U.S. Department of Justice Office of Justice Program Bureau of Justice Statistics Special Report Violent Victimization and Race, 1993-98 March 2001, NCJ 176354 -------------------------------------------------------- This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.wk1) and the full report including tables and graphics in .pdf format are available from: http://www.ojp.usdoj.gov/bjs/abstract/vvr98.htm -------------------------------------------------------- By Callie Rennison, Ph.D. BJS Statistician -------------------------------------------------------- Highlights The rates at which black and white persons experienced violent victimization were converging between 1993 and 1998. * In 1998, per 1,000 persons age 12 or older in each racial group, 110 American Indians, 43 blacks, 38 whites, and 22 Asians were victims of violence. * In each year from 1993 to 1997, black persons were victimized at rates significantly greater than those of whites. By 1998 black and white persons were victimized overall at similar rates. * The rate of violent victimization of whites fell 29% and of blacks fell 38%, 1993-98. Over the same period no measurable change in the victimization rates of American Indians or Asians occurred. * From 1993 to 1998 violent crime in which the race of the offender was known was largely intraracial for whites (66%) and blacks (76%). For American Indian and Asian victims, violent crime was primarily interracial. * From 1993 to 1998 higher percentages of black (36%) and Asian (32%) victims than of white (24%) and American Indian (28%) victims faced an armed offender. Higher percentages of black (18%) and Asian (14%) victims than of American Indian (9%) and white (8%) victims faced an offender with a firearm. * 48% of the violence against blacks, 42% against whites, 46% against American Indians, and 41% against Asians were reported to the police, 1993-98. * Blacks were disproportionately represented among homicide victims. In 1998, 4 whites, 23 blacks, and 3 persons of other races (Asians and American Indians considered together) were murdered per 100,000 persons in each group. * On average each year between 1993 and 1998, homicide rates fell 5% for whites, 7% for blacks, and 8% for persons of other races. -------------------------------------------------------- Estimates from the National Crime Victimization Survey (NCVS) indicate that between 1993 and 1998 compared to people of other races American Indians sustained violence at the highest per capita rate (119 victimizations per 1,000 American Indians age 12 or older). This rate of violent victimization is about 2 times that experienced by blacks, 2 1/2 times that sustained by whites, and 4 1/2 times that experienced by Asians. Between 1993 and 1998 violent victimization rates declined for whites and blacks. Because of a greater decline in black victimization rates during this period, by 1998 black and white overall violent crime rates were similar. Over the same period, apparent changes in the per capita violent victimization rates of American Indians and Asians were not significant. Data Findings about rape, sexual assault, robbery, and assault come from National Crime Victimization Survey (NCVS) data collected by the Bureau of Justice Statistics (BJS). Findings about homicide come from the Uniform Crime Reporting (UCR) program of the Federal Bureau of Investigation (FBI). This report presents rates and characteristics of violent victimization of whites, blacks, American Indians, and Asians. The racial categories are separate from Hispanic ethnicity and include Hispanic and non-Hispanic persons. (* For information about Hispanic and non-Hispanic victimization, see Criminal Victimization 1999, Changes 1998-99, with Trends 1993-99, August 2000, NCJ 182734.) Violent victimization and race, 1993-98 American Indians experienced overall violence, aggravated assault, simple assault, and serious violent crimes at rates higher than those for whites, blacks, and Asians, 1993-98. Asians sustained overall violence, aggravated assault, simple assault, and serious violent crimes at per capita rates lower than all other groups during the same period. On average, American Indians (10.8 robberies per 1,000 persons age 12 or older) and blacks (10.6) were victims of robbery at similar rates, 1993-98. These rates were higher than those of whites (4.4) and Asians (5.7), which, in turn, were similar. From 1993 to 1998 the average annual rate of rape or sexual assault was somewhat higher for American Indians than that for blacks, and significantly higher for American Indians than that for Asians and whites. The rates for rape or sexual assault for Asians and whites were similar during this period. Characteristics of violent crime victims, 1993-98 Victimization research consistently demonstrates a relationship between several demographic attributes and the rate of victimization. When examining victimization differences by race, it is important to account for these factors, which include gender, age, ethnicity, income, marital status and location of residence. The patterns that emerge suggest that each of these characteristics taken alone does not account entirely for the violent victimization rates among racial groups. Gender Among males, without exception, overall victimization rates for each racial group differed significantly from all others. American Indian males were victims of overall violent crime at a rate about twice the rate of blacks, 22 times the rate of whites, and 4 times the rate of Asians. Rates of aggravated assault against males also differed among the races, 1993-98. American Indians were victims of aggravated assaults at a rate greater than that of blacks, who were victimized at a rate greater than that of whites, who were victimized at a rate greater than that of Asians. ------------------------------ American Indians and violent crime For more information about American Indians and their experiences with violent crime, convictions and correctional supervision, see American Indians and Crime, BJS report, February 1999, NCJ 173386. Highlights of this report include: * Indigenous people in the United States belong to about 550 federally recognized tribes that have a distinct history and culture and often a separate language. * A higher percentage of American Indian victims of violent crime than victims of other races stated the offender was drinking or on drugs, 1992-96. * About 4% of the American Indian population age 18 or older were under the care, custody, or control of the criminal justice system on an average day, 1992-96. By comparison, an estimated 2% of white adults, 10% of black adults, and less than half of 1% of Asian adults were under correctional supervision. -------------------------------- For serious violent crime the same pattern of differences between the races emerged, with one exception: Asian and white males had similar per capita rates, 1993-98. Among females, American Indians were victims of overall violence and serious violence at per capita rates greater than those of blacks, who were victimized at rates greater than those of whites, and whites were victimized at rates greater than those of Asians. Apparent differences in rape and sexual assault rates between American Indian, white, and black females were not significant. Asian females had a rate of rape or sexual assault that was slightly lower than the rate for American Indian females and significantly lower than those for white and black females, 1993-98. Age Within each age category and with few exceptions, American Indians experienced overall violent crime at the highest rate, Asians at the lowest rate, and blacks and whites at similar rates, 1993-98. An exception was among persons ages 12 to 15 for whom apparent differences in victimization rates between American Indians, blacks, and whites were not significant. Ethnicity Ethnicity is defined independently of race. Therefore, an individual may be described as Hispanic or non-Hispanic and of any race examined. Among Hispanics, American Indians were victims of overall violence at rates greater than rates for whites, blacks, and Asians, 1993-98. During the same period Hispanic blacks were victimized at rates greater than rates for Hispanic whites and Asians. Hispanic Asians and whites were victimized at similar rates, 1993-98. Among non-Hispanics, American Indians were victims of overall violence at rates greater than those for whites, blacks, and Asians, 1993-98. During the same period blacks were victimized at rates greater than rates for whites and Asians, and Asians were victimized at the lowest rates. The same pattern existed among non-Hispanics for aggravated assault rates and for serious violent crime. Marital status Among persons who were married and those who never married, American Indians were victims of overall violence at rates greater than rates for whites and blacks, and whites and blacks were victimized at rates greater than that for Asians, 1993-98. American Indians who never married sustained overall violence at about twice the per capita rate of whites and blacks who never married. American Indians who were divorced or separated were victimized at about 3 times the rate of divorced or separated Asians during this time. Among widowed persons, blacks experienced almost twice the rate of overall violence as whites. Small sample sizes prevent testing the apparent differences across races in the widowed category. As with overall violent crime, American Indians in most marital categories experienced serious violent crime at rates greater than those for persons in other racial categories. Among those divorced or separated, American Indians were victimized at rates slightly greater than rates for Asians, and significantly greater than rates for blacks and whites. Among married persons for serious violent crime, rates for each racial group differed significantly from all others. Place of residence American Indians experienced the highest rates of overall and serious violent crimes regardless of the locality of residence considered. American Indians in urban areas sustained overall violence at about 2 times the rate of blacks and whites and 5 times the rates of Asians. American Indians in urban areas experienced serious violence at about 2 times the rate of blacks, 3 times the rate of whites, and 5 times the rate of Asians, 1993-98. Household income Generally, when considered by income levels, American Indians experienced overall violence at the highest rate, Asians experienced overall violence at the lowest rate, and blacks and whites experienced overall violence at similar rates, 1993-98. Exceptions include those earning between $15,000 and $24,999 and over $50,000 annually: blacks and American Indians experienced overall violence at similar rates. Expressed as rates per 1,000 persons, blacks were victimized more than whites among those earning from $7,500 to $34,999 annually. Fewer racial differences emerge for serious violent crime. In general, blacks and American Indians experienced serious violent crime at similar rates, and their rates were higher than rates for whites and Asians. Owned versus rented residence Within the "owned housing" group, American Indians sustained violence at a rate about twice the per capita rates for blacks and whites, and 3 1/2 times the rate for Asians. Blacks were victimized at rates higher than those for whites, and whites were victimized overall at rates greater than those for Asians. Within the "rented housing" group, American Indians sustained violence at about twice the rate for whites and blacks, and 5 1/2 times the rate for Asians. While there was no measurable difference between the rate of black and white overall victimization, both groups were victimized at rates greater than that for Asians, 1993-98. Victimization trends, 1993-98 The rate of violent victimization against whites fell 29%, and that of blacks fell 38%, 1993-98. Over the same period apparent changes in the victimization rates of American Indians and Asians were not significant. For whites, in addition to a decline in rates of overall violence, rates of specific types of violence decreased 1993-98: a 44% decline in rape and sexual assault rates, a 39% decline in aggravated assault rates, and a 30% decline in robbery rates. For blacks, the rate of all types of violence except rape and sexual assault fell 1993-98: a 62% decrease in robbery rates, a 42% decline in aggravated assault rates, and 30% decline in simple assault rates. The rate at which American Indians sustained overall violence did not change significantly between 1993 and 1998; however, the rate at which American Indians experienced serious violent crime declined slightly, 1993-98. Declines in rates of overall and serious violent crime against Asians between 1993 and 1998 were not significant. Because of a lack of sample cases, trends of rates for specific types of crime against Asians and American Indians were not examined. The rate at which non-Hispanic whites, non-Hispanic blacks and Hispanics (regardless of race) experienced overall and serious violence declined between 1993 and 1998. Non-Hispanic whites experienced violent victimization at rates 27% lower in 1998 than in 1993. Among non-Hispanic blacks violent victimization rates fell 38% from 1993 to 1998, and among Hispanics, 45%. Declines in victimization rates were more pronounced for serious violent crime than for violent crime overall. Hispanics experienced serious violent crimes at rates 54% lower in 1998 than in 1993. Serious violent victimization rates were 36% lower for white non-Hispanics and 47% lower for black non-Hispanics during this time. For non-Hispanic American Indians and Asians there were no measurable changes in overall and serious violent victimization rates from 1993 to 1998. Characteristics of the violent crime event, 1993-98 Time and location The largest percentage of violent crime occurred in open areas for each racial group. Between 1993 and 1998, 27% of Asian, 26% of black, 23% of American Indian, and 20% of white victims were victimized in an open area. Blacks were victimized at their homes in percentages similar to American Indians, and higher than whites and Asians. Blacks were victimized in commercial locations in lower percentages than other racial groups. Whites were victimized in school in higher percentages (14%) than blacks and American Indians (11% and 6%) between 1993-98. American Indians were victimized in schools in percentages lower than all other groups. Whites, blacks, and Asians were victimized during the day in similar percentages. These percentages are greater than the percentage of American Indians victimized during the day. Fifty-two percent of American Indians were victimized at night. Slightly more than 4 in 10 whites (44%), blacks (44%), and Asians (43%) were victimized at night between 1993-98. Presence of weapons All violent crime victims were more likely to face an unarmed than an armed offender, 1993-98. Black victims (36%) faced an armed offender in percentages similar to those for Asians (32%), and in percentages higher than those for white (24%) and American Indian (28%) victims. Black and Asian victims were more likely to face an offender with a firearm than were white and American Indian victims. During this time, 18% of black victims, 14% of Asian victims, 9% of American Indian victims, and 8% of white victims faced an offender with a firearm. Injuries and medical treatment Most victims of violent crime were not injured, 1993-98. Seventy-seven percent of Asian, 76% of white, 71% of black, and 65% of American Indians victims were uninjured as a result of violence that they sustained. Larger percentages of American Indians and blacks (35% and 29%) were injured from violent crime than were whites and Asians (24% and 23%), 1993-98. American Indians were seriously injured in percentages similar to blacks, slightly higher than those of Asians, and significantly higher than those of whites. Serious injuries include gunshot wounds, knife wounds, rape, internal injuries, broken bones, and being knocked unconscious. A higher percentage of American Indians received minor injuries than whites and Asians. Blacks sustained minor injuries in slightly higher percentages than whites, 1993-98. Minor injuries include bruises, cuts, scratches, black eyes, swelling, and chipped teeth. Among injured victims, whites and Asians were less likely to receive treatment than were blacks and American Indians. Reporting victimizations to the police For blacks and whites between 1993 and 1998, violence against females is reported to the police in greater percentages than violence against males. There was no difference in the percentage of violence reported to the police based on the victim's gender for American Indians and Asians. Among male victims, violence was reported in similar percentages across racial categories. Among female victims, violence against blacks (53%) and American Indians (52%) was reported to police in similar percentages. The percentage of violence against black females reported to the police was higher than violence against whites (45%) and Asians (41%). The percentage of violence against white and Asian females reported to the police was similar. The reporting of the victimization of whites has increased significantly from 40% to 45% between 1993 and 1998, as determined by a linear trend test. Apparent changes, 1993 to 1998, in percentage of victimizations reported to police were not statistically significant for other racial categories. Victims' race and the reasons for not reporting a victimization to the police were unrelated, 1993-98. The primary reason given for not reporting a victimization to police was that the incident was a "private or personal matter," that it was minor involving "no loss," or that it was "reported to another official." Percent of violence reported to the police, 1993-98 Male Female Total White 40% 45% 42% Black 43 53 48 American Indian 41 52 46 Asian 41 41 41 ------------------------------------- Intimate partner violence, 1993-98 Between 1993 and 1998, an average of 1.1 million violent crimes were committed against persons by their current or former spouses, boyfriends, or girlfriends annually. An average of 87% of this violence, termed intimate partner violence, was committed against women. Between 1993 and 1998 intimate partner victimization of women differed by race. American Indians were victimized by an intimate at rates higher than those for all other females -- 23 American Indians per 1,000 persons age 12 or older compared to 11 blacks, 8 whites, and 2 Asians. Black females were victimized at higher rates than white and Asian females, and white females experienced violence by an intimate at rates higher than Asians. Average annual rate of inmate partner violence per 1,000 persons age 12 or older, 1993-98 Female Male White 8.1 1.3 Black 11.2 2.0 American Indian 23.2 4.2* Asian 1.9 0.3* *Based on 10 or fewer sample cases. Trends in intimate partner violence The rate of intimate partner violence against white females fell significantly between 1993 and 1998, from 10 victimizations to 8 victimizations per 1,000 women. The apparent decline in the rate of intimate partner violence against black females as measured by comparing the 1993 and 1998 rates was not significant. Insufficient sample sizes for American Indian and Asian females prevented examination of trends. Female victims of intimate partner violence, by year, 1993-98 Female victims of intimate partner violence White Black Number Rate Number Rate Total 4,560,740 8.1 961,380 11.2 1993 895,090 9.8 162,600 11.9 1994 813,670 8.8 174,470 12.5 1995 731,850 7.8 188,510 13.3 1996 689,170 7.3 177,530 12.3 1997 695,930 7.4 129,610 8.9 1998 735,040 7.7 128,660 8.7 Note: Multiple-offender victimizations are classified by the most intimate relationship between the victim and one of the offenders. There were too few cases of Asian and American Indian females to provide reliable estimates for each year. Among white males there was no discernible trend in the occurrence of intimate partner violence. The rate in 1998 was similar to the 1993 rate. There were too few sample cases for black, American Indian, and Asian males for estimation of changes in rates by year. Reporting of intimate partner violence Intimate partner violence is reported to police in lower percentages than violent crime in general. The percentage of intimate partner violence against females reported to police did not differ by the race of the victim, 1993-98. The only exception was that 66% of violence by intimates against black females was reported to police compared to 51% of violence against white females. The percentage of violence against white, Asian (52%), and American Indian (51%) females reported between 1993 and 1998 was similar (not shown in table). White male victims of intimate partner violence, by year, 1993-98 Number Rate Total 713,466 1.3 1993 136,380 1.6 1994 146,610 1.7 1995 104,050 1.2 1996 96,940 1.1 1997 87,370 1.0 1998 142,120 1.6 Note: Multiple-offender victimizations are classified by the most intimate relationship between the victim and one of the offenders. There were too few cases of black, Asian, an American Indian males to provide reliable estimates for each year. Reasons for not reporting the victimization were similar across the victims' racial groups, 1993-98. Violence by intimates against females was not reported to police most commonly because the victim stated it was a "private or personal matter." Other commonly stated reasons this violence was not reported to the police was because the victim "feared reprisal" and the victim wished to "protect the offender." For additional information see Intimate Partner Violence, BJS Special Report, May 2000, NCJ 178247, and Violence by Intimates, BJS Factbook, March 1998, NCJ 167237, available on the BJS website. Reasons for not reporting intimate partner violence to the police, females, 1993-98 Female victims of intimate partner violence White Black Average Average annual Percent annual Percent Private or personal matter 124,210 34% 19,400 36% Afraid of reprisal 70,760 19 10,390 19 Protect offender 45,630 12 6,830 13 Small/no loss 27,500 7 1,380* 3* Police will not bother 21,570 6 2,850* 5* Other reason given 147,850 40 26,290 49 Note: Percentages may not total to 100% because respondents could suggest more than one reason. "Other reason given" include responses such as "police ineffectiveness or biased," "not clear a crime occurred," "inconvenient," and "reported to another official." There were too few cases of Asians and American Indians to provide reliable individual year estimates. *Based on 10 or fewer sample cases. --------------------------------------- Victim-offender characteristics, 1993-98 Victim-offender relationship and gender Males were victimized by strangers at significantly higher percentages than by nonstrangers regardless of race, 1993-98. When the offender was someone known to the male victim, it was most often a friend or acquaintance. American Indian males were more likely to identify the offender as an "other relative" than were males of other races. "Other relatives" can include a parent, stepparent, child, stepchild, brother, sister, or other related person. Asian males were victimized by a stranger at higher percentages, and by a nonstranger at lower percentages, than other males. Victim-offender relationships among females differed somewhat across racial categories. Black, white, and American Indian females were more likely to be victimized by someone they knew, 1993-98. Asian females were equally likely to be victimized by a stranger or a nonstranger during this time. Asian females were less likely to be victimized by a nonstranger, and more likely to be victimized by a stranger, than other females. Asian females were equally likely as American Indian females to be victimized by an intimate, and less likely than white or black females to be victimized by an intimate. Asian females were significantly more likely to say that the offender was a friend or acquaintance than were white or black females. Victim and offender race Violent crime was primarily intraracial for black and white victims. Sixty-six percent of white victims and 76% of black victims were victimized by an offender of the same race, 1993-98 (when the race of the offender was known). Violent crime was primarily interracial for American Indian and Asian victims. Fifty-eight percent of the American Indian victims and 35% of the Asian victims of violence stated the offender was white. Offenders were identified as black by 10% of the American Indian and 26% of the Asian victims. A quarter of American Indian and about a third of Asian victims were victimized by an offender described as "other races." Because the NCVS collects data for offenders whom victims describe as white, black, and "other race," it is not possible to report on American Indian or Asian offenders separately. -------------------------------- Further reading or other data This report and others like it, as well as the original data, are available on the BJS website, http://www.ojp.usdoj.gov. Some BJS reports that include aspects of race in their discussions of crime include Homicide Trends in the United States: 1998 Update. BJS Crime Data Brief, March 2000, NCJ 179767; American Indians and Crime, BJS report, February 1999, NCJ 173386; Criminal Victimization 1999, Changes 1998-99 with Trends 1993-99, BJS Bulletin, August 2000, NCJ 182734; and Intimate Partner Violence, BJS Special Report, May 2000, NCJ 178247. See also Michael Hindelang, Michael Gottfredson, and James Garofalo, Victims of Personal Crimes: An Empirical Foundation for a Theory of Personal Victimization, Cambridge, MA: Ballinger, 1978. ------------------------------------ Survey methodology This Special Report presents data on rape, sexual assault, robbery, aggravated assault, and simple assault from the National Crime Victimization Survey (NCVS), and data on homicide from the FBI's Uniform Crime Reporting Program. The NCVS gathers data on crimes against persons age 12 or older, reported and not reported to the police, from a nationally representative sample of U.S. households. The NCVS provides information about victims (age, gender, race, ethnicity, marital status, income, and educational level), offenders (gender, race, approximate age, and victim-offender relations) and the nature of the crime (time and place of occurrence, use of weapons, nature of injury, and economic consequences). Between 1993 and 1998 approximately 293,400 households and 574,000 individuals age 12 or older were interviewed. For the NCVS data presented, response rates varied between 93% and 96% of eligible households and between 89% and 92% of eligible individuals. In some instances the sample size used to generate an estimate is small. For instance the sample of Asian males experiencing intimate partner violence was fewer than ten for every year from 1993 through 1998. While the estimate is reliable it is also likely associated with a relatively large confidence interval and should be viewed with caution. Testing trends and annual differences in violent victimization When a statement is made describing the increase or decrease of a linear trend, it was tested using a linear trend test. This test compares the slope of the trend to a horizontal line that has a slope of zero, determining whether the slope generated from a change in an estimate differs from a flat trend. The linear trend test was conducted as a regression equation with an independent variable of time and a dependent variable of the victimization rate. A regression coefficient (b) and its corresponding standard error (F) are computed, and a t-statistic -- the ratio b/F C is calculated. If the t-statistic is greater than 1.96 for a two-tailed test (or greater than 1.645 for a one-tailed test), the slope is considered significantly different from zero. If the t-statistic is greater than 1.645 for a two-tailed test (or 1.28 for a one-tailed test), the slope is considered slightly different from zero. If the t-statistics are less than the critical values, the analyst must conclude that the change in the estimate did not differ significantly from zero and was not statistically significant. Standard error computations Comparisons of percentages and rates made in this report were tested to determine if observed differences were statistically significant. Differences described as higher, lower, or different passed a hypothesis test at the .05 level of statistical significance (95% confidence level). The tested difference was greater than twice the standard error of that difference. For comparisons that were statistically significant at the 0.10 level (90% confidence level), "somewhat," "slightly," or "marginally" is used to note the nature of the difference. Caution is required when comparing estimates not explicitly discussed in this Special Report. What may appear to be large differences may not test as statistically significant at the 95% or the 90% confidence level. Significance testing calculations were conducted at the Bureau of Justice Statistics using statistical programs developed specifically for the NCVS by the U.S. Bureau of the Census. These programs take into consideration many aspects of the complex NCVS sample design when calculating generalized variance estimates. Definitions In the NCVS the household respondent identifies the race of each household member by choosing from a flashcard. These categories and their corresponding race defined in this report follow: Flashcard category Report category White White Black Black American Indian American Indian Aleut American Indian Eskimo American Indian Asian (Japanese, Chinese Asian Korean, Asian Indian, Vietnamese, and other Asian) Pacific Islander (Filipino Asian Hawaiian, Guamanian, Samoan, and other Asian) Violent acts covered in this report include murder, rape, sexual assault, robbery, aggravated assault, and simple assault. Overall violent crime is a combination of each type of crime. Serious violent crime includes all types except simple assault: rape/sexual assault, robbery, and aggravated assault. Note that for most tables homicide was not included in the analyses because data are not available for Asians and American Indians. Definitions are as follows: Murder and nonnegligent manslaughter are defined as the willful killing of one human being by another. Rape is forced sexual intercourse, including both psychological coercion and physical force. Forced sexual intercourse means vaginal, anal, or oral penetration by the offender(s). This category includes incidents where the penetration is from a foreign object such as a bottle. Also included are attempted rapes, male and female victims, and heterosexual and homosexual rape. Sexual assault covers a wide range of victimizations distinct from rape or attempted rape. These crimes include completed or attempted attacks generally involving unwanted sexual contact between the victim and offender. Sexual assaults may or may not involve force and include such things as grabbing or fondling. Sexual assault also includes verbal threats. Robbery is a completed or attempted theft directly from a person, of property or cash by force or threat of force, with or without a weapons, and with or without an injury. Aggravated assault is defined as a completed or attempted attack with a weapon, regardless of whether or not an injury occurred, and an attack without a weapon in which the victim is seriously injured. Simple assault is an attack without a weapon resulting either in no injury, minor injury (such as bruises, black eyes, cuts, scratches or swelling) or an undetermined injury requiring less than 2 days of hospitalization. Simple assaults also include attempted assaults without a weapon. ------------------------------- The Bureau of Justice Statistics is the statistical agency of the U.S. Department of Justice. Lawrence A. Greenfeld is acting director. BJS Special Reports address a specific topic in depth from one or more datasets that cover many topics. Callie Rennison, Ph.D., wrote this report under the supervision of Michael Rand. Cathy Maston provided statistical assistance and verification. Tom Hester and Ellen Goldberg edited the report. Jayne Robinson prepared the report for publication. March 2001, NCJ 176354 End of file ih 3/2/01