stead. We do include organized crime affiliation as one of the group support/attribution attribute levels (completed on a per-offense basis), but it seemed inapt to simply designate “participation” in a group as a named offense in the hierarchy. Ultimately, we felt it best to borrow a page from, in particular, the Irish crime classification system, which is replete with references to violations of specific pieces of legislation, and so make our category 9.3.1 into violations of the (federal) Racketeer Influenced and Corrupt Organization (RICO) Act.

What Is Missing from Our Suggested Crime Classification

Consistent with the goal, as part of the properties of a proper statistical classification, of making the classification exhaustive of the phenomena of interest, we believe our suggested classification to be comprehensive. To be sure, there are still “gaps” that can be pointed to, but the ones that come readily to the fore are not included for two basic reasons. The first is that the seeming “gap” is something that may take time or investigation to rule as “crime” or not—and, hence, may not in fact be correctly designated as “crime.” So, for instance, there is no standalone category for “justifiable homicide” in our classification—properly so, since such events are not crime, by definition, if being found to be justifiable after due adjudication and investigation. Our suggested classification is not bound to any particular data collection, or to any particular time of data collection relative to the incident, and so it covers things like justifiable homicide (which requires adjudication, and would translate to murder or intentional homicide if deemed to be not justifiable), arson (which commonly requires investigation of the circumstances of burn evidence), and

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

some white-collar offense/fraud variants (where the first signal of the crime in question might not be a report to law enforcement at all, but rather the filing of charges or complaints).

The second type of gap in the classification—types or classes of crime that should or may be recoverable from a fully realized dataset built around the classification (attributes and all), but that are not explicitly called out in the classification’s listing—is probably best exemplified by cybercrime. We retain the ICCS’s per-offense attribute of cybercrime involvement—a binary yes/no flag based on whether computer systems or data were integral to the modus operandi of the offense, and the offenses listed under category 5.3 in our hierarchy are explicitly computer-centric. With that, our suggested classification has the potential to generate pieces of the overall cybercrime puzzle—harassment can be modified and reanalyzed to detect cyberharassment or cyberbullying, and likewise cyber-related stalking, identity theft (using computer means), and others. But, based on what we heard from users and stakeholders, we concluded that cybercrime per se is much like fraud or intentional homicide—a sufficiently broad and diverse concept that it could warrant a fully realized three- or four-level hierarchical classification on its own. But cybercrime poses the added complication that even the vocabulary used to describe specific variants in such a detailed classification would almost certainly be out of date by the time it was published. Hence, we deemed it best not to try to consolidate all the pieces of cybercrime in one place in the hierarchy, or to try to assess what a classification sub-tree for cybercrime might look like; instead, we opt to wait and see how the base crime + attribute combination works in practice.

Short of missing crime types, another fault that could be raised with our suggested classification—but again a point on which we made a conscious decision—is that it does not “impose” any binding, national-level constraints or parameters. Through repeated use of the word “unlawful” in the definitions and, less frequently, the category titles, we are generally deferential to the precise definitions that might apply in a particular state. We do this to set a starting position, expecting that a national-level standard, cut-off value, or the like may be revisited after some period of data collection and refinement. But, in the interim, we recognize that this does introduce some lack of “precision” in our specifications. So, for instance, in 2.3.1 abduction of a minor, 3.5 sexual exploitation of children, and 8.9 acts contrary to juvenile justice regulations or involving juveniles/minors—among other places in the classification—we decline to suggest a fixed, uniform age cut-off that distinguishes minors/juveniles/children from adults. State criminal statutes vary greatly in the age cut-offs below which a person is deemed unable to grant consent for sexual activity or other purposes, or below which referral to child protective services (or other juvenile-specific law enforcement branches)

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

or juvenile justice systems would take hold. Rather, as an interim measure, we permit the standards that would apply in a particular reporting jurisdiction.

5.3 NEXT STEPS: ISSUES GOING FORWARD

If used as a blueprint for constructing a new set of indicators of crime in the United States, the crime classification we suggest in this chapter and report has the potential to be path-breaking in advancing the nation’s understanding of crime. By dismantling the current definition of aggravated assault and collecting information on the pieces, the nation would be poised to better and more directly estimate the levels, trends, and costs of the kinds of assaults that cause greatest angst among the public. The classification opens the door, potentially, to the gathering of data on “new” frontiers of crime, at least for U.S. nationally compiled data. White-collar offenses like corruption, embezzlement, and market manipulation could finally be presented side-by-side with “street crime”; systematic collection of offenses related to the operation of the criminal justice system itself (breach of justice system authority, obstruction of justice) could yield insight into separate analyses of the functioning of the judicial branch; and damaging crimes that have heretofore been studied principally through use of limited survey data (such as harassment, stalking, and identity theft) could finally have another benchmark for comparison.

The subsequent steps of implementation and methodological development are daunting, and should begin naturally with a first revisiting of our suggested crime classification and mapping it to current (or not-as-yet created) data sources that might supply the requisite information. For this interim report, we purposefully set out not to overwhelm readers with specific findings and recommendations, instead keeping the focus on the suggested crime classification.

That said, we recognize that this particular moment in time presents a unique opportunity, coming as it does in the wake of public statements by the director of the Federal Bureau of Investigation noting frustration with the state of current crime statistics, indicating intent to sunset the UCR Program’s Summary Reporting System in favor of a full-fledged NIBRS, and elevating improvement of crime statistics (and creation of a parallel database to record law enforcement use-of-force incidents). It also comes as the Bureau of Justice Statistics (BJS) is poised, by equipping a sample of law enforcement agencies to begin reporting data in NIBRS format through the National Crime Statistics Exchange (NCS-X) Program), to finally showcase the analytic power of NIBRS because NIBRS so seeded with selected agencies will finally constitute a representative sample of the population. It also comes as the major organizational bodies of chiefs of police (including the International Association of Chiefs of Police, the Major Cities Chiefs Association, the

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

National Sheriffs’ Association, and the Major County Sheriffs’ Association) jointly issued a statement supporting the FBI’s proposed transition from Summary Reporting to NIBRS, and as the group of police chiefs and law enforcement practitioners convened by BJS (the Crime Indicators Working Group) advocates for joint analysis of crime indicator data with community and demographic information in order to put crime and its trends into proper context.

In short, this is a uniquely opportune time to state a few select conclusions that derive directly from this report’s focus on development of a crime classification and from what we learned from outreach to the broader crime statistics constituencies. We think that some basic truths need to be stated—directly and bluntly—as the debate on how to move forward with (reformed) crime statistics data collection begins, and that these conclusions do not prejudge implementation- and methodology-specific recommendations we will offer in our final report.

To begin, we repeat a point that we made earlier in this chapter describing our objectives for constructing the classification: By stating the classification now and in this form, we do not suggest or expect that it be immediately swapped in as the underlying offense code list for NIBRS, the National Crime Victimization Survey (NCVS), or any current data collection. It is, as described in Recommendation 5.1, a first and foundational step, around which a data collection system should be designed and developed after considering numerous issues and implementation challenges. The main purpose of the suggested classification is to suggest the sheer breadth of “crime in the United States” relative to the more limited picture forged over the past nine decades, and not to generate angst over how much is left unexplored in current U.S. crime statistics.

Drawing a correspondence between this suggested classification and current (or as-yet nonexistent) data collections awaits our second report. That said, even the most cursory review of the current list of crimes covered by NIBRS (Table 2.1), by the NCVS (Box 2.4), and (particularly) by the UCR Summary Reporting System (SRS; Box 2.1) in contrast with our suggested classification suggests that the gaps in coverage/knowledge will be numerous and glaring. This suggests a basic conclusion:

Conclusion 5.4: Full-scale adoption of incident-based crime reporting by all respondents or sources, that is sufficiently detailed to permit accurate classification and extensive disaggregation and analysis, is essential to achieving the kind of flexibility in crime statistics afforded by a modern crime classification.

To be clear, our use of the phrase “by all respondents or sources” is intended to connote that we anticipate that both police-report data and survey-based measures of crime and victimization (at a minimum) will have essential roles in a modern crime measurement system.

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

There are two basic corollaries to this conclusion that are supported by this report’s classification-based focus and that we feel should be made now in order to contribute in a timely manner to current debates, even though more detailed discussion awaits our second report. Both relate to the FBI’s recent decision (stated most clearly in the director’s note accompanying Federal Bureau of Investigation (2015)) to work with its law enforcement agency contributors and to transition from the SRS to full NIBRS implementation. The first corollary is that the transition away from the SRS format and content is sound and appropriate. The SRS was a major advance when created in 1929–1930 and proved instrumental for decades in shedding basic light on national crime trends, but it is simply inadequate to provide information of the quality or the level of detail demanded by modern crime data users. But the second corollary concerns the transition to NIBRS: It is important that the transition to full-up NIBRS be cast strictly as an intermediate step. A full-participation NIBRS that holds to that system’s current design and content would have great difficulty satisfying all or most crime statistics user needs. To be fair, the relatively spotty participation in NIBRS has undercut the capacity to demonstrate the full analytic power afforded by the system, because NIBRS estimates to date are not representative of broader regional or national populations. And, it is important to reemphasize that we do not expect that any single data collection (NIBRS, NCVS, or otherwise) can satisfy user needs or come close to filling out all parts of our suggested classification. Our concern is that NIBRS’s core development work and structuring took place in the late 1980s, and it is not clear that its design has kept pace with changes in reporting practices, computing technology, and emerging crime types. Upgrading from a 1929-vintage crime data management system to a 1990-vintage system falls well short of the data infrastructure that crime statistics deserve, but it would indeed be a remarkable advance.

Finally, we end this report by recognizing that implementing new technical systems—distributed across some 18,000 local, state, and federal agencies with highly differential resources and capacities—is incredibly difficult, let alone phasing in a wholesale change to the underlying classification of events that underlies the system. The 30-plus-years (and counting) history of NIBRS falling short of participation expectations (and, with them, the system’s analytic potential) is testament to that difficulty, as is the failure to move the pure attribute-based classification of the mid-1970s (SEARCH) from prototype to limited production. Sorting out the barriers, real and imagined, to NIBRS implementation over recent decades will be a critical part of our final report, as will determining what lessons may be learned from (among others) the United Kingdom’s and Ireland’s experience with major process or classification change being undermined by flaws in data collection at the source. Likewise, learning from the Australian experience of individual states maintaining their own classifications while ensuring compatibility with a modern (but fairly

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

mature) classification scheme will be important to building effective federal and subnational partnership. Classifying crime is but the first step—very complex in its own right, but arguably the “easy” part—in modernizing the nation’s crime statistics infrastructure.

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×

This page intentionally left blank.

Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 117
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 118
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 119
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 120
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 121
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 122
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 123
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 124
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 125
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 126
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 127
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 128
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 129
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 130
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 131
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 132
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 133
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 134
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 135
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 136
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 137
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 138
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 139
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 140
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 141
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 142
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 143
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 144
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 145
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 146
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 147
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 148
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 149
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 150
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 151
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 152
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 153
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 154
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 155
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 156
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 157
Suggested Citation:"5 Proposed Classification of Crime for Statistical Purposes." National Academies of Sciences, Engineering, and Medicine. 2016. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime. Washington, DC: The National Academies Press. doi: 10.17226/23492.
×
Page 158
Next: References »
Modernizing Crime Statistics: Report 1: Defining and Classifying Crime Get This Book
×
Buy Paperback | $79.00 Buy Ebook | $64.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

To derive statistics about crime – to estimate its levels and trends, assess its costs to and impacts on society, and inform law enforcement approaches to prevent it – a conceptual framework for defining and thinking about crime is virtually a prerequisite. Developing and maintaining such a framework is no easy task, because the mechanics of crime are ever evolving and shifting: tied to shifts and development in technology, society, and legislation.

Interest in understanding crime surged in the 1920s, which proved to be a pivotal decade for the collection of nationwide crime statistics. Now established as a permanent agency, the Census Bureau commissioned the drafting of a manual for preparing crime statistics—intended for use by the police, corrections departments, and courts alike. The new manual sought to solve a perennial problem by suggesting a standard taxonomy of crime. Shortly after the Census Bureau issued its manual, the International Association of Chiefs of Police in convention adopted a resolution to create a Committee on Uniform Crime Records —to begin the process of describing what a national system of data on crimes known to the police might look like.

The key distinction between the rigorous classification proposed in this report and the “classifications” that have come before in U.S. crime statistics is that it is intended to partition the entirety of behaviors that could be considered criminal offenses into mutually exclusive categories. Modernizing Crime Statistics: Report 1: Defining and Classifying Crime assesses and makes recommendations for the development of a modern set of crime measures in the United States and the best means for obtaining them. This first report develops a new classification of crime by weighing various perspectives on how crime should be defined and organized with the needs and demands of the full array of crime data users and stakeholders.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    Switch between the Original Pages, where you can read the report as it appeared in print, and Text Pages for the web version, where you can highlight and search the text.

    « Back Next »
  6. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  7. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  8. ×

    View our suggested citation for this chapter.

    « Back Next »
  9. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!