Youth Violence Prevention: The State of the Science

Mark A .Mattaini
Jane Addams College of Social Work
University of Illinois at Chicago

Christine T. Lowery (Laguna/Hopi)
University of Wisconsin-Milwaukee

and the PEACE POWER! Working Group

Violent acts perpetrated by and on youth have received enormous media and political attention in recent years, particularly high-visibility cases like those in which young men open fire with firearms on schoolmates, teachers, or family members. Such cases are shocking and naturally create considerable societal concern. Nearly a quarter of children in grades 3-12 are afraid of being hurt at or near school (and in some areas many more) (Louis Harris and Associates, 1993). Twenty percent of high school students report carrying a weapon in the past 30 days, and about half of them have done so on school property (CDC, 1998).

School violence is one aspect of the problem, but violence on the street, in homes, and in dating relationships is more than twice as common (Kaufman et al., 1998). Homicide rates are much higher in the U.S. than in any other industrialized country, as are the rates for most--but not all--other violent crimes (Reiss & Roth, 1993; Langan & Farrington, 1998). The risk of victimization by violence in the U.S. peaks between the ages of 18 and 21; while persons age 12 to 24 comprised 22% of the population, 49% of victims of serious crime were in this group (Perkins, 1997). Rates of arrests for violent crimes are also highest for men age 15 through 24 (Reiss & Roth, 1993); homicide arrests among youth 14-17 years of age increased 41% between 1989 and 1994, while decreasing by 19% for persons over 25 during the same period (CDC, 1997). People of color are at substantially higher risk than are whites in all age groups. Homicide is the leading cause of death for African-American and Hispanic youth, both male and female (CDC, 1997).

The problem of youth violence affects the quality of life for everyone, especially the most vulnerable in society. Violence prevention, therefore, has been a major focus of research funded by the Centers for Disease Control for the past 15 years (CDC, 1997), and increasing attention and funding have been directed to the issue in the 1990s by many levels of government and private funders. Behavioral science data now provide substantial guidance for selecting approaches likely to be successful, and identifying those that are not. Simplistic approaches that primarily tell young people not to fight or be violent ("Just Say No") are unlikely to be adequate given the contextual conditions and events within which youth are embedded. This is particularly true in cases where they see no other avenues for influencing their worlds--violence offers power, if only a limited and ultimately weak form of power. While individual cases may seem difficult to comprehend, the overall problem of violence among youth is consistent with a sociocultural context that is in many ways organized around coercive power, including threats of force and violence (Lowery & Mattaini, in press-b; Sidman, 1989). Nonviolence is far more likely to emerge from empowerment cultures, which can be constructed and supported in organizations, families, and communities (Lowery & Mattaini, in press-a).

This paper begins with a review of what contemporary behavioral science has learned about violence and its prevention. Related research and theory are organized in ways that suggest novel preventive strategies. Based on these reviews and analyses, this paper then points toward empirically-supported programs for constructing cultures of nonviolence and constructive, collective empowerment.

The Science of Violence Prevention

An enormous amount of research has been devoted to violence, and youth violence in particular, over the past decade. While some authors indicate that little is really known about the problem and how to prevent it (e.g., Nemecek, 1998), a great deal has actually been learned. In a comprehensive review of the state of current knowledge, the American Psychological Association (1993) noted:

Effective intervention programs share two primary characteristics: (a) they draw on the understanding of developmental and sociocultural risk factors leading to antisocial behavior, and (b) they use theory-based intervention strategies with known efficacy in changing behavior, tested program designs, and validated, objective measurement techniques to assess outcomes. (p. 53-54)
The National Research Council, after completing a thorough review of research on violence at all levels from the biological to the sociocultural, prepared a four-volume report that included a range of data-based recommendations, while also clarifying the limitations of current knowledge and areas in need of further research (Reiss, Miczek, & Roth, 1994; Reiss & Roth, 1993, 1994a, 1994b) A recent review of the current state of knowledge from the perspective of the science of behavior also suggested specific, data-based strategies (Mattaini, Twyman, Chin, & Lee, 1996).

In addition, research provides substantial guidance as to what doesn't work, or doesn't work well enough. For example, the Violence Prevention Curriculum for Adolescents (Prothrow-Stith, 1987) is a school-based program focused on anger management that may be useful in some cases. Unfortunately, however, evaluation of this somewhat unidimensional program have proven disappointing (DeJong et al., no date; Reiss & Roth, 1993), and the developers have recently moved toward approaches with a broader community focus. "Conflict resolution" programs, similarly, deal only with a narrow class of situations in which violence may occur, often direct attention only to events very late in the chain that produces violence, and may paradoxically enhance attention to, and therefore increase, undesirable behavior (Mattaini, Twyman, Chin & Lee, 1996). Comprehensive prevention requires a strategy that is responsive to the determinants of violence at multiple levels, including the biological, behavioral, and sociocultural.

Biological, Behavioral, and Cultural Factors in Violence

Skinner (1981) clarified that behavior is "selected" at three levels : the biological/genetic, behavioral, and cultural. Considerable data are now available at each of these levels, as well as information about the connections among them, which can guide the development of effective interventive strategies. This material is selectively reviewed below, with particular emphasis on manipulable variables that can be addressed by community-based prevention programs. Realistic projections indicate that integrated attention to these variables could lead to reductions in violence of up to 50% (Embry, in press).

Biological selection. While individual genetic differences affect predispositions for all forms of behavior, even sociobiologists agree that rapid societal changes observed in the incidence of violence cannot be accounted for by genetic change; genes simply don't change that quickly (Wilson & Herrnstein, 1985). Although some persons are more prone to violence than others, such variation is a universal phenomenon that some cultures apparently handle better than do others. Other physiological factors, however, apparently do affect levels of violence, including exposure to environmental toxins like lead or in utero exposure to alcohol or certain other drugs (Mirsky & Siegel, 1994).

Violence is not a simple response to biological factors. Like driving a car, violence involves physiology, but is organized at the level of environmental transactions. Biological factors can affect sensitivity to environmental transactions and conditions, however. Biobehaviorists Miczek, Mirsky, Carey, Debold, and Raine (1994), for example, indicate that:

In animals, testosterone (or its metabolites) has effects on the probability of aggressive responses to conspecifics or other environmental events ... androgens are not stimulating aggressive behavior in vacuo; more accurately, they appear to be altering the response to aggression-provoking stimuli. (p. 6)
This is one example among many in which biological factors affect the probability that violent actions will occur by changing sensitivity to the environmental events.

These biological effects operate in significant part at the level of hormones and neurotransmitters (Carlson, 1994); the immediate levels of certain neurotransmitters in critical regions of the brain and structural changes resulting from accumulated life experiences are important substrata of behavior. These effects are enormously complex and interactive (Capra, 1996), but some basic operations appear to be well established. For example, describing the effects of three crucial neurotransmitters, dopamine, serotonin, and norepinephrine, Embry and Flannery (in press), state:

A person who has just earned a primary reward, a social reward and recognition releases dopamine in the nucleus accumbens and ventral tegnmental areas--which communicate with regions in the mesocortex, cortex and frontal lobes forming the basis for long-term planning. Touch, affection and positive status release serotonin molecules, which inhibit offensive attacks by stimulating serotonergic axons in the forebrain and amygdala. Threats and aversive events (both conditioned and unconditioned) elevate norepinephrine activity from the brain stem, amygdala, and forebrain. The arousal makes the person more vigilant and defensive.

Low levels of serotonin are associated with depression and aggression, whereas high levels that result from positive social exchange are associated with experiences of safety and security and enhance learning. Adequate levels of dopamine in critical brain regions leads to improved behavioral control and social responsiveness. Dopamine can be increased by reinforcers of many kinds, including the experience of winning a coercive struggle, the use of drugs like cocaine, or experiencing a social reward for excellent performance. There are many kinds of reinforcers; fighting increases dopamine, for example, as does precision praise for positive effort (Embry & Flannery, in press). An well-established mathematical formula (the matching law, Mattaini, 1991; McDowell, 1988) predicts that in an impoverished environment, levels of violence will be high, while in environments rich in other reinforcers, the levels will be low, because reinforcers are to a significant extent substitutable. The hyperbolic effect is demonstrated in Figure 1.

hyperbolic

Figure 1. The relationship between levels of environmental reinforcers and rate of violent behavior.

Frequent, targeted rewards increase levels of dopamine, some of which is converted to serotonin. Levels of serotonin can also be increased through social and affectionate reinforcers, including kind words, praise and physical touch (all else being equal). In fact, levels of serotonin increase in both the recipient and the giver in these exchanges (Embry, in press).

Experiences of threat and coercion, on the other hand, lead to decreases in available serotonin, and the release of norepinephine, cortisol, and other hormones and neurotransmitters which increase physiological arousal, environmental scanning for threat, and readiness for aggression, protection or flight. Traumatic experiences, particularly severe and repeated traumatic experiences, also strengthen brain traces in a process of tuning, increasingly attention and sensitivity to potential threat‹a process that occurs, for example, in the development of Post Traumatic Stress Disorder (Castillo, 1997). These experiences therefore ultimately affect not only the momentary state, but also the structure, of the brain. The implications of these biobehavioral phenomena for prevention are elaborated below.

Behavioral selection. Violence is behavior. Decades of research have taught us that "intentional" or instrumental behavior is selected over time because it works. Actions that pay off tend to be repeated, and those that don't, eventually tend to fade. While there are many complexities, and learning histories are difficult to capture in simple ways, behavior in general occurs because it is in some way functional, and violence is no exception (Mattaini, Twyman, Chin & Lee, 1996). Youth who rely on coercive threats and violence have learned over time by doing so they can sometimes escape or avoid negative events, and/or obtain some level of positive payoffs. Even violence associated with "rage" appears to relieve an aversive physiological condition. Many such youth have not learned alternative ways to achieve those ends, particularly those that involve long-deferred positive outcomes.

Violence offers some "power," some influence over one's world. This power is limited and of poor quality, but like battering (which usually involves the assertion of power and control, Pence & Paymar, 1993), for youth with few options, some power is better than none. Asking youth, therefore, to "Just Say No" to violence is asking them to give up what may be the only power they have so far discovered. Behavioral theory and research consistently indicate how unlikely such a strategy is to succeed (Mattaini, Twyman, Chin, & Lee, 1996). What is needed are alternative repertoires, alternative ways to exert power, to "do power," in one's life.

In addition to material drawn from theory and the existing research base, the first author and two colleagues conducted a series of focus groups for youth and teachers at the intermediate and high school levels in two co-located schools. The purpose of the focus groups was to further specify possible functions of fighting and threats among those we were working with. Each group was conducted by a university-based researcher and a community agency-based MSW, one of whom led the discussion while the other took notes; the groups were also videotaped. The leaders discussed each group after it was completed to clarify the notes, and the leaders and a third observer (a doctoral student) watched each videotape. These three observers agreed, after extensive discussion, that the basic clusters of functional outcomes produced by violent or potentially violent episodes as identified by the participants, could be described as those shown in Figure 2.

focus

Figure 2. Clusters of violent and coercive incidents identified by teachers and students in focus groups.

The major functional outcomes identified through this qualitative strategy are shown in the circles, while occasions and motivating events in which each was seen as likely are connected to the circles by arrows. Many episodes involve more than one outcome, so this figure should be viewed as an interconnected network.

At least four of the major clusters produce rewards of one kind or another. Perhaps the most frequent cluster involved gaining status or face; many different situations occasioned this pattern, and there were many varieties, from bragging and showing off to "snapping" (rapid, reciprocal, and often creative put-downs). A second cluster of behaviors produced sexual attention. A third cluster of behaviors were seen as motivated by getting others to do what one wishes, to assert coercive power and control. Some of the examples given involved threats and violence within dating relationships, but the general pattern clearly operates in other situations as well. The fourth cluster, consistent with the biological research, consisted of aggressive acts for "fun" (perhaps achieving a "dopamine high," Embry & Flannery, in press).

Both students and teachers were clear that there was a separate cluster of actions that produced escape or relief from aversive conditions, including, for example, situations where academic deficiency became obvious, embarrassment about lack of expensive clothing or jewelry, or the pain involved with thinking about problems at home. Both the reward and escape categories were important among this sample, and probably are in others as well (Mattaini, Twyman, Chin, & Lee, 1996). An effective program or strategy for prevention needs to help youth find alternative repertoires that are functionally similar, that produce these valued outcomes in other ways. These may include, for example, finding new ways to gain recognition, new ways to relate to boyfriends and girlfriends that produce security, and new ways to avoid or obtain relief from aversive conditions.

Finally, a sixth cluster was identified, one which is not perhaps functionally different than the others discussed, but needs to be noted. Some of the behavior the youth manifested involved patterns that were functional in other settings, for example on the street in certain areas. When these repertoires generalized to the school setting, however, they led not to survival and problem-solving, but to confrontations with school staff and peers. In other words, the issue is not always lack of skill, but in some cases one of stimulus overgeneralization. Maintaining a "tough" stance on the street is sometimes a survival skill, but the same pattern at school can lead to failure.

Both family and peers may also shape coercive and violent behaviors. There is often a false dichotomy drawn between the two, particularly in the popular literature. Crucial longitudinal studies conducted by Patterson and his colleagues (Patterson, DeBaryshe, & Ramsey, 1989; Patterson, Reid, & Dishion, 1992) suggest that the most common developmental pathway for antisocial youth is as follows:

1) Early in life, parenting is chaotic and coercive; an accumulation of severe environmental and intrafamilial aversives (for example, poverty, drug abuse, and family violence) is often present. Antisocial models are common in these homes, as is extensive reliance on physical punishment and other forms of harsh, inconsistent discipline (Biglan, 1995). By early childhood, antisocial children learn that aggressive and other antisocial repertoires help them to escape aversives and obtain reinforcers in the family, and often have learned no alternative repertoires for doing so.
2) When the child begins school, he (most but not all are male) brings those social repertoires with him, and is therefore rapidly rejected by both prosocial peers and teachers.
3) Academic failure follows quickly, and bullying may develop as one way to obtain some level of social contact and control.
4) The child often finds that his only available peer group consists of other antisocial children, whose repertoires are familiar. As described by Walker, Colvin, and Ramsey (1995), by about grade 4 or 5, many antisocial children become members of such deviant peer groups, which then encourage further antisocial behavior. Some, however, remain socially isolated, and may engage primarily in covert antisocial behaviors.
Both the microculture of the family, and groups of antisocial peers contribute to these patterns. These, in turn, are often shaped by larger sociocultural factors, the third level of behavioral selection, in a society that glorifies violence in the media, and relies heavily on coercive and adversarial processes in many realms, including economics, the justice system, education, and informal social relations (Lowery & Mattaini, in press-b). At least for these youth, perhaps Vizenor (1991) is not far wrong in characterizing ours as a "culture of death."

Cultural selection. Violent acts are often shaped and supported not only by family and friends, but by larger cultural forces, including the culture of the street, the media, and even, paradoxically, the justice system. In his book, Fist Stick Knife Gun, Geoffrey Canada (1995) describes the changes he has seen over four decades in the life experience of children in Harlem, where he grew up and now works with youth. As indicated in the title, one crucial shift has been the continuing escalation in lethality of violent episodes due to the availability of weapons. Issues that once might have been resolved with fists may now involve automatic weapons. In addition, Canada talks about how the culture of the street shapes and demands coercive and violent action, and how families may reluctantly come to believe that they must encourage their children to participate in that culture for their own protection, and that of the family.

A culture of violence, however, is not confined to low-income inner cities. Butterworth (1998) reports in the New York Times on a culture of "honor" in the rural Southern states, which appears to be the root of most of the excess homicides in the U.S. as compared with European countries. The 1996 murder rate in Louisiana, for example, was 17.5 murders per 100,000 people, while that of South Dakota was 1.2 per 100,000. The majority of homicides in the former slaveholding states involve personal affronts, while most murders in other parts of the country begin with other crimes like robbery. Slavery appears to have reinforced the need to be ready to take violent action on a moment's notice, and research (e.g., Cohen & Nisbett, 1994; Cohen, Nisbett, & Bowdle, 1996) indicates that this culture persists to the present.

Despite occasional claims to the contrary, it is now clear that there is, for many youth, a strong association between media violence and aggressive behavior (American Psychological Association, 1993; Reiss & Roth, 1993). There has been a dramatic rise in exposure to increasingly realistic models of violence in movies, television and video games in recent years. While these may not significantly affect behavior among those with strong rule-governed repertoires, the data clearly suggest that many youth without those repertoires can be powerfully affected by such exposure. When combined with models, support, and even demands for coercive and violent action from peers and family, and given the ready availability of firearms, it is not surprising that the incidence and lethality of violent events in the US is as high as it is.

In addition, the justice system in the US incarcerates a far higher proportion of the population than any other developed country (Reiss and Roth, 1993), and increasingly turns to capital punishment, even though such punishment is recognized in the Universal Declaration of Human Rights (United Nations, 1948) as a violation of basic human rights. The justice system relies on threats and punishment, and it is often difficult to even imagine alternatives. Interestingly, however, both indigenous justice models (Ross, 1996; see below) and the science of behavior (Cohen & Filipczak, 1971) indicate that alternatives that emphasize the construction of responsible repertoires and healing processes can be more effective. Both also clarify that current practices may, in fact, intensify rather than attenuate the level of violence in society.

The scientific data make clear that the exercise of coercive power is ultimately weak, and produces a wide range of undesirable side-effects, ranging from counter-aggression to depression (Sidman, 1989). Coercive arrangements are ever at risk of collapse, and therefore require constant surveillance. They also damage the web of community relationships on which all of society depends. Asking anyone, be it an individual young person or the justice system, to simply give up even this low-quality power is unrealistic, however, unless we can offer other means of exercising some influence over one's world.

Constructing Alternatives to Violence

The PEACE POWER! strategy is designed to offer youth access to constructive and nonviolent power, and to construct cultures that support such action in organizations and communities. PEACE POWER! is not a program or a curriculum; rather it is a behavioral strategy that can be flexibly adapted to meet local conditions. Fawcett (Fawcett, Mathews, & Fletcher, 1980; Fawcett, 1991) has emphasized that effective community research and action requires strategies that are decentralized, sustainable with local resources, and in particular flexible enough to be adapted to meet local needs and values. Rather than a rigid program, therefore, the PEACE POWER! approach is to provide empirical information about "behavior-environment relations of importance to communities" (Fawcett, 1991, p. 629), and an extensive menu of possible action steps that the community can take or adapt consistent with the current state-of-the-science.

As is true of the PeaceBuilders(tm) program for violence prevention with younger children (Embry, Flannery, Vazsonyi, Powell, & Atha, 1996), PEACE POWER! relies heavily on principles of social learning, in which common language, repetition with variations, and extensive modeling ensure that the core program practices become deeply embedded in the cultural entities (schools, agencies, communities) sponsoring the program. Everyone, including youth and adults, is immersed in these practices. Program delivery, therefore, relies on prompting and reinforcing actions by teachers, parents, staff, and other community members, as well as for youth themselves. Prompts for positive action, including materials for self-monitoring, and structured recognition for participation, are structured for each group of stakeholders; some examples are provided in later sections of this paper. The general delivery strategy involves workshops for each group of stakeholders, during which tools and materials are provided from which they can construct their own local program. On-going consultation is then provided to assist in operationalizing and implementing these efforts. Complementary programming efforts, including Life and Leadership Skills Training for youth, and organizational consultation around implementing consistent positive structure, can further enhance local programs.

The Core PEACE POWER! Practices

Four clusters of cultural practices provide the foundation of the PEACE POWER! strategy. If these core practices are adequately embedded in a local effort, the major determinants of violent behavior at all three levels of selection discussed above will be addressed. Each practice is consistent with the basic principles of the science of behavior, and each has extensive empirical support. The four core cultural practices included in the PEACE POWER! strategy are shown in Figure 3. These four practices are recursive and interconnected. Available data suggest that emphasis on each may be required for optimal results.

Wheel

Figure 3. Core data-based practices embedded in an empowered culture of non-violence.

Programs based on these practices have consistently demonstrated positive outcomes. For example, several programs have emphasized high levels of recognition and reinforcement. Mayer and his colleagues (Mayer, Butterworth, Nafpaktitis, & Sulzer-Azaroff, 1983) implemented a program in elementary and junior high school in Los Angeles County in the early 1980s in which high levels of social recognition produced dramatic reductions in vandalism costs (often an accessible marker variable for the range of antisocial behaviors), and improved discipline significantly. Embry and his colleagues (Embry, Flannery, Vazsonyi, Powell, & Atha, 1996; Embry & Flannery, in press) extended Mayer's work in developing the PeaceBuilders(tm) program. Their preliminary results indicate reductions of disciplinary actions and violence of between 50 and 60%, and 77% of teachers indicated that the program made their jobs easier (see http://www.peacebuilders.com). Mayer, Mitchell, Clementi, Clement-Robertson, Myatt, & Bullara (1993) implemented a drop-out prevention program that relied on changing the balance of reinforcement versus punitive exchanges for high-risk ninth graders and staff. Students involved in this program, who had formerly been on-task from 8 to 35% of the time, were found to be on task 70 to 100% of the time after implementation; suspensions dropped, and their drop-out rate declined to less than the district average. Data for violence were not reported, but these variables tend to be significantly correlated with aggression, coercion, and violence. School staff who have worked with the PEACE POWER! project report that the PEACE POWER! practices are useful at the intermediate and high school levels (Mattaini, Lowery, Herrera, & DiNoia, under review).

The PEACE POWER! Working Group has produced a range of tools that can be adapted by local programs (Mattaini & the PEACE POWER! Working Group, in press). For more information on the PEACE POWER! strategy, tools, training and consultation, or the PeaceBuilders program, click on those links. For further scientific information about the biological, behavioral, and cultural factors that support these approaches, and their outcomes, please contact Behaviorists for Social Responsibility.


Return to State of the Science Papers


References

American Psychological Association. (1993). Violence and youth: Psychology's response. Washington, DC: American Psychological Association.

Biglan, A. (1995). Changing cultural practices. Reno, NV: Context Press.

Butterworth, F. (1998, July 26). Why the South's murder rate is so high. The New York Times on the Web. (http://www.nytimes.com)

Canada, G. (1995). Fist stick knife gun. Boston: Beacon Press.

Capra. F. (1996). The Web of Life. New York: Anchor Books.

Carlson, N. R. (1994). Physiology of behavior. Boston: Allyn & Bacon.

Centers for Disease Control and Prevention (CDC) (1997). Youth Violence in the United States. Washington, DC: Author. Retrieved December 4, 1998, from the World Wide Web: http://www.cdc.gov/ncipc/dvp/yvfacts.htm

Centers for Disease Control and Prevention (CDC) (1998). Facts About Violence Among Youth and Violence in Schools. Washington, DC: Author. Retrieved December 4, 1998, from the World Wide Web: http://www.cdc.gov/od/oc/media/fact/violence.htm

Cohen, D., Nisbett, R. E., & Bowdle, B. F. (1996). Insult, aggression, and the southern culture of honor: an "experimental ethnography." Journal of Personality and Social Psychology, 70, 945-960.

Cohen, D., & Nisbett, R. E. (1994). Self-protection and the culture of honor: Explaining Southern violence. Personality and Social Psychology Bulletin, 20, 551-567.

DeJong, W., Spiro, A. III, Wilson-Brewer, R., Vince-Whitman, C. & Prothrow-Stith, D. (no date). Evaluation Summary: Violence Prevention Curriculum for Adolescents. Newton, MA: Educational Development Center, Inc.

Embry, D. D. (in press). Reasons for hope: Creating a climate for change and resiliency. In W. L. Reed (Ed.), Violence and childhood trauma: Understanding and responding to the effects of violence on young children.

Embry, D.D., & Flannery, D. J. (in press). Two sides of the coin: Multi-level prevention and intervention to reduce youth violent behavior. In D.J. Flannery and C. Ronald Huff (eds.), Youth Violence: Prevention, Intervention and Social Policy. Washington, DC: American Psychiatric Press.

Embry, D. D., Flannery, D. J., Vazsonyi, A. T., Powell, K. E., & Atha, H. (1996). PeaceBuilders: A theoretically driven, school-based model for early violence prevention. American Journal of Preventive Medicine, 12 (Supplement 2), 91-100.

Fawcett, S. B. (1991). Some values guiding community research and action. Journal of Applied Behavior Analysis, 24, 621-636.

Fawcett, S. B., Mathews, R. M., & Fletcher, R. K. (1980). Some promising dimensions for behavioral community technology. Journal of Applied Behavior Analysis, 13, 505-518.

Kaufman, P., Chen, X., Choy, S., Chandler, K. A., Chapman, C. D., Rand, M. R., & Ringel, C. (1998). Executive summary: Indicators of school crime and safety, 1998. Washington, DC: U.S. Departments of Education and Justice. NCES 98-251/NCJ-172215.

Langan, P. A., & Farrington, D. P. (1998). Crime and justice in the United States and in England and Wales, 1981-96. Washington, DC: U.S. Department of Justice. NCJ-173402.

Louis Harris and Associates, Inc. (1993). The Metropolitan Life survey of the American teacher 1993. New York: Louis Harris and Associates, Inc.

Lowery, C. T. & Mattaini, M. A. (in press-a). The co-construction of empowerment cultures in child welfare. For publication in W. Shera & L. Wells (Eds.), International perspectives on empowerment practice. New York: Columbia University Press.

Lowery, C. T., & Mattaini, M. A. (in press-b). Shared power in social work: A Native American perspective. In H. Briggs and K. Corcoran (Eds.), Structuring Change (2nd ed.). Chicago: Lyceum.

Mattaini, M. A. (1991). Choosing weapons for the war on "crack": An operant analysis. Research on Social Work Practice, 1, 188-213.

Mattaini, M. A., and the PEACE POWER! Working Group (in press). Constructing cultures of non-violence: The PEACE POWER! toolkit. In D. S. Sandhu (Ed.), Violence in American schools: A practical guide for counselors. American Counseling Association.

Mattaini, M. A., Lowery, C. T., Herrera, K., & DiNoia, J. (under review). PEACE POWER! Developmental research for constructing alternatives to youth violence.

Mattaini, M. A., Twyman, J. S., Chin, W., Nam, K. (1996). Youth violence. In M. A. Mattaini & B. A. Thyer (Eds.), Finding solutions to social problems: Behavioral strategies for change (pp. 75-111). Washington, D.C.: American Psychological Association Books.

Mayer, G. R., Butterworth, T., Nafpaktitis, M., & Sulzer-Azaroff, B. (1983). Preventing school vandalism and improving discipline: A three-year study. Journal of Applied Behavior Analysis, 16, 355-369.

Mayer, G. R Mitchell, L. K., Clementi, T., Clement-Robertson, E., Myatt, R, & Bullara, D. T. (1993) A dropout prevention program for at-risk high school students: Emphasizing consulting to promote positive classroom climates. Education & Treatment of Children, 16, 135-146.

McDowell, J. J. (1988). Matching theory in natural human environments. The Behavior Analyst, 11, 95-109.

Miczek, K. A., Mirsky, A. F., Carey, R., Debold, J., & Raine, A. (1994). An overview of biological influences on violent behavior. In A. J. Reiss, Jr., K. A. Miczek, & J. A. Roth (Eds.), Understanding and preventing violence: Volume 2: Biobehavioral influences (pp. 1-20). Washington, DC: National Academy Press.

Mirsky, A. F., & Siegel, A. (1994). The neurobiology of violence and aggression. In A. J. Reiss, Jr., K. A. Miczek, & J. A. Roth (Eds.), Understanding and preventing violence: Volume 2: Biobehavioral Influences. Washington, DC: National Academy Press.

Olweus, D., Mattsson, A., Schalling, D., & Low, H. (1988). Circulating testosterone levels and aggression in adolescent males: A causal analysis. Psychosomatic Medicine, 50, 261-272.

Patterson, G. R., DeBaryshe, B. D., & Ramsey, E. (1989). A developmental perspective on antisocial behavior. American Psychologist, 44, 329-335.

Patterson, G. R., Reid, J. B., & Dishion, T. J. (1992). Antisocial boys: Vol. 4. A social interactional approach. Eugene, OR: Castalia.

PeaceBuilders (tm) (1997). ActionGuide. Tucson: Author.

Pence, E., & Paymar, M. (1993). Education groups for men who batter: The Duluth model. New York: Springer.

Perkins, C. A. (1997). Age patterns of victims of serious violent crime. Washington, DC: U.S. Department of Justice. NCJ-162031.

Prothrow-Stith, D. (1987). Violence prevention curriculum for adolescents. Newton, MA: Educational Development Center.

Reiss, A. J., Jr., Miczek, K. A., & Roth, J. A. (Eds.) (1994). Understanding and preventing violence: Volume 2: Biobehavioral Influences. Washington, DC: National Academy Press.

Reiss, A. J., Jr., , & Roth, J. A. (Eds.) (1993). Understanding and preventing violence. Washington, DC: National Academy Press.

Reiss, A. J., Jr., & Roth, J. A. (Eds.) (1994a). Understanding and preventing violence: Volume 4: Consequences and Control. Washington, DC: National Academy Press.

Reiss, A. J., Jr., & Roth, J. A. (Eds.) (1994b). Understanding and preventing violence: Volume 3: Social Influences. Washington, DC: National Academy Press.

Ross, R. (1996). Returning to the teachings: Exploring aboriginal justice. Toronto: Penguin.

Sidman, M. (1989). Coercion and its fallout. Boston: Authors Cooperative.

Skinner, B. F. (1981). Selection by Consequences. Science, 213, 501-504.

United Nations (1948). Universal declaration of human rights. New York: Author.

Vizenor, G. (1991). The heirs of Columbus. New York: Quality Paperback Book Club by arrangement with Wesleyan University Press.

Walker, H. M., Colvin, G., & Ramsey, E. (1995). Antisocial behavior in school: Strategies and best practices. Pacific Grove, CA: Brooks/Cole.

Wilson, J. Q., & Herrnstein, R. J. (1985). Crime and human nature. New York: Simon & Schuster.

Latest Revision: June 6, 1999