SECTION V The data in Tables 5.2 and 5.3 in the special report "Violence Among Family Members and Intimate Partners" were transposed. The numbers have been corrected and the study has been revised as appropriate. The FBI regrets this error. January | VIOLENCE AMONG FAMILY MEMBERS AND INTIMATE PARTNERS INTRODUCTION The phenomenon of violence among family members has been present in Western society throughout its history. It is a significant societal as well as an individual problem, but it has not always been considered a crime. History records instances of wife beating as early as the time of the Roman Empire. Further, the English common law as codified by jurist Sir William Blackstone in 1768 affirmed the right of a husband to physically chas- tise his wife as long as "the stick was no bigger than his thumb." This right was upheld by an appellate court in North Carolina as late as 1867. 1 M. A. Straus and R. J. Gelles, who have authored several works about fam- ily violence, also categorized instances of child abuse throughout history. Some of the cases they examined date to bibli- cal times. "Infanticide, mutilation, and other forms of violence were legal pa- rental prerogatives from ancient Rome to colonial America." 2 Child abuse was identified as a so- cial problem by church and social work- ers in the last quarter of the nineteenth century. However, it was not until C. Henry Kemper published his 1962 study, "The Battered Child Syndrome" 3 that child abuse found its way onto the pub- lic agenda. Likewise, it was not until the 1970s that wife beating was recognized as a problem and that significant schol- arly research on spousal abuse began. In their writings, Straus and Gelles (1988) and Straus (2000) listed some of the factors that led to the reformation in our society's view of family violence. Those factors included the social move- ments of the 1960s that undertook to aid oppressed groups; the growth in paid employment of married women; the re- emergence of the women's movement in the 1970s; the provision of shelters for battered women; public abhorrence of violence evidenced by the rising homi- cide and assault rates; violent political and social protests; assassinations; terror- ist activity; the Vietnam War; the critical reassessment of the family; and changes in theoretical perspectives in sociology, family studies, and criminology. 4 Measuring Domestic Violence The subject of domestic violence is broad in scope and there are many ways to measure it. For example, the Department of Justice's National Crime Victimization Survey (NCVS) questions individuals regarding their victimiza- tion experiences. Investigators from other agencies examine hospital records and physicians' reports to determine the frequency of broken bones and use that information as evidence of child or spousal abuse. 5 The present work investigates the problem of violence among intimate partners and other family members by examining the incidents reported to law enforcement who, in turn, submitted data to the Uniform Crime Reporting (UCR) Program. The years considered are 1996 through 2001. Although there are other studies of this criminal phe- nomenon from the vantage point of the victim or from a public health perspec- tive, this study is confined to the experi- ences of victims in close relationships with their offenders. Some additional data presented in this report are from other sources and are tendered to un- derline the nature of the phenomenon. However, those data are presented only as background information. Data from the UCR Program clearly demonstrate that violence among family members is a prevalent prob- lem. For instance, the Program's 1996 Supplementary Homicide Report 6 (SHR) showed that 30 percent of all female victims of murder or nonnegligent man- slaughter in the U.S. were killed by their husbands, ex-husbands, or boyfriends. 7 The 2000 SHR data indicated that of the 3,173 women homicide victims for which supplemental data were provided, 1,029 were killed by their husbands, former husbands, or boyfriends. Further, data from the UCR Program's National Inci- dent-Based Reporting System (NIBRS) for 2001 showed that an estimated 38,614 women were beaten and/or sexually as- saulted by family members. 8 Intimate Partner and Spousal Abuse Domestic violence takes many forms including intimate partner and spousal abuse, child abuse, and elder abuse. Regarding spousal abuse, data from the American Psychological Associa- tion (APA) 9 indicate that one-third of all adult women will be assaulted by a partner during adulthood. The Centers for Disease Control and Prevention re- ported that "nearly two-thirds of women who reported being raped, physically assaulted, or stalked since the age of 1 8 were victimized by a current or former husband, cohabiting partner, boyfriend, or date." 10 Further, one in three of these women were injured. 11 Reports from the NCVS from 1992 to 1996 showed that, without adjusting for socioeconomic status, an average of 12 per 1,000 black women experienced violence by an intimate partner compared to an estimated 8 per 1,000 white women. 12 SPECIAL REPORT 339 In studies of visits to hospital emergency rooms in 1994, the Bureau of Justice Statistics reported that women accounted for nearly 40 percent of all the patients in need of treatment for vio- lent victimizations. Thirty- six percent of these victims were attacked by their intimate partners. 13 Female victims were more likely than male victims to require medical attention, take time off work, and spend more days in bed. 14 More- over, the National Research Council argues that the psychological costs for these victims are quite high and "can in- clude depression, suicidal thoughts and attempts, lowered self-esteem, alcohol and other drug abuse, and post-traumatic stress disorder." 15 According to Straus and Gelles, perpetrators of violence are more likely to have had a history of physical or sexual abuse themselves or were victims of threats of abuse. Furthermore, men who abuse their partners are more likely to abuse their children. 16 Both victims and perpetrators of domestic violence are more likely to abuse alcohol. Statistics from the Na- tional Institute on Alcohol Abuse and Alcoholism show that more than 50 per- cent of male batterers and 20 percent of female victims are alcohol abusers. 17 Surveys taken by the NCVS be- tween 1992 and 1996 indicated that financial losses to women victims of non-lethal intimate violence amounted to more than $150 million per year. This amount was made up of medical costs (approximately 40 percent), property losses (about 44 percent), and the rest comprised lost pay. 18 Child Abuse Men are more likely to be the offenders in cases of physical and sexual abuse against children. Approximately 10 percent of all injuries to children under 7 years of age who are examined in emer- gency rooms come from abuse. 19 More than 50 percent of murder victims under the age of 12 are killed by a parent. About 3.3 million children each year witness acts of violence by family members against their mothers or female caretakers. The APA estimates that 16 to 34 percent of girls and 10 to 20 percent of boys are sexually abused, most often by a family member or trusted family friend. The APA has for a long time indicated that children who experience violence are at greater risk of becoming adult abusers. The Associa- tion terms this the "cycle of violence." 20 Children at risk for being abused include those who are unwanted, who have physical or mental disabilities, and whose parents are under stress (e.g., par- ents with more than four children, those who make less than $15,000 annually, those who abuse drugs, or young moth- ers who are isolated from others outside the family.) 21 The U.S. Advisory Board on Child Abuse and Neglect reports that there are particular characteristics that are associ- ated with child abusers. Usually, the offenders are in their mid- 20s, do not have high school educations, live at or below the poverty level, suffer from de- pression, and may have difficulty coping with stressful situations. 22 Elder Abuse Elder abuse affects thousands of indi- viduals each year, but according to the National Center on Elder Abuse, 23 the incidents are underreported. Few stud- ies examine this topic; however, a 1997 study of case reports of various protec- tive agencies by the National Center on Elder Abuse found that neglect is the most common form of elder maltreat- ment in domestic settings, and adult children are the most frequent abusers of the elderly. From the data that were available, authors Tatara, Kuzmeskus, and Duckhorn (1997) found that cases of elder neglect increased substantially over the years 1990 to 1996, rising from 47 percent in 1990 to 55 percent in 1996. 24 Also according to Tatara, Kuzmeskus, and Duckhorn, most elderly victims of abuse were female, but from 1990 to 1996, the gap between male and female victims narrowed somewhat, changing from 68.3 percent female/31.5 percent male in 1990 to 67.3 percent female/32.4 percent male in 1996. 25 Ad- ditionally, they found that nearly a third of the murders of victims 60 years of age or older were committed by a fam- ily member. Further, most elder abuse was committed by someone with whom the elderly victim lived. Because most caregivers for the elderly are women, they found that most of the neglect cases were committed by female family members. On the other hand, the most frequent offenders of physical abuse against the elderly were male family members. 26 OBJECTIVES This study examines violent crime inci- dents in which at least one of the offend- ers and one of the victims are related within the family. The crimes included in this analysis are murder and nonneg- ligent manslaughter, forcible rape, robbery, aggravated assault, simple assault, intimi- dation, forcible sodomy, sexual assault with an object, and forcible fondling. The relationships included in this study fall into the categories of fam- ily members and intimate partners and include spouse, common-law spouse, parent, sibling, child, grandparent, grandchild, in-law, stepparent, stepchild, stepbrother or stepsister, boyfriend, girl- 340 CRIME IN THE UNITED STATES friend, child of boyfriend/girlfriend, ex- wife, ex-husband, and other family member. The general objective of this study is to analyze the domestic violence data that are provided in the UCR Program's NIBRS data. It will show the types of crimes that are committed in domestic disputes (e.g., assaults, rapes, and sexual assaults). The relationships of the individuals involved (i.e., partner or ex-partner, parent, or other relationship) are examined. Further, variables such as the number and degree of injury in the cases, the weapons used, and the severity of the sustained injuries are included. Study Question 1 — Characteristics of the Incidents and Offenses The level of analysis in this study question is the incident itself. In the NIBRS data that were used in this study, an incident includes all the family violence offenses within a single incident, whether the offense is against an intimate partner, a child, or an elder. Variables that describe the incident such as the number of incidents per year, the use of alcohol, and the violence involved (i.e., homicides, injuries, and types of weapons used) are addressed in question 1 . Study Question 2 — Victims, Offenders, and Relationship Status Question 2 concerns the victim and offender characteristics. The age, sex, and race of the victims and offenders are examined here. The incidents are broken down by the selected relationships of victim to offender (intimate partner, child/offspring, or elderly relative). This question also concerns the relationships of the victims to the offenders. In this section, different crime categories are examined by types of incidents to show the similarities and differences between them. DATA The UCR's National Incident-Based Reporting System (NIBRS) data from 1996-2001 Data for this study came from the UCR Program's NIBRS database. The NIBRS, which is the redesigned, expanded version of the Program's original Summary system, was established in the 1980s, and a limited number of agencies began submitting data to the FBI via the NIBRS in January 1989. This database contains information on incidents and arrests reported by the participating local, county, and state law enforcement agencies. The NIBRS collects data for 22 crime categories and includes information about each incident, the offenses committed within the incident, and details about the victim and offender. The data collected by this method provide a rich, disaggregated source of information that can be used to enhance law enforcement and crime research as well as assist officials in strategic and administrative decision-making. METHODS The years considered for this study are 1996 through 2001. Frequency distributions and cross tabulations are used to explore the data and to address the Study Questions. For this report, relationships that fall into the spousal abuse category are defined as those in which the victim and offender were related as spouse, common-law spouse, ex-spouse, boyfriend, or girlfriend. Child abuse cases are defined as those in which at least one victim was below age 18. However, when relationships are considered in the data presented in this study, the term child can also mean the offspring (adult or juvenile) of a victim or offender. Footnotes are provided in the appropriate tables to clarify how this classification applies. Elderly abuse cases are defined as those in which as least one victim was above age 65 and had a familial relationship to one of the offenders. FINDINGS Incident Characteristics Number of Incidents and Offenses Table 5.1 shows the total number of incidents reported to the UCR Program via the NIBRS for each year from 1996 through 2001. The number of incidents reflect violent, property, and society crimes. As expected, the numbers steadily increased over the period as more jurisdictions began reporting data via the NIBRS. The total number of incidents over the period was 12,545,546, and of those, 2,985,101 (23.8 percent) contained at least one violent offense. Table 5.2 presents the number of incidents reported each year containing at least one violent offense. During the timeframe of this study, simple assault was the most prevalent violent crime, present in 58.5 percent of the total violent incidents. Aggravated assault and intimidation followed comprising 16.2 percent and 15.6 percent, respectively. Table 5.1 Number of Incidents Reported in NIBRS, 1996-2001 Percentage Incidents of Incidents containing at containing at least one violent least one Year Total Incidents crime violent crime 1996 1,064,763 249,872 23.5 1997 1,426,978 325,921 22.8 1998 1,822,675 424,728 23.3 1999 2,157,326 518,975 24.1 2000 2,841,523 689,641 24.3 2001 3,232,281 775,964 24.0 Total 12,545,546 2,985,101 23.8 SPECIAL REPORT 341 Table 5.2 Number of Incidents with a Violent Crime by Crime Type, 1996-2001 Number of 7996 1997 1998 7999 2000 2001 Incidents Murder/Nonnegligent Manslaughter 594 702 929 1,139 1,570 1,820 6,754 Negligent Manslaughter 67 59 95 104 146 138 609 Justifiable Homicide 1 18 20 14 25 40 58 175 Forcible Rape 4,929 6,819 9,755 10,931 14,298 16,204 62,936 Forcible Sodomy 1,327 1,726 2,466 2,946 3,392 3,819 15,676 Sexual Assault With An Object 737 1,073 1,364 1,568 2,163 2,163 9,068 Forcible Fondling 5,854 7,733 10,289 12,856 16,603 17,796 71,131 Incest 194 282 296 363 433 458 2,026 Statutory Rape 951 1,056 1,387 1,820 2,344 2,806 10,364 Robbery 11,805 14,708 19,440 22,680 34,569 42,855 146,057 Aggravated Assault 49,083 60,523 72,815 80,723 105,398 114,002 482,544 Simple Assault 144,060 187,779 248,069 305,199 406,996 454,558 1,746,661 Intimidation 35,263 47,385 62,066 83,743 109,189 127,512 465,158 1 While technically not a crime, justifiable homicide was included in the violent offense group due to its particular relationship to family violence situations. Table 5.3 also shows the prevalence of the violent offenses examined in this study. The highest percentages of violent offenses for each year were for simple assault (58.2 percent overall), aggravated assault (16.7 percent overall), and intimidation (14.8 percent overall). Each of the remaining crime categories reflected percentages less than 5.0 percent. Relationships in Violent Offenses Tables 5.4 and 5.5 show the relationships of the victims to the offenders in the violent offenses studied. Table 5.4 shows all relationship categories that were available and that applied to this study. Of the 3,368,347 violent offenses reported during the period, there were 3,534,254 confrontations for which the UCR Program knew the relationships of victims to offenders. Of these, 1,551,143 were familial relationships, and the totals for each of these categories are provided in Table 5.5, broken down by year. The most prevalent relationship was boyfriend/ girlfriend (29.6 percent) followed by spouse (24.4 percent). When spouse, common-law spouse, and ex- spouse were considered together, the percentage of the total rose to 32.4 percent. Table 5.6 shows violent incidents by the type of abuse being studied. Simple assault was the most prevalent offense in all three relationship categories followed by aggravated assault and intimidation in the spousal abuse category. In the case of child abuse, simple assault was the most prevalent offense, followed by combined sexual assaults, then aggravated assault, and intimidation. In the elderly abuse categories, simple assault comprised the largest offense total, followed by intimidation, robbery, and aggravated assault. Most sex offenses (i.e., forcible rape, forcible sodomy, sexual assault with an object, forcible fondling, incest, and statutory rape) fall in the category of child abuse, comprising approximately 17 percent of those offenses. Weapons The weapons used in violent offenses in which there was a familial relationship are broken down by year in Table 5.7. For each year studied, the most prevalent weapon used were those categorized by the UCR as "personal" (i.e., hands, fists, or feet). Nearly 70 percent of violent offenses involving familial relationships were carried out using this type of weapon. Following personal weapons, no weapons (1 1.4 percent) and knives, handguns, and blunt objects, (between 3 and 5 percent) were used most often to carry out the most common offenses reported for the period. Weapon use in offenses involving familial relationships broken down by the three types of relationships studied is presented in Table 5.8. Personal weapons were used most often in cases of spousal and child abuse, (78.4 percent and 73.3 percent, respectively). However, in the elderly abuse category, personal weapons constituted only 36.1 percent of the offenses involving weapons. Almost 44 percent of elder abuse offenses involved the use of a handgun. 342 CRIME IN THE UNITED STATES Table 5.3 Number of Offenses, by Violent Crime, 1996-2001 1996 1997 1998 1999 2000 2001 Total Offenses Percent Offenses Percent Offenses Percent Offenses Percent Offenses Percent Offenses Percent Offenses Percent Murder/Nonnegligent Manslaughter 643 0.2 749 0.2 1,007 0.2 1,275 0.2 1,695 0.2 1,958 0.2 7,327 0.2 Negligent Manslaughter 72 0.0 62 0.0 112 0.0 127 0.0 158 0.0 142 0.0 673 0.0 Justifiable Homicide 1 18 0.0 21 0.0 15 0.0 25 0.0 40 0.0 59 0.0 178 0.0 Forcible Rape 5,065 1.8 7,039 1.9 10,567 2.2 11,778 2.0 14,745 1.9 16,847 1.9 66,041 2.0 Forcible Sodomy 1,496 0.5 1,935 0.5 2,975 0.6 3,585 0.6 3,816 0.5 4,316 0.5 18,123 0.5 Sexual Assault With An Object 800 0.3 1,150 0.3 1,540 0.3 1,774 0.3 2,362 0.3 2,364 0.3 9,990 0.3 Forcible Fondling 6,692 2.4 8,888 2.4 12,450 2.6 15,302 2.6 18,805 2.4 20,063 2.3 82,200 2.4 Incest 219 0.1 320 0.1 331 0.1 425 0.1 471 0.1 524 0.1 2,290 0.1 Statutory Rape 1,004 0.4 1,086 0.3 1,582 0.3 2,022 0.3 2,422 0.3 2,888 0.3 11,004 0.3 Robbery 11,805 4.2 14,708 4.0 19,440 4.0 22,680 3.9 34,569 4.5 42,855 4.9 146,057 4.3 Aggravated Assault 56,805 20.2 70,172 19.2 85,472 17.7 94,265 16.0 122,391 15.8 134,270 15.3 563,375 16.7 Simple Assault 159,792 56.7 208,193 57.1 279,670 58.0 344,155 58.6 456,389 58.9 513,592 58.5 1,961,791 58.2 Intimidation 37,389 13.3 50,315 13.8 66,986 13.9 90,010 15.3 116,854 15.1 137,744 15.7 499,298 14.8 Total Violent Offenses 281,800 100.0 364,638 100.0 482,147 100.0 587,423 100.0 774,717 100.0 877,622 100.0 3,368,347 100.0 1 While technically not a crime, justifiable homicide was included in the violent offense group due to its particular relationship to family violence situations. Due to rounding, the percent of total may not add to 100.0 percent. Table 5.4 Number of Confrontations Between Victim and Offender by Relationship, 1996-2001 Year Relationship 1996 1997 1998 1999 2000 2001 Total to Offender Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent Spouse 33,432 11.3 42,880 11.2 54,552 10.8 67,662 11.0 87,681 10.8 92,896 10.1 379,103 10.7 Common-Law Spouse 7,225 2.5 9,371 2.5 13,253 2.6 14,910 2.4 17,229 2.1 17,105 1.9 79,093 2.2 Parent 6,910 2.3 9,222 2.4 12,200 2.4 15,842 2.6 21,028 2.6 24,171 2.6 89,373 2.5 Sibling 7,505 2.5 9,605 2.5 12,815 2.5 16,177 2.6 21,206 2.6 24,588 2.7 91,896 2.6 Child 7,312 2.5 10,012 2.6 14,313 2.8 18,185 3.0 23,943 2.9 26,567 2.9 100,332 2.8 Grandparent 315 0.1 452 0.1 618 0.1 783 0.1 1,122 0.1 1,348 0.2 4,638 0.1 Grandchild 502 0.2 606 0.2 939 0.2 1,095 0.2 1,344 0.2 1,440 0.2 5,926 0.2 In-Law 2,346 0.8 2,854 0.8 3,595 0.7 4,450 0.7 5,389 0.7 6,001 0.7 24,635 0.7 Stepparent 1,165 0.4 1,485 0.4 2,258 0.5 3,000 0.5 3,804 0.5 4,345 0.5 16,057 0.5 Stepchild 1,920 0.7 2,413 0.6 3,464 0.7 4,300 0.7 5,509 0.7 6,037 0.7 23,643 0.7 Stepsibling 326 0.1 493 0.1 720 0.1 890 0.1 1,266 0.2 1,335 0.1 5,030 0.1 Other Family Member 5,784 2.0 7,313 1.9 10,947 2.2 14,336 2.3 19,112 2.4 22,435 2.4 79,927 2.3 B oy friend/ Girlfriend 35,805 12.1 48,650 12.8 62,133 12.3 77,247 12.6 108,280 13.3 126,556 13.7 458,671 13.0 Child of Boyfriend/Girlfriend 808 0.3 1,054 0.3 1,482 0.3 1,840 0.3 2,404 0.3 2,894 0.3 10,482 0.3 Ex-Spouse 3,879 1.3 4,922 1.3 6,213 1.2 7,911 1.3 10,055 1.2 11,659 1.3 44,639 1.3 Acquaintance 72,411 24.5 88,514 23.2 110,506 21.8 124,845 20.3 157,013 19.3 176,002 19.1 729,291 20.6 Friend 8,379 2.8 10,051 2.6 13,796 2.7 17,086 2.8 22,720 2.8 24,422 2.7 96,454 2.7 Neighbor 4,703 1.6 6,389 1.7 8,670 1.7 10,845 1.8 13,950 1.7 15,863 1.7 60,420 1.7 Babysittee (the baby) 346 0.1 455 0.1 595 0.1 665 0.1 742 0.1 778 0.1 3,581 0.1 Homosexual relationship 296 0.1 492 0.1 757 0.2 968 0.2 1,578 0.2 2,027 0.2 6,118 0.2 Employee 725 0.3 1,123 0.3 1,338 0.3 1,742 0.3 2,388 0.3 2,821 0.3 10,137 0.3 Employer 572 0.2 862 0.2 1,034 0.2 1,391 0.2 1,731 0.2 1,922 0.2 7,512 0.2 Stranger 36,725 12.4 44,114 11.6 53,302 10.5 60,622 9.9 83,666 10.3 95,169 10.3 373,598 10.6 Victim was Offender 1 17,447 5.9 22,003 5.8 27,873 5.5 34,016 5.5 47,280 5.8 55,772 6.1 204,391 5.8 Otherwise Known 13,692 4.6 20,979 5.5 30,918 6.1 41,675 6.8 57,798 7.1 67,205 7.3 232,267 6.6 Unknown 24,920 8.4 35,092 9.2 58,134 11.5 72,986 11.9 95,850 11.8 110,058 11.9 397,040 11.2 TOTAL 295,450 100.0 381,406 100.0 506,425 100.0 615,469 100.0 814,088 100.0 921,416 100.0 3,534,254 100.0 'The category "Victim was Offender" was used in cases where all of the participants in an incident were victims and offenders of the same offense such as domestic disputes where both husband and wife are charged with assault, double murders, etc. Due to rounding, the percent of total may not add to 100.0 percent. SPECIAL REPORT 343 Table 5.5 Number of Confrontations Between Victim and Offender by Family Relationship 1 1996-2001 Year Relationship 1996 1997 1998 1999 2000 2001 Total to Offender Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent Number Percent Spouse 33,432 26.4 42,880 25.9 54,552 25.0 67,662 24.9 87,681 24.3 92,896 22.8 379,103 24.4 Common-Law Spouse 7,225 5.7 9,371 5.7 13,253 6.1 14,910 5.5 17,229 4.8 17,105 4.2 79,093 5.1 Parent 6,910 5.5 9,222 5.6 12,200 5.6 15,842 5.8 21,028 5.8 24,171 5.9 89,373 5.8 Sibling 7,505 5.9 9,605 5.8 12,815 5.9 16,177 6.0 21,206 5.9 24,588 6.0 91,896 5.9 Child 7,312 5.8 10,012 6.0 14,313 6.6 18,185 6.7 23,943 6.6 26,567 6.5 100,332 6.5 Grandparent 315 0.3 452 0.3 618 0.3 783 0.3 1,122 0.3 1,348 0.3 4,638 0.3 Grandchild 502 0.4 606 0.4 939 0.4 1,095 0.4 1,344 0.4 1,440 0.4 5,926 0.4 In-Law 2,346 1.9 2,854 1.7 3,595 1.7 4,450 1.6 5,389 1.5 6,001 1.5 24,635 1.6 Stepparent 1,165 0.9 1,485 0.9 2,258 1.0 3,000 1.1 3,804 1.1 4,345 1.1 16,057 1.0 Stepchild 1,920 1.5 2,413 1.5 3,464 1.6 4,300 1.6 5,509 1.5 6,037 1.5 23,643 1.5 Stepsibling 326 0.3 493 0.3 720 0.3 890 0.3 1,266 0.4 1,335 0.3 5,030 0.3 Other Family Member 5,784 4.6 7,313 4.4 10,947 5.0 14,336 5.3 19,112 5.3 22,435 5.5 79,927 5.2 B oy friend/ Girlfriend 35,805 28.3 48,650 29.3 62,133 28.4 77,247 28.4 108,280 30.0 126,556 31.1 458,671 29.6 Child of Boyfriend/Girlfriend 808 0.6 1,054 0.6 1,482 0.7 1,840 0.7 2,404 0.7 2,894 0.7 10,482 0.7 Ex-Spouse 3,879 3.1 4,922 3.0 6,213 2.8 7,911 2.9 10,055 2.8 11,659 2.9 44,639 2.9 Victim was Offender 9,744 7.7 12,318 7.4 15,868 7.3 19,239 7.1 27,397 7.6 32,634 8.0 1 17,200 7.6 Otherwise Known 393 0.3 572 0.3 1,226 0.6 1,943 0.7 2,301 0.6 2,506 0.6 8,941 0.6 Unknown 1,373 1.1 1,586 1.0 1,890 0.9 1,975 0.7 2,256 0.6 2,477 0.6 11,557 0.8 TOTAL 126,744 100.0 165,808 100.0 218,486 100.0 271,785 100.0 361,326 100.0 406,994 100.0 1,551,143 100.0 'The additional relationship codes were included as family relationships in this table. The category Victim was Offender is used in cases where all of the participants in an incident were victims and offenders of the same offense such as domestic disputes were both husband and wife are charged with assault, double murders, etc. Otherwise Known and Unknown categories were also included due to the possibility that many of these encounters would include family relationships not otherwise captured. Due to rounding, the percent of total may not add to 100.0 percent Table 5.6 Number of Incidents Containing at Least One Violent Offense by Family Relationship, 1996-2001 Spouse Child 1 Elderly relative Murder/Nonnegligent Manslaughter 1,226 1,061 444 Negligent Manslaughter 38 200 51 Justifiable Homicide 7 14 5 Forcible Rape 8,195 33,644 432 Forcible Sodomy 852 12,112 78 Sexual Assault with an Object 612 6,588 71 Forcible Fondling 1,920 57,941 363 Incest 88 1,789 5 Statutory Rape 2,765 10,570 3 Robbery 1,801 19,080 7,140 Aggravated Assault 115,769 102,675 6,919 Simple Assault 647,286 397,775 20,955 Intimidation 93,173 76,303 11,229 1 The category of child refers to victims under the age of 18. Victims, Offenders, and Relationships Substance Abuse Table 5.9 shows the number of family violence incidents in which substance abuse was involved. The overwhelming majority of these situations involved alcohol, which was used in approximately 99 percent of violent family incidents for which there was a substance abuse code. Substance abuse in offenses involving the three domestic relationships studied is presented in Table 5.10. In all three relationships, more than 99 percent of the offenses involving abused substances involved alcohol. Gender, Race, and Age Victims of violent family crimes tend to be female. Table 5.11 presents the overall breakdown, indicating that 74.8 percent of the victims were female. 344 CRIME IN THE UNITED STATES Table 5.7 Percent of 1996 Percent 7997 Percent 1998 Percent 1999 Percent 2000 Percent 2001 Percent Total total 1,678 0.8 2,184 0.8 2,506 0.7 3,478 0.8 5,195 0.9 5,245 1.0 20,286 0.9 9,330 4.4 10,808 3.9 13,214 3.7 15,165 3.6 22,821 4.1 24,791 4.6 96,129 4.1 712 0.3 857 0.3 1,084 0.3 1,213 0.3 1,481 0.3 1,418 0.3 6,765 0.3 1,313 0.6 1,471 0.5 1,741 0.5 2,076 0.5 2,545 0.5 2,485 0.5 11,631 0.5 384 0.2 507 0.2 757 0.2 681 0.2 969 0.2 965 0.2 4,263 0.2 11,399 5.3 13,421 4.9 16,535 4.6 19,214 4.5 24,747 4.4 25,510 4.7 110,826 4.7 11,128 5.2 12,222 4.4 13,676 3.8 14,582 3.4 19,813 3.5 19,841 3.7 91,262 3.9 2,445 1.1 3,093 1.1 4,313 1.2 5,123 1.2 7,101 1.3 7,811 1.5 29,886 1.3 >4,076 38 42 117 21 5,084 16,112 13,879 72.0 0.0 0.0 0.1 0.0 2.4 7.5 100.0 197,632 38 37 145 34 9,385 23,832 275,666 71.7 0.0 0.0 0.1 0.0 3.4 8.7 100.0 247,287 67 72 147 59 10,991 46,168 358,617 69.0 0.0 0.0 0.0 0.0 3.1 12.9 100.0 287,413 72 104 205 111 13,969 60,623 424,029 67.8 0.0 0.0 0.1 0.0 3.3 14.3 100.0 381,916 79 132 275 133 27,199 65,381 559,787 68.2 0.0 0.0 0.1 0.0 4.9 11.7 100.0 369,559 89 125 293 100 21,992 59,032 539,256 68.5 0.0 0.0 0.1 0.0 4.1 11.0 100.0 1,637,883 383 512 1,182 458 88,620 271,148 2,371,234 69.1 0.0 0.0 0.1 0.0 3.7 11.4 100.0 Table 5.8 Use of Weapons Within Family Violence Incidents by Family Relationship, 1996-2001 Spouse Percent Child 1 Percent Elderly relative Pei Firearm (Type Unknown) 1,679 0.2 91 0.1 2,253 Handgun 8,997 1.2 450 0.6 11,989 Rifle 1,464 0.2 115 0.1 67 Shotgun 2,189 0.3 130 0.2 103 Other Firearm 202 * 29 * 7 Knife/Cutting Instrument 26,415 3.5 1,287 1.6 396 Blunt Object 17,721 2.3 2,078 2.6D 593 Motor Vehicle 6,867 0.9 369 0.5 72 Personal Weapons (hands, fists, feet, etc.) 599,072 78.4 58,141 73.3 9,889 Poison 78 * 5 * 10 Explosives □ 22 * 2 m.- 1 Fire/Incendiary Device 331 * 73 O.I 19 Asphyxiation 200 * 37 0.1 8 Unknown Weapon 19,787 2.6 3,490 4.4 475 No Weapon 79,397 10.4 13,040 16.4 1,542 TOTAL 764,421 100.0 79,337 100.0 27,424 4 3 10 Use of Weapons in Violent Offenses Within Family Relationships, 1996-2001 Firearm (Type Unknown) Handgun Rifle Shotgun Other Firearm Knife/Cutting Instrument Blunt Object Motor Vehicle Personal Weapons (hands, fists, feet, etc.) Poison Explosives Fire/Incendiary Device Asphyxiation Unknown Weapon No Weapon TOTAL Due to rounding, the percent of total may not add to 100.0 percent. The races of the victims of violent crime are presented in Table 5.12. Over 70 percent of the victims were white; black victims accounted for a little more than 27 percent of all victims. Age groups of victims of domestic □ violence are presented in Table 5.13. As we would expect, the 1 8 to 65 age group □ is the most prevalent, comprising 83.4 percent of the victims. The two juvenile groups follow. The elderly group is very small; the over-65 age group accounted for 1.1 percent of the violent crime victims in familial relationships. □ Table 5.14 displays the number of □ confrontations of the victims and their □ offenders by the age ranges reported to the UCR Program via the NIBRS from □ 1996-2001. A confrontation can be thought of as each unique combination of victim and offender within a single □ incident. For example in an incident □ with two victims and two offenders, there will be four confrontations (i.e., victim #1 /offender #1; victim #2/offender #1; victim #1 /offender #2; victim #2/offender #2). This also means that each victim or offender may be present in more than one category 1 The category of child refers to victims under the age of 18. Due to rounding, the percent of total may not add to 100.0 percent. *Less than 1 one-tenth of 1 percent. ultimately depending upon the total number of victims and offenders in the incident. Table 5.14 provides information on the relationship of victim to offender within each confrontation. When all possible relationships are considered, most of the confrontations 8.2 3.7 0.2 0.4 * 1.4 2.2 0.3 6.1 * m.- O.I * 1.7 5.6 0.0 involved victims and offenders in the 18 to 65 age category. After the incidents with familial relationships are isolated from the total, again, the highest number of confrontations occurred where both victims and offenders were in the 18 to 65 age group. For those incidents SPECIAL REPORT 345 Table 5.9 Number of Family Violence Incidents Involving Substance Abuse, 1996-2001 Substance 1996 7997 1998 7999 2000 2001 Total Alcohol 39,486 47,088 59,189 68,079 89,877 79,972 383,691 Drugs 27 52 100 163 188 173 703 Table 5.10 Number of Offenses Involving Substance Abuse by Family Relationship, 1996-2001 Substance Spouse Child 1 Elderly relative Alcohol Drugs 182,822 77 10,691 47 3,476 3 1 The category of child refers to victims under the age of 18. in which the confrontations could be considered spousal abuse, or the victims were "significant others" of the offenders, the majority (891,514) also occur with victims and offenders in the 18- to 65-year-old group. The number of confrontations within incidents involving a child and an individual in a parenting or care-giving role by the ages of the victims and offenders shows that most of these occurred where the offenders were in the 18-to 65-year-old category. Finally, confrontations involving elderly victims again confirm that majority of these incidents (27,574) occur with offenders 18- to 65- years old. Injuries The types of injuries suffered by victims of domestic, or family, violence during the study period are presented in Table 5.15. Major injuries are defined as those in which the victims suffered broken bones, possible internal injuries, loss of teeth, severe lacerations, or unconsciousness. Minor injuries and no injuries were nearly equal in number for every year except 1997 when the data showed a few more minor injuries than none. Overall, 46.0 percent of the injuries reported were minor, and 4.7 percent were major. No injuries were indicated in 49.3 percent of the reports. The numbers and types of injuries by certain types of abuse (i.e., spousal, child, and elderly) are presented in Table 5.16. In the spousal abuse category, the most prevalent type of injuries was minor. In child and elderly abuse situations, a majority of the cases involved no reported injuries (50.0 percent in child abuse cases and 50.4 percent in elderly abuse situations). Nearly 47 percent of child abuse cases and 45.1 percent of elderly abuse cases involved minor injuries. In all three categories, less than 5 percent of the cases involved major injuries. LIMITATIONS There are several limitations to this study. The UCR Program's Summary data, which comprise approximately 80 to 85 percent of the Program's database, could not be used to develop an in-depth study of this type. Those data are submitted as summary counts for the seven Part I crimes — murder, rape, robbery, aggravated assault, Table 5.11 Number of Victims of Violent Crime in Family Relationships by Gender, 1996-2001 Gender Number Percent of total Female 1,041,498 74.8 Male 348,267 25.0 Unknown 2,156 0.2 TOTAL 1,391,921 100.0 Due to rounding, the percent of total may not add to 100.0 percent. Table 5.12 Number of Victims of Violent Crime in Family Relationships by Race, 1996-2001 Race Asian/Pacific Islander Black American Indian/ Alaskan Native Unknown White TOTAL Number 6,676 379,884 5,320 22,707 977,334 1,391,921 Percent of total 0.5 27.3 0.4 1.6 70.2 100.0 Due to rounding, the percent of total may not add to 100.00 percent. Table 5.13 Number of Victims of Violent Crime in Family Relationships by Age, 1996-2001 Age 0-11 12-17 18-65 66 and up TOTAL Number 92,865 122,948 1160300 15800 1,391,921 Percent of total 6.7 8.8 83.4 1.1 100.0 346 CRIME IN THE UNITED STATES Table 5.14 Number of Confrontations Specific to Incidents involving Family Relationships by Age of Victim and Offender, 1996-2001 Offender age 0-11 Offender age 12-17 Offender age 18-65 All Offender age 66 and up Offender all ages Victim age 0-11 Victim age 12-17 Victim age 18-65 Victim age 66 and up Victim all ages 69,911 41,498 202,189 5,513 319,111 60,730 228,466 192,708 2,936 484,840 163,449 244,432 2,095,617 27,574 2,531,072 2,892 2,495 18,335 5,576 29,298 296,982 516,891 2,508,849 41,599 3,364,321 All Family Relationships Victim age 0-11 8,473 14,031 75,443 965 98,912 Victim agel2-17 2,915 30,951 98,374 807 133,047 Victim age 18-65 19,579 73,687 1,077,014 6,995 1,177,275 Victim age 66 and up 289 1,102 11,828 3,284 16,503 Victim all ages 31,256 119,771 1,262,659 12,051 1,425,737 Significant Other 1 Victim age 0-11 1,707 196 12,180 84 14,167 Victim age 12-17 576 8,829 23,497 54 32,956 Victim age 18-65 11,821 8,548 891,514 4,037 915,920 Victim age 66 and up 46 22 2,422 2,658 5,148 Victim all ages 14,150 17,595 929,613 6,833 968,191 Parent-Child 2 Victim age 0-11 1,877 2,278 52,831 734 57,720 Victim age 12-17 680 4,399 51,559 614 57,252 Victim age 18-65 564 1,464 29,631 1,683 33,342 Victim age 66 and up 12 5 86 28 131 Victim all ages 3,133 8,146 134,107 3,059 148,445 Elderly Relative Victim age 0-11 Victim age 12-17 Victim age 18-65 Victim age 66 and up 5,5 13 1 Includes the categories of spouse, common-law spouse, boy/girlfriend, and ex-spouse as victim. □ 2 Includes the categories of child, grandchild, stepchild, babysittee ("the baby"), and child of boy /girlfriend as victim. □ - - - 2,936 27,574 5,576 41,599 SPECIAL REPORT 347 Table 5.15 Number and Type of Injuries in Violent Offenses, 1996-2001 Percent of Percent of Percent of Percent of Percent of Percent of Number Percent of 1996 total 1997 total 1998 total 1999 total 2000 total 2001 total of injuries total None 102,488 48.5 129,192 47.1 176,386 48.7 215,217 49.6 285,087 49.8 276,795 50.4 1,185,165 49.3 Minor 94,422 44.7 130,012 47.4 168,686 46.5 200,054 46.1 262,678 45.9 250,099 45.5 1,105,951 46.0 Major 14,573 6.9 15,239 5.6 17,434 4.8 19,091 4.4 24,794 4.3 22,891 4.2 114,022 4.7 TOTAL 211,483 100.0 274,443 100.0 362,506 100.0 434,362 100.0 572,559 100.0 549,785 100.0 2,405,138 100.0 Due to rounding, the percent of total may not add to 100.0 percent. Table 5.16 Number of Injuries in Violent Offenses by Victim Category, 1996-2001 □ Spousal Percent of Percent of Elderly Percent of abuse total Child abuse total abuse total Major 24,769 3.2 2,546 3.0 648 4.5 Minor 425,267 54.4 39,714 46.9 6,436 45.1 None 332,142 42.5 42,344 50.0 7,195 50.4 TOTAL 782,178 100.0 84,604 100.0 14,279 100.0 burglary, larceny, and motor vehicle theft — and cannot be disaggregated to study an incident. Further, the use of the Hierarchy Rule in the Summary system limits the reporting of data to the offenses that fall inside the "hierarchy" structure of Part I crimes as defined by the UCR Program. In multiple offense situations, this procedure requires the reporting agency to count only the highest offense on the hierarchy list and ignore all others. For example, if a man beat, raped, and murdered his wife, the only offense that would be reported to the UCR Program (if the law enforcement agency was not reporting data via the NIBRS) is murder — the highest crime in the hierarchy. The other offenses would simply be lost data. NIBRS data are richer and more disaggregated than Summary data. However, NIBRS data are not as universally submitted as are Summary data. Over the decade of the nineties, more states became certified NIBRS participants and began submitting NIBRS data. The number of states submitting NIBRS data has grown from year to year. Even so, as of 2002, there were only 4,239 law enforcement agencies from 24 states using the NIBRS. This number represents 17 percent of the U.S. population and 18 percent of the crime statistics collected by the UCR Program. These data do not represent a scientific sample to reflect the national phenomenon. There are no cities participating in the NIBRS that have populations of 1 million or more inhabitants. There are only 11 cities or consolidated counties that contribute NIBRS data whose populations are 250,000 or more. A regional analysis would be valuable in this study. Regional variances could indicate cultural differences that could be studied to determine the causes and effects of domestic violence. For this study, however, regional analysis may hide more than it shows. NIBRS data for the period 1996-2001 are available for 20 states and the District of Columbia. Many of these states joined the program some time in the late nineties; therefore, the data for some of these states are not complete for that period. This study may better have been conducted by examining states for a particular year for which each had NIBRS data available. Even so, until more states contribute NIBRS data, regional analysis will be limited. For example, the West is defined by the UCR Program as Arizona, Colorado, Idaho, Montana, Nevada, New Mexico, Utah, Wyoming, Alaska, California, Hawaii, Oregon, and Washington. As of 2003, the only states from this region that were participating in the NIBRS program were Arizona, Colorado, Idaho, and Utah. The largest state in the region is California, which has the highest population and the highest number of crimes. In addition, it has many of the largest cities in the region. Its absence from regional statistics could present an inaccurate crime picture of the West. With these limitations, NIBRS data may not represent the crime experience in the entire United States. Due to these limitations, the results of this study must be interpreted with caution and with the noted caveats. SUMMARY AND CONCLUSIONS The objective of depicting violence among family members and intimate partners as reported in the data collected by the FBI's UCR Program has been met. Even though the findings in this report cannot be generalized to the entire country, it has demonstrated the utility of NIBRS data for analyses of this type. Moreover, other crimes or crime categories can be examined at a more 348 CRIME IN THE UNITED STATES in-depth level using NIBRS data. The simple methods used here demonstrate that characteristics of incidents, offenses, victims, and offenders can be examined across data segments. These findings are interesting and have significant implications for law- and policymakers. This study and other research concerning the demographic characteristics of the victims, offenders, and locations of domestic violence and information on prior criminal history and probationary status of offenders could be used to paint a fuller picture of the problem. This information could be valuable in enabling law enforcement policymakers, state legislatures, and Congress to develop better, more effective strategies for preventing spousal, child, and elderly abuse. ENDNOTES 1 Straus, M. A. and R. J. Gelles. (1988). "Violence □ in American Families: How Much is There and □ Why Does It Occur?" In E. W. Nunnally, C. S. □ Chilman, and F. M. Cox. Troubled Relationships. □ Families in Trouble Series, v. 3. Newbury Park, □ CA: Sage (1988), p. 141. □ 2 Straus, M. A. (2000). "Family Violence." □ In E. G. Borgatta and M. L. Borgatta (eds), □ Encyclopedia of Sociology, Second Edition, v. 2. □ NY: Macmillan, pp. 981-987. Straus and Gelles. □ (1988), p. 142.D 3 Kemper, C. Henry. (1962). "The Battered Child □ Syndrome," Journal of the American Medical □ Association. □ 4 Straus and Gelles. pp. 141-142.D 5 Abbott, J., R. Johnson, J. Kaziol-McLain. (1995). □ "Domestic Violence Against Women: Incidence and Prevalence in an Emergency Department Population." Journal of the American Medical Association. V. 272: 1763-1767. 6 Supplementary Homicide Report. (2002). U.S. Department of Justice, Federal Bureau of Investigation. 7 Crime in the United States. (1996). U.S. Department of Justice, Federal Bureau of Investigation. 8 National Incident-Based Reporting System. (2002). U.S. Department of Justice, Federal Bureau of Investigation. 9 American Psychological Association http://www.apa.org. 10 Tjaden, P., N. Thoennes. (2000). Full Report of the Prevalence, Incidence, and Consequences of Intimate Partner Violence Against Women: Findings from the National Violence Against Women Survey. Report for grant 93-IJ-CX-0012, funded by the National Institute of Justice and the Centers for Disease Control and Prevention. Washington, D.C.: NIJ. n Tjaden and Thoennes. (2000). 12 Rand, M. R. (1997). Violence-Related Injuries Treated in Hospital Emergency Departments. Bureau of Justice Statistics, Special Report. Washington, D.C.: U.S. Department of Justice. 13 Rand. (1997). 14 Stets, J. E., M. A. Straus. (1990). "Gender Differences in Reporting Marital Violence and its Consequences." Straus, M. A. and R. J. Gelles, editors. Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 Families. New Brunswick, NJ: Transaction Publishers, (1990): pp. 151-165. 15 National Research Council. (1996). Understanding Violence Against Women. Washington, D.C.: National Academy Press, 79-90. 16 Straus, M. A. and R. J. Gelles, eds., (1990). Physical Violence in American Families: Risk Factors and Adaptations to Violence in 8,145 Families. New Brunswick, NJ: Transaction Books. 17 Roizen, J. (1993). "Issues in the Epidemiology of Alcohol and Violence." In S. E. Martin, ed., Alcohol and Interpersonal Violence: Fostering Multidisciplinary Perspectives. Bethesda, MD: National Institute on Alcohol Abuse and Alcoholism. NIAAA Research Monograph No. 24, pp. 3-36. 18 National Crime Victimization Survey, 1992- 1996. U.S. Department of Justice, Bureau of Justice Statistics. 19 APA Online Press Release, http://www.apa.org/ releases/facts. html. 20 APA Online Press Release, http://www.apa.org/ releases/facts. html. 21 APA Online Press Release, http://www.apa.org/ releases/facts. html. 22 U.S. Advisory Board on Child Abuse and Neglect. (1995). A Nation's Shame: Fatal Child Abuse and Neglect in the United States. Washington, D.C.: U.S. Department of Health and Human Services. 23 Tatara, T., L. M. Kuzmeskus, and E. Duckhorn. (1997). "Trends in Elder Abuse in Domestic Settings," Elder Abuse Information Series No. 2. Report for grant 90-am-0660, funded by the National Center on Elder Abuse. Washington, D.C.: NCEA. 24 Tatara, Kuzmeskus, and Duckhorn. (1997). 25 Tatara, Kuzmeskus, and Duckhorn. (1997). 26 Tatara, Kuzmeskus, and Duckhorn. (1997). SPECIAL REPORT 349 HOMICIDE AS A COMMUNITY PROBLEM IN THE UNITED STATES The Importance of Homicide as a Community Problem in the United States Within the realm of criminal justice, probably no offense is as studied as that of homicide. In the United States, most certainly no other offense is treated more seriously in the criminal justice system. The most severe punishments are re- served for those individuals found guilty of homicide, and no statute of limita- tions exists for homicide as is found in most other types of offenses. Homicide, or fear of homicide, garners an immedi- ate reaction from communities. As such, law enforcement continually focuses efforts on the issues surrounding violent crime such as illegal gun or drug crime crackdowns. Ultimately, they hope to lessen the likelihood of homicide vic- timization and, as a result, fear within a community. Patterns of Homicide To understand the dynamics of any criminal offense, one would focus upon general patterns, and homicide is no exception. These patterns can be expressed in terms of temporal pat- terns (how homicide has changed over time), spatial patterns (how homicide changes over regions or locations), pat- terns in the inherent characteristics in terms of victims, offenders or other qualities of the homicide incident, or any combination of these qualities. To provide a thorough review of the stud- ies conducted about homicide is beyond the scope and purpose of this paper. Suffice it to say, there have been numer- ous studies focusing on any one of the major themes listed above. However, in terms of recognizing patterns in the homicide data, these studies traditionally limit themselves to analyzing only a few dimensions at one time. For example, age- specific rates for victims or offend- ers may be analyzed over time or space. Even though theoretically possible, diffi- culties in interpretation arise when tradi- tional pattern analysis techniques, such as cross-tabulations or scatter plots, are used to analyze more than three dimen- sions at one time. To try to view large complex data sets such as can be found with homicide in order to recognize pat- terns requires a more sophisticated ap- proach. Newer technologies have made these sophisticated pattern analysis tech- niques more accessible through both the availability of computing resources and the ease of use. Pattern Recognition and Data Mining The underlying goal of pattern recogni- tion is, ultimately, data reduction. In- stinctively, humans reduce the amount of information in the world by organizing its constituents into a series of concep- tual types. This is accomplished through highlighting important characteristics that define the differences and disregard- ing the details that add little value. The details of criminal incidents are invalu- able to law enforcement to achieve the goals of solving and reducing crime. However, making sense of those details is a challenge when the scope is broader than a single incident. In order to truly use the data, the analyst must first orga- nize the data. This can be done in one of two ways: either conceptual classifica- tion or numerical classification. Conceptual classification methods are often used to identify ideal or polar (extreme) types and are more likely to be drawn from a collective set of ex- periences that represent the concepts rather than an actual set of cases. Nu- merical classification uses quantitative techniques to identify like cases that translate into classes. Numerical clas- sification has been more readily used in biology and other sciences rather than in the social realm. However, that practice is beginning to change as the technology involved has become more accessible to a wider audience (Bailey 1994). The amount of information available to law enforcement via their incident reports and external sources, such as medical examiners offices, drug laboratories, and other sources of intel- ligence, have placed new demands on already overstretched resources which have limited the time and attention that each law enforcement employee can spend on investigation and analysis. Law enforcement would benefit from the application of newer computer technolo- gies in both hardware and software to help cut a clearer path through their data in order to help define problems in their communities. For these and many other reasons, the pattern recognition capabili- ties in data mining have become increas- ingly popular with law enforcement. Crime analysts also can use pat- tern recognition techniques common to data mining to examine large criminal justice data sets. In the case of homi- cide, the Uniform Crime Reporting (UCR) Program through its Supplemen- tary Homicide Report (SHR) collects information on the incident for the vast majority of reported homicides from state and local law enforcement. By us- ing pattern recognition techniques, such as cluster analysis, with homicide inci- dent information, the patterns that occur naturally within the data set can be used to provide a deeper understanding of the SPECIAL REPORT 351 dynamics of homicide in the United States. This can aid more in- depth studies that focus on the underly- ing causes of or concurrent factors that contribute to homicide. Objectives of the Study By using incident-specific information available through the SHR, pattern rec- ognition or data mining techniques can be applied to discern any patterns that exist in the homicide data. One meth- odology employed is cluster analysis. Cluster analysis uses a series of math- ematical computations to identify groups that occur in data sets. These groups, or clusters, could be used to delve beneath the surface of the homicide rate so often quoted to reveal the nature of those rates and to explore them in a regional and temporal context. The objectives could be summarized by the following ques- tions: • □ What do the characteristics of homi- cide incidents reveal about the nature of homicide in the United States? • DHow might the homicide data be de- composed to research it over time and space? • DHow have the characteristics of homi- cide changed regionally since 1980? Methodology Data The UCR Program defines murder and nonnegligent homicide as "the willful (nonnegligent) killing of one human being by another" (USDOJ 1984, p. 6). This definition does not include suicides, accidental deaths, assaults to murder, traffic fatalities, or attempted murders. Although justifiable homicides by law enforcement officers in the line of duty or private citizens during the commis- sion of a felony are considered willful killings, they have not been used in the calculation of published murder rates. Since 1962, 1 the FBI has collected infor- mation on homicide incidents that can be employed by a variety of users to ex- plore the nature of homicides. This in- formation, the Supplementary Homicide Report, is collected in addition to the official reports of crime used to calculate the national murder rate. The attributes collected on the form include the age, sex, and race of both victim and of- fender, weapons, relationship of victim to offender, circumstances of the homi- cide, as well as information on multiple victims and offenders. Additionally, law enforcement submits information on those incidents classified as justifiable homicide by UCR definitions via the SHR. It is included in this analysis. The SHR data from 1980 through 2002 were recoded to express each incident in terms of 36 characteristics. These characteristics include the pres- ence or absence of such things as juve- nile offenders or victims, male victims or offenders, a firearm, or familial rela- tionships between victims and offenders to name a few. More detail can be found on these characteristics in the Appendix of this study. These variables become the basis for the cluster analysis. For the remainder of the analysis, the homicide data are assigned to a county based upon the location and jurisdiction of the re- porting agency. These county-level files form the basis of the geographic analy- sis. Because the incident-level homicide data can have fluctuations based upon the reporting history of the agencies involved as well as rare or extreme events, the data used for this study are the 3 -year centered moving averages for each county. For example, the 3 -year centered moving average rep- resenting levels of homicide for a county for 1982 is the average of the reports for 1981, 1982, and 1983. By using this data smoothing technique, however, the first and last years in the series are lost. The final time series used in this study represent the years 1981 to 2001. Cluster analysis of homicides incidents to determine types The methodology of this study focuses on two primary areas: the use of cluster analysis to detect patterns inherent in the data set itself and the mapping of those homicides to a location with the help of geographic information system (GIS) technology. As stated previously, a cluster analysis was performed on the recoded SHR data in order to discern any pat- terns inherent in the data. Cluster analysis allows for the data to drive the determination of types or groups rather than preconceived ideas of how homicides occur in the nation. It uses measurements of similarity based upon the characteristics of each homicide incident to allow "clustering" of types to be identified. Once the valid and reli- able "types" of homicide are determined for each year, each reporting agency's types of homicide will be analyzed within each region to track their spatial movement and to see if they are related to one another or could be considered to be related to population movements. For a more thorough discussion of the technique, including the checks on the validity and reliability of the results, a Technical Note is available upon request from the FBI's Crime Analysis, Re- search and Development Unit, telephone (304) 625-3600. GIS analysis of homicide incidents In addition to cluster analysis, the homicide information was tracked through space with the aid of GIS tech- nology. Initially, the location of the incident-level homicides is determined 352 CRIME IN THE UNITED STATES Figure 5.1 Mean Center of Homicides Reported as Incident Data 1981-2001 (3-year centered moving average) o o Springfield*^ 00 00° o Total Homicides Reporting Population Murder Rate per 100,000 O 1981 -1984 O 1981 -1984 O 5.5-6.3 1985-1988 1989-1992 1993-1996 1997 -2001 Std Dev (+1) -2001 1985-1988 1989-1992 1993-1996 1997 -2001 Std Dev (+1) -2001 O 6.4 -7.4 Q 7.5 -8.3 8.4-9.1 9.2 -9.8 Places (100,000+ population) o Q> SPECIAL REPORT 353 based upon the county in which the re- porting agency resides. These incidents are spatially attributed to the point of the county centroid, which is the spatial center of the county. The mean center weighted by the number of reported homicides for each year, as well as by cluster type, is calculated for the Nation and each of the four regions. The mean center is the geographic equivalent to the average in a data set, and can be thought of as the "balancing point." Addition- ally, the standard deviation for the mean center is calculated for each year to indicate the dispersion of the data. This standard deviation is calculated for both the x- and y-dimension in space and is represented by an ellipse. For a more thorough discussion of the calculation of the weighted mean center and standard deviational ellipse, a Technical Note is available upon request from the FBI's Crime Analysis, Research and Develop- ment Unit, telephone (304) 625-3600. Incidents of Homicide in the United States Spatial trend of homicide incidents from 1981 to 2001 for the Nation An examination of the calculated mean centers for the reports of incident-level homicide revealed that they appeared to be balanced near the geographic center of the United States. However, the stan- dard deviational ellipse for the 2001 cen- tered value indicated that the majority of the homicide incidents drifted to the east coast, and the reporting population was more geographically dispersed than the reports of homicide. This is primar- ily reflected in the northwest extant of the standard deviational ellipse. In other words, homicides were more concentrat- ed than the reporting populations of the same counties for the same time period (2001). Additionally, there appears to be an urban bias to the homicide incidents shown in this map. Approximately 66 percent of the Nation's homicides took place in an area that encompassed nearly 5 1 percent of the urban places with a population of 100,000 or more. (See Figure 5.1.) The study period (1981-2001) be- gan with a murder rate of approximately 9.8 murders per 100,000 inhabitants. However, the murder rate declined in subsequent years. The rate again rose to the same high in 1991 and then declined to the lowest point in 21 years by 2000 to 5.5 murders per 100,000 inhabitants (USDOJ 2003). Over time, the popula- tion covered by agencies reporting SHR data has shown a nearly true westerly progression from east to west. Interest- ingly, the Census Bureau reported the same momentum for the U.S. popula- tion. During the same time period, the incidence of homicide appears to have a western bias in its trend, and there is less movement for reported homicides. When the homicide rate is taken into consideration, there is a more easterly bias in higher crime rate years than the lower crime rate years in the latter part of the study period. This may be a reflection of a decline in the homicide rates in the East. (See Figure 5.1, Inset.) Spatial trend of homicide incidents from 1981 to 2001 for each region The UCR Program divides the United States into four geographic regions for data analyses: the Northeast, the Mid- west, the South and the West. When the homicide data are analyzed by region, there appears to be strong evidence that the level of urbanity is tied to the inci- dence of homicide on a regional level, as well. Visually, the data show that urban areas are in a more geographically dispersed pattern in the Midwest and South. Although the West accounts for approximately 40 percent of the Nation's populated places with 100,000 or more in population, they are almost all con- centrated around San Francisco and Los Angeles. In the Northeast, these larger urban centers are almost all concentrated around the New York City and Boston areas. Additionally, the Northeast pro- portionally contributes only about 10 percent to the total number of the larger urban centers for the Nation. (See Figure 5.2.) The regional dispersion of homi- cides in relation to reporting population nearly mimics the results of the national analysis. Except for the Northeast, ho- micides tend to be more concentrated than reporting population over time. With the exception of the South, most mean centers are also geographically close to major urban centers of the region (for example, New York City, Chicago/Detroit, and Los Angeles/San Francisco). The South shows a much wider dispersion of homicide, which may reflect the geographic pull of more widely dispersed urban centers (for ex- ample, New Orleans, Washington, D.C., Houston, and Atlanta). (See Figure 5.2.) This urban pull is also reflected in the geographic progression of the mean centers through time. Although the reporting population of the Midwest appears to split the difference between Chicago and Detroit, there appears to be a bias towards Chicago in terms of the homicide reports. The reporting popu- lation in the Northeast appears to be geographically stagnant near New York City. However, the incidence of ho- micide drifts westerly during the study period. The reporting population mean center of the West has been moving in a southerly direction towards Los Angeles, but homicides show a southeasternly pull between Los Angeles and Las Vegas. Again, the South does not show a bias towards any particular urban area with the reporting population mov- 354 CRIME IN THE UNITED STATES hicago O aO A o o 9 o° o6> Midwest Stamford Yonkers rson J 4 v York Elizabeth Northeast Total Homicides Reporting Population O 1981 -1984 O 1981 -1984 O 1985 -1988 O 1985 -1988 O 1989 -1992 • 1989 -1992 • 1993 -1996 • 1993 -1996 • 1997 -2001 • 1997 -2001 CZ^Std Dev (+1)-2001 C Dstd Dev (+1) -2001 Places (100,000+ population) South B?rmil£$iam o ^qp Montgomery Figure 5.2 Mean Center of Homicides Reported as Incident Data by Region 1981-2001 (3-year centered moving average) SPECIAL REPORT 355 Figure 5.3 50 Murder Rate and Percent of Category of Unknowns, 1981-2001 14 ing towards the northwest and homicide incidents moving towards the northeast. In general, one does not see evidence that the incidence of homicide appears to be linearly related to the reporting population during the study period. Characteristics of Detected Homicide Clusters The results of the cluster analysis pro- duced two highly robust types of homi- cide that can be tracked for the entire study period. The first robust cluster is comprised primarily of incidents that involve an unknown offender, and as such, much of the information about the offender and the circumstances sur- rounding the incident are reported as unknown. The remaining incidents are either clustered together as one all other type, or in some years, the remaining cases are divided based primarily on the race of the offender. In those years, there are two remaining clusters: one with black offenders and one with white and all other race 2 offenders. Given the consistency (reliability) of the results from each year, the clusters produced by this analysis will form the basis of the remaining analyses. The category unknowns accounts for approximately 30 percent of the 417,505 homicides reported through the SHR during the study period. These are the homicides in which little to noth- ing is known about the offender at the time of the incident report. There is a slight urban bias to these unknowns, and the victims are more likely to be black, male, and adult than the rest of the vic- tims of the homicides. However, when the unknowns are compared to all other homicides, both groups appear to be very similar in terms of weapons used. (See Table 5.17.) Over the length of the study period, there appears to be a slight increase in the proportion of unknowns. (See Figure 5.3.) Regionally, the data reflected many of the same patterns concerning age and weapons associated with the two identified types of homicide. However, some striking differences are obscured by the national figures. In the West, one sees a significant increase in the propor- tion of white victims (65 and 70 per- cent, respectively for unknowns and all other homicides) when compared to the national figures (46 and 5 1 percent, re- spectively). Additionally, both the South and the West show a higher incidence of each type of homicide in suburban areas than do the remaining regions. In gen- eral, although many of the differences between the two types are subtle, the South also showed much more similarity 356 CRIME IN THE UNITED STATES Figure 5.4 Murder Rate and Percent of Category of Unknowns by Region, 1981-2001 14 Northeast - 12 - 10 - 8 - 6 - 4 c CD CD t3 CD o o o o o I— I tr p cr i— ■ • P Percent Unknowns Murder Rate - 2 J I I I I I I I I I I I I I I I I L oo oo oo -J oo to o o between the unknowns and all other ho- micides than did the other regions. (See Table 5.17.) Since 1981, the proportion of unknowns on a regional level does not consistently track with fluctuations in the murder rate. In the Midwest, the South, and the West, the percent of un- knowns either grew or remained stable during times of declining murder rate. It was only in the Northeast that the trends of the murder rate and percentage of unknowns appear to move in congruent directions (See Figure 5.4.) Spatial Occurrence of Homicide Clusters Spatial trend of homicide clusters from 1981 to 2001 for the Nation Unknowns show a lot more geographic progression in a westerly direction dur- ing the study period than do all other homicides. In the early part of the study period, the mean center for unknowns appeared more to the east than the mean center for the remaining homicides. That bias is almost nonexistent by the end of the study period. For the entire study period, the unknown offender homicides also show a more northern bias than other homicides. The standard deviational ellipse for the centered aver- age value for 2001 (the most current data year in the study) shows a wider dispersion for the unknowns on the east- to-west axis. However, the remaining homicides are more dispersed on the north-to-south axis. This indicates a more concentrated band of unknowns that incorporates the effects of urban areas in the northern part of the North- east (Boston/Connecticut) as well as in southern California. The all others seem to incorporate the effect of more south- erly urban areas such as Atlanta, New Orleans, and Houston. (See Figure 5.5.) Spatial trend of homicide clusters from 1981 to 2001 for each region Within the Nation's four regions, the data reflected a similar pattern where each type of homicide has distinct mean centers. These differences for the two types reflect the influence of diverse communities even within the same re- gion. All regions but the West showed the same spatial progression through time for the two types of homicide. In all cases, this progression runs counter to the progression for the reporting pop- ulation. Interestingly, unknowns show an almost true southerly progression in the West while all other homicides drift to the northeast. The standard deviation- al ellipse for the most recent data year (the centered average value for 2001) showed that the dispersion of unknowns was less than or equal to the remaining homicides in all regions but the South. (See Figure 5.6.) Discussion and Conclusion The differing rates of homicide amongst the regions, particularly in the South and West, have been noted in the past by law enforcement and researchers alike. However, instead of these differences in the levels of homicide being a result of global processes within a region, there appears to be evidence of more subtle processes that are connected to local urban centers. The regional incidence of homicide may be a reflection of the level of urbanity or change in urbanity rather than strictly the numbers of people that reside there. Since the patterns of Table 5.17 Characteristics of detected homicide clusters Percent of total Nation Northeast Midwest South West Unknowns All Other Unknowns All Other Unknowns All Other Unknowns All Other Unknowns All Other MSA status 94.2 84.8 98.3 94.7 96.1 90.4 89.7 76.3 95.5 90.3 Suburban 18.1 21.7 8.3 17.1 11.1 15.0 22.8 24.8 25.8 24.0 Black Victim 50.6 45.9 53.1 49.2 67.8 57.8 56.2 51.5 27.7 23.4 White Victim 45.9 51.4 43.3 48.2 31.1 40.7 41.6 47.1 65.3 70.3 Female Victim 19.8 24.3 17.1 24.8 21.2 25.3 21.4 24.0 18.8 23.8 Male Victim 81.3 77.4 83.7 76.9 80.5 76.5 79.4 77.5 82.4 78.0 Juvenile Victim 7.0 10.7 6.7 11.8 7.4 12.6 5.8 8.7 8.6 12.0 Firearm Used 65.4 64.5 66.6 55.7 67.1 63.2 64.2 68.4 64.7 63.7 Other Serious Weapon Used 34.6 35.6 33.4 44.3 32.9 36.9 35.8 31.8 35.3 36.3 358 CRIME IN THE UNITED STATES Figure 5.5 Mean Center of Homicide by Type 1981-2001 (3-year centered moving average) o 1980-1983 O 1980-1983 o 1984-1988 O 1984-1988 o 1989-1992 • 1989-1992 • 1993-1997 • 1993-1997 • 1998 -2002 • 1998 -2002 Std Dev (+1 ) -2001 Std Dev (+1 ) -2001 Places (100,000+ population) SPECIAL REPORT 359 Chicago Gary • CD ° 0. • ^ o -o 00 CT OCD o O O O Midwest o ftp Fresno \ West Las Vegas Bakersfield \ r Paterson ..^086^ ° t>% V If - Newark ¥ _ ; Northeast Ne ^ York Unknowns All Other Homicides 1981 -1984 1981 -1984 1985-1988 1985-1988 1989-1992 • 1989-1992 1993-1996 • 1993-1996 • 1997 -2001 • 1997 -2001 ) Std Dev(+1) -2001 ) Std Dev(+1) -2001 Places (100,000+ population) "cr Huntsville South O o ^ o o • 9d Atlanta O Birmingham Figure 5.6 Regional Mean Center of Homicide by Type 1981-2001 (3-year centered moving average) 360 CRIME IN THE UNITED STATES growth and decline vary by region, this could be one explanation for the varying results in movement and types of homi- cide among regions. The unknowns are difficult to draw too many conclusions about, since, by definition, little is known of the cir- cumstances surrounding the homicide. However, it could easily be seen how these homicides may differ qualitatively from the remaining homicides that are often between people who know one another and in many cases are the result of arguments. The regional differences in the trends and proportions of these unknown homicides appear to be influ- enced by communities different from the remaining homicides. Again, this points to the dominance of communities differ- ent from those that drive the trends for the remaining homicides. The results of this analysis show that there is a definite need for further exploration of what is driving regional trends in homicide. The evidence seems to point to a complex interaction between regional differences in the underlying factors affecting homi- cide and the regional differences in the types of homicide itself. References Aldenderfer, Mark S. and Roger K. Blashfield. (1984). Cluster Analysis. Newbury Park, CA: Sage Publications. Bailey, Kenneth D. (1994). Typologies and Taxonomies: An Introduction to Classification Techniques. Newbury Park, CA: Sage Publications. Goldsmith, Victor, Philip G. McGuire, John H. Mollenkopf, Timothy A. Ross, eds. (2000). Analyzing Crime Patterns: Frontiers of Practice. Thousand Oaks, CA: Sage Publications, Inc. Lang worthy, Robert H. and Eric S. Jef- feris. (2000). The Utility of Standard Deviation Ellipses for Evaluating Hot Spots. In Goldsmith, Victor, Philip G. McGuire, John H. Mollenkopf, Timothy A. Ross, (Eds.), Analyzing Crime Pat- terns: Frontiers of Practice. Thousand Oaks, CA: Sage Publications, Inc. Levine, Ned. (May 2002). CrimeStatll: A Spatial Statistics Program for the analysis of Crime Incident Locations. Houston, TX: Ned Levine & Associa- tions and Washington, D.C.: the National Institute of Justice. McCue, Colleen, Emily S. Stone, and Teresa P. Gooch. (2003). "Data Mining and Value-Added Analysis," Law En- forcement Bulletin. November, p. 1-5. Romesburg, H. Charles. (1984). Cluster Analysis for Researchers. Belmont, CA: Lifetime Learning Publications. SPSS. "The SPSS TwoStep Cluster Component," Technical Report down- loaded from www.spss.com on April 1, 2004. U.S. Department of Justice. Federal Bureau of Investigation (2003). Crime in the United States, 2002. Washington, D.C. U.S. Department of Justice. Federal Bureau of Investigation (1984). Uniform Crime Reporting Handbook. Washing- ton, D.C. Endnotes 1 Supplementary homicide information has been collected since the beginning of the UCR Program in the early 1930s. However, this information was not made available for general dissemination until 1962 and has gone through various revi- sions since that time. The data used for this analysis reflect the latest version of information collected which has re- mained the same since 1980. 2 These race categories include Asian and Other Pacific Islander and Ameri- can Indian and Alaskan Native. SPECIAL REPORT 361 Appendix Variable Description MSA MSA location Suburban Suburban location Single Victim Incident involved only a single victim Single Ottender Incident involved only a single ottender Unknown Ottender Incident involved an unknown ottender Juvenile Victim Victim under the age ot 1 8 years old Male Victim Male victim involved Female Victim Female victim involved Unknown Victim Unknown victim involved White Victim White victim involved Black Victim Black victim involved AIAN Victim American Indian or Alaskan Native victim involved AOPI Victim A * * i~* T 11 * a 9 9 11 Asian or Pacific Islander victim involved Juvenile Offender y**v c*c* i ii c* i i i * i i Ottender under the age ot 1 8 years old involved Male Offender Male offender involved Female Offender Female ottender involved Unknown Ottender T "T" 1 C*C* 1 * 11 Unknown ottender involved White Offender T "T T1 * C*C* 1 * 11 White ottender involved Black Offender Black offender involved AIAN Offender American Indian or Alaskan Native offender involved AOPI Offender Asian or Pacific Islander offender involved Firearm used Includes handgun, rifle, shotgun, other gun, and general firearm vjLiier oenous weapon Intimate Relationship Other Family Relationship Otherwise Known iiiciuues Kiine/ cutting msLruiiieiiL, diuiil opject, personal weapons, poison, pusiieu/tiirown out oi winuow, explosives, fire, narcotics/drugs, drowning, strangulation, asphyxiation, and other Includes husband, wife, common-law husband, common-law wife, boyfriend, girlfriend, ex-husband, ex-wife, and nomosexuai reiationsnip Includes mother, father, son, daughter, brother, sister, in-law, stepfather, stepmother, stepson, stepdaughter, and other family Til "11 * 1 1 f* " 1 1 A 1 "1 * J * Includes neighbor, acquaintance, employee, employer, mend, and otherwise known to victim Not Known to Victim Includes stranger. Unknown Relationship All instances where relationship of victim to offender cannot be determined. Felony Type Circumstance - Violent Includes rape and robbery Felony Type Circumstance - Drug Includes narcotic drug laws Felony Type Circumstance - Other Other Circumstance - Arguments Other Circumstance - Organized Includes burglary, larceny, motor vehicle theft, arson, prostitution and commercialized vice, other sex offenses, abor- uon, gamonng, ana omer— noi specmea Includes lover's triangle, brawl due to influence of alcohol, brawl due to influence of narcotics, argument over mon- ey or property, and other arguments. Includes gangland killings, juvenile gang killings, and institutional killings. Other Circumstance - Other Includes child killed by babysitter, sniper attack, and other. Suspected Felony Circumstances indicate possible felony type murder, but sufficient facts to identify type of felony not available. Justifiable Homicide The intentional killing of a person without evil design and under such circumstance of necessity or duty as to render the act proper. Includes felons killed by either private citizen or police. The preceding list reflects particular characteristics captured in a homicide incident reporting through the SHR. These characteristics were recoded to reflect whether or not that characteristic was present or not present. If the characteristic was present on the incident, that variable was coded as the value 1. Otherwise, the variable was set to the value of 0. The cluster analysis algorithm described in detail in the Technical Note used these values in its calculations. The Technical Note is available upon request from the Crime Analysis, Research and Development Unit, telephone (304) 625-3600. 362 CRIME IN THE UNITED STATES