Center Papers

Youth Violence: An Overview
Dr. Delbert Elliott, Ph.D.
Director, Center for the Study and Prevention of Violence
March 1994

Paper presented at The Aspen Institute's Children's Policy Forum
"Children and Violence Conference" February 18-21, 1994
Queenstown, MD

Across America, people are afraid. This fear is not restricted to those living in the most disadvantaged neighborhoods in our large cities, but extends to residents of affluent suburban communities, and even small towns and rural areas. For many, the violence epidemic of the 1990s signals a general breakdown in the social order. There seem to be no safe places; the violence extends into our homes, neighborhoods, schools, day care facilities, shopping malls, and workplaces. The perpetrators are often relatives, friends or acquaintances of the victim. So much of the violence seems petty, senseless or random, suggesting a wanton disregard for human life. And both perpetrators and victims are increasingly our adolescents and children.

This paper presents an overview of theory and research on youth violence as a background for specific presentations and discussion. Four key questions are addressed: 1) Are todays youth really more violent? 2) How many youth are victims of violence or committing violent acts themselves? 3) What are the major causes of youth violence? and 4) What is known about the prevention and control of youth violence?

ARE TODAYS YOUTH MORE VIOLENT* AND VICTIMIZED BY VIOLENCE?

Is the current violence epidemic a distortion resulting from excesses in media coverage or does it reflect a real change in the behavior of youth? The evidence suggests the following conclusions about trends in youth violence over the past decade: 1) There is a substantial increase in the violence victimization rates for adolescents, particularly for 12-15 year olds; 2) There has been a relatively small increase (8-10 percent) in the proportion of adolescents involved in some type of serious violent offending; and 3) There has been a dramatic increase in adolescent homicide rates, beginning in 1988.

In essence, todays youth are more frequently the victims of violence; but about the same proportion of youth are committing serious violent offenses today as in 1980 and their frequency of offending is approximately the same. One important dimension of youth violence has clearly changed. Todays violent acts are more lethal, a larger proportion of these acts result in serious injury or death. The fact that the adolescent homicide rate has more than doubled since 1988 (while the overall rate has remained relatively constant) is grim evidence of this increased lethality. And this dramatic increase in the lethality of adolescent violence is explained almost entirely by the increased use of handguns in these violent exchanges.

These findings do not suggest a massive failure in our social institutions or a dramatic decline in the commitment of most young persons to responsible, lawful behavior. While levels of youth participation in violent behavior are unacceptably high (see below) and constitute a serious crime and public healthproblem that must be addressed, what is different about youth violence in the 90s is its lethality. This should focus our concern on the dramatic increase in the use of handguns in juvenile assaults at a time when handgun crimes in general are declining, and on what has happened in the lives of these violent young people that they have so little respect for human life.

HOW MANY YOUTH ARE SERIOUS VIOLENT OFFENDERS OR VICTIMS?

National self-report studies indicate that the age of highest risk for the initiation of serious violent behavior is age 15-16 and that the risk of initiating violence after age 20 is very low. If persons have not initiated serious violent behavior by age 20, it is unlikely that they will ever become serious violent offenders. The highest rates of participation in serious violence are at ages 16-17. At these ages, 20-25 percent of males and 4-10 percent of females report one or more serious violent acts. After age 17 however, participation rates drop dramatically. Approximately 80 percent of those who were violent during their adolescent years will terminate their violence by age 21.

Estimates of physical abuse of children and youth suggest that as many as 10 percent are assaulted by family members and care givers each year. Among children and youth, preschoolers were most likely to be victims of physical assault, followed by elementary school-aged children, teenagers, and infants. Children under age 4 are more likely to be killed than older children. Infants are more often killed by their mothers than fathers. Between 2-4 percent of youth are victims of a violent personal assault each year while at school. Overall, teenagers are twice as likely to be victims of an assault as persons aged twenty and older. Suicide is now the third leading cause of death for youth. Clearly our children and teenagers are the most frequent victims of violence. Young violent offenders and victims have similar personal characteristics. This is because many youthful victims of violence are also involved in violent behavior themselves, or are at high risk of becoming violent during their adolescent years. While both offenders and victims are disproportionately male, black, urban, and from low income and single-parent families, this characterization of violent youth is misleading. Among children, the gender difference in victimization is small, whereas among adolescent victims and violent offenders, it is quite strong. Among violent offenders, race/ethnic and social class differences are small during adolescence; they become substantially greater during the adult years. For example, by age 18, the cumulative proportion of blacks involved in serious violent offending is only 18 percent greater than that of whites. There is little evidence from the national self-report studies for any difference in predisposition to violence by race, once social class is taken into account. Youth violenceis thus very widespread in our society. It is not just a problem for the poor, or minorities, or those in our large cities. It crosses all class, race, gender and residence boundaries. It is a problem for all Americans.

WHAT ARE THE MAJOR CAUSES OF YOUTH VIOLENCE?

Most violent behavior is learned behavior. We all have some potential for violent behavior; we have observed others using violence and know how to do it. But while it may be a part of nearly everyone's behavioral repertoire, most persons have non-violent ways of achieving their purposes which are effective in most situations. Further, their commitment to conventional norms and values inhibits their use of violent behavior and they are embedded in social networks (family and friends) and situations where this type of behavior would have serious negative ramifications. Under these circumstances, violent behavior becomes irrational.

Unfortunately, for too many youth, violence is either the only or the most effective way to achieve status, respect, and other basic social and personal needs. There is little prosocial modeling of alternative ways of dealing with conflict. Like money and knowledge, violence is a form of power, and for some youth, it is the only form of power available. When such limited alternatives are combined with a weak commitment to moral norms (internal controls) and little monitoring or supervision of behavior (external controls), violent behavior becomes rational. The potential rewards are great, the perceived costs minimal.

The Family Context

The initial causes of violence are found in the early learning experiences in the family. They involve 1) weak family bonding, ineffective monitoring and supervision; 2) exposure to and reinforcement for violence in the home, and 3) the acquisition of expectations, attitudes, beliefs and emotional responses which support or tolerate the use of violence.

Early exposure to violence in the family may involve either witnessing violence or physical abuse. Research suggests that these forms of exposure to violence during childhood increase the risk of violent behavior during adolescence by as much as 40 percent. Still, most youth who are victims of physical abuse do not go on to become serious violent offenders. While exposure to real violence and physical abuse on the part of family members have stronger modeling effects, heavy exposure to violence on television is also causally linked to later violence. In many homes, television is the de facto baby sitter, with little or no monitoring or supervision of content. When there is strong family bonding, effective teaching of moral values and norms, and effective monitoring of behavior, the effect of exposure to violence on TV is probably negligible; without this protection, its effect can be quite strong. What is learned is not only how to do violence, but a desensitization to violence and rationalizations for disengaging one's moral obligations to others.

Even if violence is not modeled in the home, research suggests that the absence of effective social bonds and controls, together with a failure of parents to teach (and children to internalize) conventional norms and values, puts children at risk of later violence. In fact, parental neglect may have an even stronger effect than physical abuse on later violence, as it appears to be more damaging to the subsequent course of youth development and involves three times as many youth.

There is also evidence that certain individual temperaments and acquired biological deficits may complicate or interfere with parents' efforts to develop good internal controls in their children. Antisocialpersonality and attention deficit disorders, a fearless and impulsive temperament, exposure to lead and other neurotoxins, and serious head injuries, for example, may make it difficult for even the best parents to develop strong family bonding, good internal controls and provide effective monitoring of their children.

Families with a high risk for child abuse are those with parents or caretakers who have limited problem solving skills, poor impulse control and a history of violent behavior during adolescence. These caretakers are frequently young, low income, single parent, minority women with four or more children in the household. Fathers, when present, tend to be part-time employed and have a limited education. These families have few resources and are experiencing both social isolation and economic stress. They have few alternatives and limited social supports from extended family or friendship networks which might provide social controls on their behavior and non-violent alternatives for managing their children.

The Neighborhood Context

Some neighborhoods also provide opportunities for learning and engaging in violence. The presence of gangs and illegal markets, particularly drug distribution networks, not only provide high levels of exposure to violence, but violent role models, and positive rewards for serious violent activity. Single parent families, ineffective parenting, violent schools, high dropout rates, high adolescent pregnancy rates, substance abuse and high unemployment rates are all concentrated in such neighborhoods.

While these neighborhoods are areas with high rates of concentrated poverty, the critical feature of such neighborhoods that is most directly related to the high rates of violence, crime and substance use, is the absence of any effective social or cultural organization in these neighborhoods. High levels of transiency make it difficult to establish common values and norms, informal support networks and effective social controls. High chronic unemployment results in social isolation from legitimate labor markets, and undermines the relevance of completing school. Illegitimate enterprises and gangs emerge in these neighborhoods, in part because the neighborhood has no effective means of resisting such activity, and in part as a means of providing some stable social organization for youth and some economy for the neighborhood. Not all poor neighborhoods are disorganized however; and those that are effectively organized have low rates of violent behavior, crime and substance use. Poverty is linked to violence through disorganized neighborhoods.

The effect of living in such neighborhoods can be devastating on the family's attempt to provide a healthy, conventional upbringing for their children. Not only are there few social reinforcements for conventional lifestyles to support this type of parenting, but conventional opportunities are limited by racism, discrimination, social isolation from the labor market and few resources. There are often greater opportunities for participation in gangs and the illicit economy which offer relatively quick and substantial rewards that seem to offset the risks associated with violence. One effect of participation in these types of activities is that youth are at high risk for becoming victims as well as perpetrators of violence; a second is that such youth frequently abandon the pursuit of more conventional goals, drop out of school, get pregnant, and become enmeshed in health compromising and dysfunctional lifestyles which arrest the normal course of adolescent development. Such youth are ill-prepared to enter conventional adult roles.

The School and Peer Context

While patterns of behavior learned in early childhood (e.g., aggressiveness) carry over into the school context, the school has its own potential for generating conflict and frustration and violent responses to these situations. A successful non-violent social adjustment at home increases the likelihood but does not guarantee a successful non-violent adjustment to school and peers. These are new social systems which have to be negotiated, where one must find her or his own niche. They each have their own performance demands and developmental tasks to complete. Failure to meet these school and peer performance expectations (e.g., academic success, peer approval, personal competence and independence, self-efficacy, and a capacity for developing and maintaining interpersonal relationships and intimacy) creates stress and conflict. The combination of new conflicts and reduced levels of monitoring and supervision in these contexts, increases the likelihood that violence will emerge in response to these problems.

During junior and senior high school, a clear adolescent status hierarchy emerges, and much of the violence at school is related to competition for status and status-related confrontations. Ability tracking also contributes to a collective adaptation to school failure and peer rejection by grouping academically poor students and those who are aggressive troublemakers together in the same classes. Delinquent peer groups tend to emerge out of these classes and individual feelings of anger, rejection and alienation are mutually reinforced in these groups. The strongest and most immediate cause of the actual onset of serious violent behavior is involvement with a delinquent peer group. It is here that violence is modeled, encouraged, and rewarded; and justifications for disengaging one's moral obligation to others are taught and reinforced. The effects of early exposure to violence, weak internal and family controls and aggressive behavior patterns developed in childhood all influence the type of friends one chooses, and the type of friends, in turn, largely determines what behavior patterns will be modelled, established and reinforced during adolescence. However, a strong bond to parents is a protective factor which insulates youth from the influence of delinquent friends as long as the friendship network is not dominated by such youth.

Gangs are a subtype of adolescent peer group, with a more formal identity and membership requirements. They tend to involve more homogeneously delinquent youth, often actively recruiting persons for their fighting skills or street smarts. In some instances membership entails violent behavior as an initiation ritual. However, not all gangs are involved in serious violent behavior or drug distribution. They often serve some positive functions, particularly in disorganized neighborhoods. They not only provide youth a sense of acceptance, belonging and personal worth (which most friendships do), but also a safe place to stay, food, clothing and protection from abusive parents. But like delinquent groups more generally, joining a gang greatly increases the risk of serious violence, both perpetration and victimization. Likewise, leaving a gang or delinquent peer group substantially reduces the risk of serious violence.

Alcohol, Illicit Drugs and Firearms

The relationship between substance use and violence is complicated. Alcohol is implicated in over half of all homicides and of assaults in the home. Parents who abuse alcohol (and illicit drugs) are more prone to be physically abusive and neglectful of their children. But while problem drinkers are more likely to have a history of violent behavior, they are not disproportionately represented among violent offenders as compared to non-violent offenders. Pharmacological studies find no simple dose-response relationship between alcohol use and violent behavior. While alcohol is clearly implicated in violent behavior, the exact mechanism has not yet been established.

In general, the use of psychoactive drugs has not been linked pharmacologically to violent behavior. The effect of marijuana and opiate drugs actually appears to inhibit violence, although withdrawal may precipitate an increased risk of violence. There is some evidence that drug addicts commit violent crimes to support their drug habit, but this appears to be a relatively rare phenomenon. The clearest drug-violence connection is for selling drugs; the drug distribution network is extremely violent.

Since 1985 the firearm-related homicide rate for adolescents has increased over 150 percent and firearms now account for nearly three-fourths of all homicides of young black men. Surveys estimate that 270,000 guns are taken to school each day. It is not clear that the increase in gun-related violence is simply the result of greater gun availability. However, violent events involving guns are 3 to 5 times more likely to result in death than those involving knives, the next most lethal weapon.

Not much is known about why today's youth, in increasing numbers, are carrying guns. Anecdotal evidence suggests it is to "show off", to insure "respect" and acquiescence from others, or for self-defense. In part, it appears to be a response to the perception that public authorities cannot protect youth or maintain order in their neighborhoods or at school. There is evidence that dropouts, drug dealers and those with a prior record of violent behavior are more likely to own a gun than are other adolescents. And the vast majority (80 percent) of firearms used in crimes are obtained by theft or some other illegal means.

The Adolescence-Adulthood Transition

The successful transition into adult roles (work, marriage, parenting) appears to reduce involvement in violent behavior. In one national study, nearly 80 percent of adolescents who were serious violent offenders reported no serious violent offenses during their adult years (to age 30). However, nearly twice as many black as white youth continued their offending after age 21. Among those employed at age 21, rates of continuity were low and there were no differences in rates of continuity by race.

As noted earlier, race and class differences in serious violent offending are small during adolescence, but become substantial during the early adult years. This difference does not appear to be the result of differences in predispositions to violence, but in the continuity of violence once initiated. Race, in particular, is related to finding and holding a job, and to marriage and stable cohabiting rates. In essence, race and poverty are related to successfully making the transition out of adolescence and into adult roles.

It appears that growing up in poor, minority families and disorganized neighborhoods has two major effects directly related to violent behavior. First, when it comes time to make the transition into adulthood, there are limited opportunities for employment which, in turn, reduces the chances of marriage. These are two primary definers of adult status. Second, there is evidence that growing up in poor, disorganized neighborhoods inhibits a normal course of adolescent development. Youth from these neighborhoods have lower levels of personal competence, self- efficacy, social skills, and self-discipline. Many are not adequately prepared to enter the labor market even if jobs were available. They are, in some ways, trapped in an extended adolescence and continue to engage in adolescent behavior.

WHAT IS KNOWN ABOUT THE PREVENTION AND CONTROL OF YOUTH VIOLENCE?

Since most violent behavior is learned behavior, the general strategy for prevention and treatment interventions should be 1) to reduce the modeling and reinforcement of violence as a means of solving problems and manipulating or controlling the behavior of others and 2) to ameliorate those social conditions which generate and support violent lifestyles. The most effective strategy for accomplishing this is to insure a healthy course of child and adolescent development for all youth, so they are prepared to enter productive, responsible adult roles; and to insure that these roles are accessible.

Individual Level Interventions

Several individual level interventions appear promising: Head Start programs, Teaching Family group homes, parent effectiveness training, behavioral skill training and some types of employment programs. The reductions in crime, violence and substance use are relatively modest from these programs, and may be relatively short-term effects. Teaching family programs, for example, demonstrate good effects while youth are in these homes, but when they leave this treatment setting and return to their own homes and neighborhoods, these effects are quickly lost. Other programs are too narrowly targeted to a specific context or focus upon improving personal competence without any significant changes in opportunity structures. They are also frequently used as general interventions when they are developmentally appropriate for selected age-groups. But these interventions hold promise as components for a more comprehensive, integrated intervention effort. Counseling and casework approaches had no significant effects and some programs, e.g., shock incarceration and boot camps, appeared to have negative effects.

Neighborhood or Community Interventions

This approach is a comprehensive one which attempts to bring together all of the primary institutions that serve youth, e.g., families, health agencies, schools, employment, and justice, in an integrated, coordinated effort to develop an effective neighborhood organization and deliver the full range of needed services at a single site under a single administrative structure. Such programs include family support programs, community development corporations, and school-based clinics. Unfortunately, there are few good evaluations of these neighborhood level approaches. In too many cases, neighborhood programs fail to develop a comprehensive range of services or a cohesive neighborhood organization which is an essential to this approach. However, the evidence indicates that when such programs are well implemented, they improve the emotional well-being of families, expand and develop informal socialnetworks, and facilitate a successful course of youth development. Theoretically, if sustained over five years or more, this approach should have the greatest payoff in reducing violence, crime and drug abuse, and facilitating a successful course of child and adolescent development.

Gun Control Policies

There is relatively little rigorous research on the effectiveness of various gun control policies. However, there is some evidence for the effectiveness of restrictive handgun laws and mandatory sentences for firearm offenses. In the case of restrictive handgun laws, several studies have found significant declines in homicide rates with no evidence that other weapons were used as substitutes for firearms. There was some evidence that other weapons were being employed as substitutes in the studies of mandatory sentencing laws. The little evidence on the effectiveness of waiting periods suggests little or no effect on homicide rates. More research is needed in this area to establish the effects of various gun control measures.

Justice System Responses

Since 1985, waivers to adult court have increased dramatically for violent offenses and drug-related offenses. Research on the effects of this policy indicates: 1) longer processing time and longer pre-trial detention, 2) higher conviction rates and longer sentences, 3) disproportionate use of waivers for minority youth, and 4) a substantially lower probability of treatment while in custody. The last two findings raise serious questions about the use of waivers. Restitution is an effective policy. The compliance rate for restitution orders is over 90 percent, and there are modest reductions in recidivism. There is no clear evidence that increases in sentence length or confinement in adult institutions have any significant deterrent effect over shorter sentences and confinement in juvenile institutions. If we consider the situation relative to drug-related offenses, research demonstrates that the decline in adolescent drug use was primarily the result of educational awareness programs and community-based prevention programs; drug enforcement policies involving mandatory sentences and stronger sanctions appear to have had very small deterrent effects. While prevention programs take longer and are more difficult to implement, the violence reduction effects of prevention programs are substantially greater and probably cost no more.
________________________________________________________

* Throughout this report, the term "violence" will be used to describe physical assaultive behavior and the term "serious violent offenses" will refer to aggravated assaults, forcible rapes, robberies and homicides.