THE TWO OFFICIAL SOURCES OF DATA
As early as the 1800s, public officials began keeping records about lawbreaking to gauge the “moral health” of society. Then, as now, high rates of inter- personal violence and theft were taken as signs of social pathology—indications that something was
V I C T IM I Z AT ION IN T H E UN I T ED S T AT ES : AN OVE RV I E W 69
Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. WCN 02-200-203
profoundly wrong with the way people generally interacted in everyday life. Annual data sets were compiled to determine whether illegal activities were being brought under control as time passed. Monitoring trends is even more important today because many innovative but intrusive and expensive criminal justice policies intended to curb crime have been implemented.
Two government reports published each year contain statistical data that enable victimologists to keep track of what is happening out on the streets. Both of these official sources of facts and figures are disseminated each year by the U.S. Depart- ment of Justice in Washington, D.C. Each of these government data collection systems has its own strengths and weaknesses in terms of provid- ing the information victimologists and criminolo- gists are seeking to answer their research questions. The Federal Bureau of Investigation’s Uniform Crime Report (UCR): Crime in the United States is a massive compilation of all incidents “known” to local, county, and state police departments across the country. The FBI’s UCR is a virtual “bible” of crime statistics based on victims’ com- plaints and direct observations made by officers. It is older and better known than the other official source, the Bureau of Justice Statistics’ National Crime Victimization Survey (NCVS): Criminal Victimization in the United States. The BJS’s NCVS provides projections for various regions and for the entire country based on incidents vol- untarily disclosed to interviewers by a huge national sample drawn from the general public.
Facts and Figures in the Federal Bureau of Investigation’s Uniform Crime Report (UCR)
The UCR was established in 1927 by a committee set up by the International Association of Chiefs of Police. The goal was to develop a uniform set of definitions and reporting formats for gathering crime statistics. Since 1930, the FBI has published crime data in the UCR that was forwarded volun- tarily to Washington by police departments across the United States. In recent years, more than 18,000 village, town, tribal, college, municipal,
county, and state police departments and sheriff’s departments in all 50 states, the District of Columbia, and several territories that serve about 97 percent of the more than 300 million inhabit- ants of the United States participate in the data collection process, usually by sending periodic reports to state criminal justice clearinghouses. Several federal law enforcement agencies now contribute data as well (FBI, 2014).
The FBI divides up the crimes it tracks into two listings. Unfortunately, both Part I and Part II of the UCR have been of limited value to those interested in studying the actual flesh-and-blood victims rather than the incidents known to police departments or the characteristics of the persons arrested for allegedly committing them.
Part I of the UCR focuses on eight index crimes, the illegal acts most people readily think about when they hear the term “street crime.” Four index crimes count violent attacks directed “against persons”: murder, forcible rape, robbery, and aggravated assault. The other four monitor crimes “against property”: burglary, larceny (thefts of all kinds), motor vehicle theft, and arson. (The category of arson was added in 1979 at the request of Congress when poor neighborhoods in big cities experienced many blazes of suspicious origin. However, incidents of arson are still unreliably measured because intentionally set fires might remain classified by fire marshals as being “of unknown origin.”) These eight crimes (in actual practice, seven) are termed index crimes because the FBI adds all the known incidents of each one of these categories together to compute a “crime index” that can be used for year-to-year and place-to-place comparisons to gauge the seriousness of the problem. (But note that this grand total is unweighted; that means a murder is counted as just one index crime event, the same as an attempted car theft. So the grand total [like “com- paring apples and oranges”] is a huge composite that is difficult to interpret.)
The UCR presents the number of acts of vio- lence and theft known to the authorities for cities, counties, states, regions of the country, and even many college campuses (since the mid-1990s; see
70 CH APT ER 3
Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. WCN 02-200-203
Chapter 11). For each crime, the FBI compiles information about the number of incidents reported to the police, the total estimated losses in billions of dollars due to property crimes, the proportion of cases that were solved, and some characteristics of the suspects arrested (age, sex, race)—but, unfortu- nately for victimologists, nothing about the people who suffered harm and filed the complaints.
In Part II, the UCR only provides counts of the number of people arrested for 21 assorted offenses (but no estimates of the total number of these illegal acts committed nationwide). Some Part II crimes that lead to arrests do cause injuries to individuals, such as “offenses against women and children,” as well as “sex offenses” other than forc- ible rape and prostitution. Others do not have clearly identifiable victims, such as counterfeiting, prostitution, gambling, drunkenness, disorderly conduct, weapons possession, and drug offenses. Still other Part II arrests could have arisen from incidents that directly harm identifiable individuals, including embezzlement, fraud, vandalism, and buying/receiving/possessing stolen property.
The Uniform Crime Reporting Division fur- nishes data in separate annual publications about how many hate crimes were reported to police departments (see Chapter 11). It also issues a yearly analysis of how many law enforcement officers were feloniously assaulted and slain in the line of duty, the weapons used against them, and the assignments they were carrying out when they were injured or killed (also analyzed in Chapter 11).
From a victimologist’s point of view, the UCR’s method of data collection suffers from sev- eral shortcomings that undermine its accuracy and usefulness (see Savitz, 1982; and O’Brien, 1985). First of all, underreporting remains an intractable problem. Because many victims do not inform their local law enforcement agencies about illegal acts committed against them and their possessions (see Chapter 6), the FBI’s compilation of “crimes known to the police” is unavoidably incomplete. The annual figures about the number of reported crimes recorded as committed inevitably are lower than the actual (but unknown) number of crimes that really were carried out. Second, the UCR
focuses on accused offenders (keeping track of the age, sex, and race) but does not provide any infor- mation about the complainants who reported the incidents. Only information about murder victims (their age, sex, and race) is collected routinely (see Chapter 4). Third, the UCR lumps together reports of attempted crimes (usually not as serious for vic- tims) with completed crimes (in which offenders achieved their goals). Fourth, when computing crime rates for cities, counties, and states, the FBI counts incidents directed against all kinds of targets, adding together crimes against impersonal entities (such as businesses and government agencies) on the one hand, and individuals and households on the other. For example, the totals for robberies include bank holdups as well as muggings; statistics about bur- glaries combine attempted break-ins of offices with the ransacking of homes; figures for larcenies include goods shoplifted from department stores in addition to thefts of items swiped from parked cars.
Another shortcoming is that the FBI instructs local police departments to observe the hierarchy rule when reporting incidents: List the event under the heading of the most serious crime that took place. The ranking in the hierarchy from most ter- rible to less serious runs from murder to forcible rape, robbery, aggravated assault, burglary, vehicle theft, and finally larceny. For instance, if an armed intruder breaks into a home and finds a woman alone, rapes her, steals her jewelry, and drives off in her car, the entire incident will be counted for record-keeping purposes only as a forcible rape (the worst crime she endured). The fact that this person suffered other offenses at the hands of the criminal isn’t reflected in the yearly totals of known incidents. If the rapist is caught, he could also be charged with armed robbery, burglary, motor vehicle theft, possession of a deadly weapon, and possession of stolen property, even though these lesser offenses are not added to the UCR’s compilations.
Phasing in a National Incident-Based Reporting System Fortunately, the UCR is being overhauled and is becoming a much more useful source of information about individuals who are harmed
V I C T IM I Z AT ION IN T H E UN I T ED S T AT ES : AN OVE RV I E W 71
Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. WCN 02-200-203
by lawbreakers. The FBI is converting its data col- lection format to a National Incident-Based Reporting System (NIBRS).
Law enforcement officials began to call for a more extensive record-keeping system during the late 1970s. The police force in Austin, Texas, was the first to switch to this comprehensive data collection and reporting system. However, other big-city police departments that deal with a huge volume of crime reports have had trouble meeting NIBRS goals and timetables, so complete imple- mentation has been postponed repeatedly. North Dakota and South Carolina were the first two states to adopt NIBRS formatting in 1991. As of 2008, NIBRS data collection was taking place in 32 states and the District of Columbia, covering 25 percent of the nation’s population, 26 percent of its reported crimes, and 37 percent of its law enforce- ment agencies (IBR Resource Center, 2011). But the switchover seems to be proceeding very slowly. A 2012 NIBRS national report compiled figures submitted by 6,115 law enforcement agencies (only 33 percent of the over 18,000 participating departments) in 32 states that covered about 30 per- cent of the nation’s population and 28 percent of all crimes known to police departments across the country (FBI, 2014b).
One major change is the abandonment of the hierarchy rule of reporting only the worst offense that happened during a sequence of closely related events. Preserving a great many details makes it pos- sible to determine how often one crime evolves into another, such as a carjacking escalating into a kidnap- ping, or a robbery intensifying into a life-threatening shooting. For the incident cited above that was cate- gorized as a forcible rape under the hierarchy rule, this new record-keeping system also would retain information about the initial burglary; the resulting robbery; the vehicle theft; other property stolen from the victim and its value; other injuries sus- tained; the woman’s age, sex, and race; whether there was a previous relationship between her and the intruder; and the date, time, and location of the incident. Until the advent of the NIBRS com- puter database, only in cases of homicide were some of these facts extracted from police files and retained.
Another significant aspect of the overhaul is that the NIBRS provides expanded coverage of additional illegal activities. Instead of just eight closely watched UCR index offenses, FBI compu- ters are now prepared to keep track of 46 Group A offenses derived from 22 categories of crimes. In addition to the four “crimes against persons” and the four “against property” of the UCR’s Part I, the new Group A monitors offenses that formerly had been listed in Part II or just not collected at all. Victim-oriented data are becoming available for simple assault (including intimidation), vandalism (property damage and destruction), blackmail (extortion), fraud (swindles and con games), forcible sex crimes (sodomy, sexual assault with an object, and fondling), nonforcible sex offenses (statutory rape and incest), kidnapping (including parental abductions), and the nonpunishable act of justifiable homicide. The data collected about a victim of a Group A offense include these variables: sex, age, race and ethnicity, area of residence, type of injury, any prior relationship to the offender, and the circumstances surrounding the attack in cases of aggravated assault and murder. If items were stolen, the details preserved about them include the types of possession taken and their value, and in cases of motor vehicle theft, whether the car was recovered. Also, the NIBRS makes a distinction between attempted and successfully completed acts (from the criminal’s point of view) (IBR Resource Center, 2011).
Already, some data mining studies based exclu- sively on the NIBRS archives from selected states and cities address some intriguing issues. For exam- ple, an analysis of roughly 1,200 cases of abductions in the 12 states that had switched over to NIBRS by 1997 shed light on a previously overlooked subcat- egory of “holding of a person against his or her will,” called acquaintance kidnapping. This newly recognized offense includes situations such as when a teenage boy isolates his former girlfriend to punish her for spurning him, to pressure her to return to him, to compel her to submit sexually, or to evade her parents’ efforts to break them up. Also included are incidents in which street gang members spirit off rivals to intimidate them, retaliate against them, or
72 CH APT ER 3
Copyright 2016 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. WCN 02-200-203
even recruit them. In these types of hostage takings, the perpetrators tend to be juveniles just like their teenage targets, the abductions often take place in homes as opposed to public places, and the cap- tives are more likely to be assaulted (Finkelhor and Ormrod, 2000).
Several studies using NIBRS data from certain jurisdictions have revealed important findings about various types of murders. In general, elderly men are murdered at twice the rate as elderly women (Chu and Kraus, 2004). Specifically, older white males are the most frequent victims of “eldercide”; they are killed predominantly by offenders who are below the age of 45 and tend to be complete stran- gers. Female senior citizens are more likely to be slain by assailants who are older than 45 and are often either a spouse or grown child (Krienert and Walsh, 2010). Murders of intimate partners tend to be committed late at night, during weekends, and in the midst of certain holidays more often than at other times (Vazquez, Stohr, and Purkiss, 2005). Also, the higher homicide rate in Southern cities actually may not be due to a presumed subculture of violence or “code of honor” that supposedly compels individuals likely to lose fights and suffer beatings to nevertheless stand up to the aggressors to save face (Chilton, 2004).