Most information about law enforcement’s response to juvenile crime comes from the FBI’s UCR Program
Since the 1930s, police agencies have reported to the UCR Program
Each year, thousands of police agencies voluntarily report the following data to the Federal Bureau of Investigation’s (FBI’s) Uniform Crime Reporting (UCR) Program:
- Number of Index crimes reported to law enforcement (see sidebar).
- Number of arrests and the most serious charge involved in each arrest.
- Age, sex, and race of arrestees.
- Proportion of reported Index crimes cleared by arrest and the proportion of these Index crimes cleared by the arrest of persons under age 18.
- Police dispositions of juvenile arrests.
- Detailed victim, assailant, and circumstance information in murder cases.
What can the UCR arrest data tell us about crime and young people?
The UCR arrest data can provide estimates of the annual number of arrests of juveniles* within specific offense categories. UCR data can also provide detail on juvenile arrests by sex, race, and type of location (urban, suburban, or rural area). The data can be used to compare the relative number of arrests of adults and juveniles within offense categories, to develop estimates of change in arrests over various periods, and to monitor the proportion of crimes cleared by arrests of juveniles.
What do UCR data count?
UCR data document the number of crimes reported to police, not the number committed. The UCR Program monitors the number of Index crimes that come to the attention of law enforcement agencies. Although this information is useful in identifying trends in the volume of reported crime, it is important to recognize that not all crimes are brought to the attention of law enforcement.
Crimes are more likely to be reported if they involve a serious injury or a large economic loss and if there is a desire to have law enforcement involved in the matter. Therefore, some crimes are more likely to come to the attention of law enforcement than are others. For example, the National Crime Victimization Survey for 1998 found that victims reported 80% of motor vehicle thefts to police, 63% of robberies, 58% of aggravated assaults, 50% of burglaries, 41% of simple assaults, 32% of sexual assaults, and 29% of thefts. Overall, victims reported to law enforcement 46% of violent crimes and 36% of property crimes.
Changes in the proportion of crimes reported may, therefore, reflect more than changes in the number of crimes actually committed. They may also reflect changes in the willingness of victims to report crimes to law enforcement agencies.
|
What are the Crime Indexes? The designers of the UCR Program wanted to create an index (similar in concept to the Dow Jones Industrial Average or the Consumer Price Index) that would be sensitive to changes in the volume and nature of reported crime. They decided to incorporate specific offenses into the index based on several factors: likelihood of being reported, frequency of occurrence, pervasiveness in all geographical areas of the country, and relative seriousness. The Crime Index is divided into two components: the Violent Crime Index and the Property Crime Index: Violent Crime Index —Includes murder and nonnegligent manslaughter, forcible rape, robbery, and aggravated assault. Property Crime Index —Includes burglary, larceny-theft, motor vehicle theft, and arson. Crime Index —Includes all eight crimes in the Violent Crime Index and Property Crime Index. Although some violent crimes, such as kidnapping and extortion, are excluded, the Violent Crime Index contains what are generally considered to be serious crimes. In contrast, a substantial proportion of the crimes in the Property Crime Index are generally considered less serious crimes, such as shoplifting, theft from motor vehicles, and bicycle theft, all of which are included in the larceny-theft category. |
UCR arrest data reflect only the most serious offense for which a person was arrested. An arrest of a person for both robbery and weapons possession would appear in the UCR data as one robbery arrest. The UCR data on number of weapons arrests, therefore, reflect only those arrests in which a weapons charge was the most serious offense charged. This aspect of the UCR counting rules must be taken into consideration when the data are used to analyze arrest volume and trends for less serious offenses.
Clearance data provide another perspective on law enforcement. A crime is considered cleared if someone is charged with the crime or if someone is believed to have committed the crime but for some reason (e.g., the death of the suspect, unwillingness of the victim to prosecute) an arrest cannot be made. If a person is arrested and charged with committing 40 burglaries, UCR records 40 burglary clearances. If three people are arrested for robbing a liquor store, UCR records one robbery cleared.
Knowing both the number of crimes reported and the number cleared in a year makes it possible to compute the proportion of crimes cleared in a year. A much greater proportion of violent crimes than property crimes are cleared.
| Most serious offense | Percent of all crimes cleared in 1999 |
|---|---|
 | |
| Violent Crime Index | 50% |
| Murder | 69 |
| Forcible rape | 49 |
| Robbery | 29 |
| Aggravated assault | 59 |
| Property Crime Index | 18 |
| Burglary | 14 |
| Larceny-theft | 19 |
| Motor vehicle theft | 15 |
| Arson | 17 |
UCR data capture the proportion of crimes cleared by juvenile arrest. UCR data also document the proportion of cleared crimes that were cleared by the arrest of persons under age 18. Assessments of the juvenile contribution to the crime problem are often based on this proportion. Clearance and arrest statistics give very different pictures of the juvenile contribution to crime.
| 1999 Juvenile proportion | ||
|---|---|---|
| Most serious offense | Arrests | Crimes cleared |
| ||
| Violent Crime Index | 16% | 12% |
| Murder | 9 | 6 |
| Forcible rape | 17 | 12 |
| Robbery | 25 | 15 |
| Aggravated assault | 14 | 12 |
| Property Crime Index | 32 | 22 |
| Burglary | 33 | 19 |
| Larceny-theft | 31 | 23 |
| Motor vehicle theft | 35 | 19 |
| Arson | 54 | 49 |
How should juvenile arrest and clearance data be interpreted?
Considerations in interpreting UCR data on juvenile arrests and clearances can be demonstrated by attempting to answer a typical question about juvenile crime: "In 1999, what proportion of all robberies were committed by juveniles?" The UCR data show that 25% of all persons arrested for robbery in 1999 were under age 18 and that 15% of all robberies cleared in 1999 were cleared by the arrest of persons under age 18.
The key to reconciling the difference between the two percentages is the fact, noted previously, that juveniles are more likely than adults to commit crimes in groups. If a police department cleared all seven of its robberies in a year by arresting two juveniles for one incident and six different adults for the other six incidents, the juvenile proportion of persons arrested for robbery would be 25% (2 in 8), and the juvenile proportion of robberies cleared would be 14% (1 in 7). Arrest percentages are offender-based; clearance percentages are offense-based.
Clearance data are a better choice than arrest data for determining the juvenile proportion of all robberies committed. There are, however, questions about what clearance figures actually represent.
One question stems from the fact that a crime cleared by the arrest of a juvenile and the arrest of an adult is classified by the FBI as an adult clearance. Therefore, some cleared crimes involving juvenile offenders are not counted in the proportion of crimes cleared by juvenile arrest, a factor that makes the juvenile clearance proportion an underestimate of juvenile involvement in cleared crimes.
Another question is whether it is safe to assume that characteristics of robberies cleared are similar to characteristics of robberies not cleared. For example, were the 29% of robberies cleared in 1999 like the 71% not cleared?
A study of more than 21,000 robberies in 7 States between 1991 and 1993 found that robberies by juveniles were more likely to result in arrest than were robberies by adults (Snyder, 1999). The FBI's National Incident-Based Reporting System (NIBRS) data from these States gave the victim's perception of the age of the offender and indicated whether the offender was arrested. This study found that robberies by juveniles were 23% more likely to result in arrest than were robberies by adults. Therefore, the juvenile proportion of cleared robberies was substantially greater than the proportion of robberies actually committed by juveniles. Based on this finding, it appears that UCR clearance percentages overestimate the juvenile responsibility for crime because juvenile offenders are more likely to be arrested.
Arrest data and clearance data can be used to explore different types of questions. Arrest data provide a rough estimate of how many juveniles entered the justice system in a given year; but it must be remembered that a particular individual may have been arrested more than once during the year (and therefore counted more than once), and that a particular arrest may have involved more than one offense (with only the most serious charge counted). Clearance data are more useful than arrest data in estimating the proportion of crimes committed by juveniles, but evidence that juveniles are more likely than adults to be arrested for their crimes indicates that clearance percentages also exaggerate juveniles' actual share of crime.
However, and most important, the trends, or changes, in arrest data are likely to reflect actual changes in the number of juveniles entering the juvenile justice system, whereas changes in clearance proportions can be used to monitor changes in the relative responsibility of juveniles for crime.
How accurate are the UCR-based juvenile arrest and clearance trends?
Annually, the FBI generates national estimates of reported crimes for the 8 Index offenses and national estimates of total arrests in 29 offense categories. The Bureau does not currently produce national estimates of juvenile arrests or arrest rates (although it did produce tables of juvenile arrest rates up through the early 1990s). For those interested in juvenile arrest trends, the FBI’s annual Crime in the United States reports do contain numerous tables showing juvenile arrest counts reported to the FBI by that year’s contributing law enforcement agencies.
Statisticians characterize these annual samples as “opportunistic” samples—that is, each sample contains data from every agency that was willing and able to report to the FBI in that year. The essential problem is that the sample is not scientifically determined; therefore, no one can assume that the sample’s characteristics (e.g., juvenile arrest proportions, juvenile arrest rates) are representative of all the law enforcement agencies in the U.S.
For example, let us assume that one sample contained a disproportionate number of agencies from large metropolitan areas or cities. If so, then the arrest tables in Crime in the United States would present a picture of juvenile arrests with a more urban character than the U.S. as a whole. This would mean, compared with the U.S. overall, the data from the reporting sample would have a higher percentage of violent crime arrests, a higher percentage of juvenile arrests, higher rates of juvenile arrests for violent crimes, and higher percentages and rates of arrest for black juveniles across offense categories.
In all, the quality of the juvenile arrest rate trends derived from the sample data reported in Crime in the United States is dependent on the consistent representativeness of the annual reporting samples. There are currently no assessments of the representativeness of the annual samples. What is known is that the coverage of the sample has changed substantially in recent years. For 1999, law enforcement agencies with jurisdiction over 63% of the U.S. population contributed data on arrests, a proportion lower than at any time in the prior 20 years.
The traditional approach to the development of national estimates of juvenile arrests (and clearances) is based on the assumption that the reporting samples in the Crime in the United States series are nationally representative. The more this assumption is violated, the less reliable are the estimates. It is possible to adjust for some of the known, or measurable, biases in the samples, but this work has not been done. Even if such adjustments were made, the validity of the estimates would still be in question because of the inherent weaknesses of an opportunistic sample.
From a pragmatic standpoint, those who wish to study arrest and clearance trends are encouraged to turn to the FBI’s UCR Program and its Crime in the United States reports. This resource is the best information available, even with its weaknesses. Users, however, should always be aware of the potential biases in the data and the potential effects of these biases.
* In this Bulletin, “juvenile” refers to persons under age 18. This definition is at odds with the legal definition of juveniles in 1999 in 13 States—10 States where all 17-year-olds are defined as adults and 3 States where all 16- and 17-year-olds are defined as adults.
| Previous | Contents | Next |
 |
|
| Law Enforcement and Juvenile Crime | December 2001 |