Development of the Social Ecology Model of Adolescent Interpersonal Violence Prevention (SEMAIVP). (Research Papers).
Schools receive adolescents who daily must deal with emerging lifestyles that tolerate weapons, drugs, and dysfunctional behavior. Adolescents bring the impact of these experiences to school. Social ecology offers a framework for understanding the complexity of adolescent violence. It provides an inclusive basis for schools, youth service agencies, juvenile justice organizations, and community groups to collaborate and provide multilevel prevention and intervention strategies. (1,2)
Adolescent violence represents a significant public health problem ranging in seriousness from pushing, shoving, and fighting to criminal acts in which adolescents become both victims and perpetrators. The incidence of adolescent interpersonal violence began to increase in the 1980s, peaked in the 1990s and, while declining since 1994, remains higher now than in the mid 1980s. (3) The Youth Risk Behavior Survey recorded this decline in interpersonal violence. Between 1991 and 1997, the percentage of 12- to 17-year-old adolescents who reported involvement in a physical fight decreased from 42.5% to 36.6%; the percentage of students injured in a physical fight decreased from 4.4% to 3.5%; and the percentage of students in a physical fight on school property decreased from 16.2% to 14.8%. (4)
Interpersonal violence is often viewed as one of several delinquency problems associated with adolescence. Social ecology models have been used to study drug use, (5) alcohol use, (6,7) problem behaviors, (8) and aggression. (9-11) A social ecology model views patterned behavior of individuals or aggregates as the outcomes of interest. (12-14) In this study the patterned behavior was avoidance of or engagement in interpersonal violence. Behavior was viewed as being both affected by, and affecting, multiple levels of influence. A social ecology model can accommodate these complex, multifaceted variables in the intrapersonal, interpersonal, and physical environments in which adolescents live, go to school, and spend their leisure time. (8-10,12)
From a health promotion viewpoint social ecology addresses the importance of interventions directed at interpersonal, organization, community, and public policy, all factors which support and maintain healthy behavior. Social ecology models assume that appropriate changes in the social environments will produce desirable changes in individuals, and that the support of individuals in the population is essential for implementing environmental changes. (14)
This project verified and extended the Social Development Model (9) and the Social Ecology Mode (15) with the addition of neighborhood aesthetics, neighborhood fighting, and anger control as proposed in the Social Ecology Model of Adolescent Interpersonal Violence Prevention (SEMAIVP). The SEMAIVP (15) was developed to identify variables associated with violence avoidance and violence engagement among adolescents. The model's three independent components include health behavior interventions, population demographics, and the environment (Figure 1). These three components influence an adolescent's avoidance of or engagement in violence.
The health behavior intervention component includes skill development opportunities, involvement in activities, and positive recognition for nonviolence efforts, all components of the Social Development Model, (9) prominent in delinquency theory (16) and in social ecology interventions. (17,18) According to Hawkins and Weis, (9) positive socialization occurs when youth become involved in conforming activities, when they develop skills necessary to be successfully involved in social activities, and when those with whom they interact consistently reward their desired behaviors.
The population demographics component reveals differences in participation in various violent acts as influenced by gender, race, age, and socioeconomic status. Greater incidence of involvement in violence is associated with being Black, (4,19,20) being male, (4,19,20) being a younger adolescent, (1,2,17,19,21) and being in a family with lower parent socioeconomic status. (10,22)
The environment component includes neighborhood, personal, and interpersonal characteristics. The neighborhood environment includes neighborhood aesthetics and neighborhood fighting. The neighborhood where youth live provides the most immediate environment outside the family and shows how life is to be lived. (23) In his ecological research Stark, (24) concluded that poor, dense, mixed-use neighborhoods have high transience rates, resulting in weakened extrafamilial attachments, reduced levels of conformity, and dilapidated buildings. Dilapidation is seen as a social stigma for residents and, in turn, is associated with a reduction in an individual's stake in conformity. (24,25) Previous exposure to neighborhood violence and victimization provides a strong predictor for use of violence among adolescents. Exposure is associated with influencing personal norms regarding the circumstances under which youth perceive use of violence as justified. (22-25)
Personal characteristics include self-efficacy, anger control beliefs, and anger control strategies. Self-control, similar in many ways to self-efficacy, is a basic factor influencing the engagement in delinquency. (5,10,26) Gottfredson and Hirschi (27) believe persons who lack self-control will be impulsive, insensitive, physical (as opposed to mental), risk-taking, short-sighted and nonverbal, and they will engage in delinquent acts. Adolescent anger can be viewed as resulting from lack of knowledge and ability to ask. Rather than being able to demand, negotiate, compromise, or otherwise respond appropriately to conflict, youth strike out physically. (23,26) This component includes beliefs about anger control and the potential strategies for use in responding to anger.
Interpersonal characteristics include family climate, school climate, and peer influence. Family climate represents the basic bond between parent and child essential in the developmental process leading to conformity in society. (14) A negative parent-adolescent relationship represents a risk factor for involvement in fighting. (5,10) School climate establishes norms and standards for behavior. A school climate in which students feel valued, and where norms are established and create an orderly environment, is associated with fewer student-to-student disputes. (5,10,16,28,30) Peer influence becomes more important during adolescence. Association with delinquent peers is often associated with increased engagement in delinquent acts. (10,16,17,29)
PRELIMINARY TESTING OF THE MODEL
The Youth Conflict Mediation Survey (YCMS), compiled by Riner, (15) was used to operationalize the SEMAIVP components. Items for the YCMS came from the Teen Conflict Survey developed by Bosworth, (30) and the Effective School Battery scale developed by Gottfredson. (31) In addition, Riner (15) developed items related to the neighborhood, family, and health education intervention.
The YCMS was pilot tested in one school with 115 students. The responses were submitted for factor analysis which in turn confirmed 14 separate dimensions. Twelve scales met criteria for inclusion into the final survey instrument and analysis. They included 10 dependent and two independent psychometric scales (Table 1). The activity involvement and school bonding scales were deleted due to low reliability.
Health behavior intervention was assessed with two measures. The first measure, skill development opportunity, assessed frequency of a student's participation in anger control groups, peer mediation training or mediation interventions, and being taught how to avoid fights. Higher scores indicate more frequent opportunities. The second measure, recognition for violence avoidance, assessed the frequency students are recognized for walking away from a fight or talking another student out of fighting. Higher scores indicate more frequent recognition by teachers and others.
The neighborhood environment was assessed with three measures. The first measure, neighborhood aesthetics, assessed the degree that the student's neighborhood is perceived as safe, homes are well-kept and it is, overall, a good place to live. Higher scores indicate greater agreement. The second measure, neighborhood fighting, assessed frequency of exposure to being in a fight or watching a friend engage in a fight. Higher scores indicate no involvement or no opportunity.
The personal environment was assessed with three measures. The first, self-efficacy, assessed the confidence of students to avoid engaging in physically violent behaviors. Higher scores indicate a higher level of confidence. The second, anger control belief, assessed the acceptability of using nonphysically violent strategies to resolve conflict such as talking it out, refusing to fight because friends will think they are afraid, being aware of other ways to deal with being angry, and not hitting someone who hits them. Higher scores indicate stronger agreement. The third scale, anger control strategies, focuses on actions versus beliefs. It assessed the likeliness of engaging in nonphysically violent strategies for resolving conflict. Higher scores indicate greater likelihood of using nonphysical aggression.
The interpersonal environment was assessed using three measures. The first, family climate, assessed perceived parent/guardian cooperation with school issues and students' respect for and liking of their parents/guardians. Higher scores indicate greater respect and liking. The second measure, school climate, assessed the student's perception of feeling liked and respected by peers and teachers, fairness of school rules and evenness of punishment for breaking rules, positive perception of learning, and being proud of belonging to the school. Higher scores indicate stronger positive regard for the school climate. The third, peer influence, assessed involvement with antisocial friends, delinquent behavior of friends, and having a serious conflict with a friend. Higher scores indicate less involvement.
The dependent variables were assessed using two measures. The first, violence avoidance behavior (VAB), assessed self-reported active avoidance of, or encouragement from another student to avoid involvement in potentially violent situations. Higher scores represent more frequent avoidance behaviors. The second, violence engagement behavior (VEB), assessed self-reported engagement in interpersonal violence. Higher scores represent more frequent involvement.
FIELD TESTING THE MODEL
Three suburban middle schools in a large, midwestern, metropolitan community participated in field testing between May 1, and December 31, 1997. Each school operated a peer mediation program to train students as mediators and to provide student mediation services. Study approval was received from the Indiana University Institutional Review Board, the school corporation superintendent, and from each participating school's principal. By state law active permission from parents and students was required for students to participate. Nine classes in three schools were selected randomly from the sixth, seventh, and eighth grades for a total of 27 classes. Three hundred and eighteen students returned signed Parent/Guardian Permission and Student Assent forms for a 47% participation rate.
Ninety-seven percent of students were between 11 and 14 years of age, and 53% were female. The sample included African Americans (17%), Asian Americans (5%), Caucasian non-Hispanics (56%), Caucasian Hispanics (3%), Native Indian Americans (4%), and Others (14%), which fairly well reflected the school and community profile. Fifty-two percent of students reported that their fathers and 48% of their mothers completed college. This finding was higher than the community's education profile in which 27% of adults older than age 25 hold a college degree.
Data analysis included descriptive statistics with appropriate frequencies, means, and standard deviations. Factor analysis was used to identify psychometrically similar items to form the scales used in the study. Multiple regression analyses were conducted to assess the usefulness of the models in explaining variance in participants' responses. The Statistical Program for Social Sciences-Personal Computer software was used to analyze the data.
Violence Avoidance Behavior
Regression analysis was conducted for the total sample and for gender and race. The violence avoidance behavior models (Table 2) included 11 psychometric scales and three demographic variables as did the violence engagement behavior equations (Table 3).
The violence avoidance behavior regression analysis for the total sample yielded a significant equation with an R squared of .48 (Table 2). The three statistically significant contributing factors included higher nonviolence self-efficacy beliefs, more frequent positive recognition for violence avoidance, and more frequent use of violence engagement behaviors. Results of the violence engagement behavior equation yielded a significant equation with an R squared of .56 (Table 3). The six significant contributing factors included having more skill development opportunities, more exposure to neighborhood fights, fewer prosocial anger control beliefs, fewer prosocial anger control strategies, lower nonviolence self-efficacy beliefs, and more frequent use of violence avoidance behaviors.
For boys, the VAB model explained 56% and the VEB model explained 60% of the variance (Tables 2 and 3). For the VAB model, recognition for violence avoidance and violence engagement behaviors were significant. The VEB model indicates that more skill development opportunities, more exposure to neighborhood fighting, fewer prosocial anger control beliefs, and more violence avoidance behaviors were significant for boys.
For girls, the VAB model explained 49%, and the VEB model explained 53%, of the variance (Tables 2 and 3). For the VAB model, more recognition for violence avoidance and a less democratic school climate were significant for girls. In the VEB model, having had more skill development opportunities, more exposure to neighborhood fighting, possessing a lower sense of nonviolence self-efficacy, and fewer prosocial anger control beliefs all were significant for girls.
For Black students, the VAB model explained 67% of the variance and the VEB model explained 68% of the variance (Tables 2 and 3). A single factor explained all the variance for Black students for both the VEB and the VAB models. Recognition for violence avoidance was the single predictor for the VAB model for Black students. Possessing a lower sense of nonviolence self-efficacy was the single predictor for the VEB model for Black students.
For White students, the VAB model explained 46% of the variance, and the VEB model explained 51% of the variance (Tables 2 and 3). Three factors contributed to the VAB and to the VEB models with anger control beliefs being the only factor contributing to both models for White students. In addition, for the VAB model, more recognition for violence avoidance, and more violence engagement behaviors, were important for White students. For the VEB model, more exposure to neighborhood fighting and more violence avoidance behaviors were important for White students.
RESULTS FROM FIELD TESTING
Results from the field testing support potential usefulness of the SEMAIVP model in identifying variables associated both with violence avoidance behavior and violence engagement behavior among a suburban adolescent population. The model concepts operationalized from the research and theoretical literature explain a moderate amount of violence engagement behavior and violence avoidance behavior. These findings, in conjunction with those of social ecology theorists (12-14) and adolescent delinquency researchers, (5,6,8-11) suggest that use of social ecology can prove useful in understanding the multiple and complex variables underlying adolescent violence. Use of the SEMAIVP also can focus school administrators, teachers, and staff efforts on identifying multilevel interventions and community agencies with whom to establish partnerships.
Positive recognition for more frequent violence avoidance was the only variable contributing to each violence avoidance model. This finding supports the belief of Hawkins and Weiss (9) that recognition represents a key component in preventing adolescent delinquency. When students are asked to engage in or to avoid specific actions, they need to be recognized and rewarded by teachers and administrators for doing what they were asked to do. Positive reinforcement provides a strong bond with important people in one's life, provides internal satisfaction to an adolescent as being competent and capable of meeting society's valued expectations, and provides feedback to adolescents that they are meeting expectations and conforming to the valued social norm. It gives adolescents an incentive to continue to contribute, and it reinforces their skillful performance. (32)
However, adolescents first must be taught avoidance skills through skill development opportunities. More participation in these opportunities was associated with VEB for the total sample and each gender, but not for any of the VAB models. This positive relationship could reflect actual school practice. When conflict escalates or students engage in a fight, they are often asked or required to participate in anger control programs, attend sessions on how to avoid fighting, or become involved in peer mediation interventions. Future research should examine these earlier efforts, possibly at the school environment and classroom levels, that prevent the need for secondary intervention when conflict escalates, and at the tertiary level when violence occurs. Strategies for helping students develop effective relationship skills at the classroom level include appreciation of diversity, recognition of emotions, and engaging in prosocial responses in conflict situations.
The finding of exposure to more frequent neighborhood fighting as a correlate of engagement in adolescent fighting is supported by previous research. (23-25) As one of the most consistent correlates in this study and over time, creative and comprehensive methods to understand neighborhood violence need to be included in future intervention studies. Community agencies, including neighborhood watch programs and community policing programs, need to be incorporated into adolescent violence prevention programs.
Young adolescents need help in developing prosocial anger control beliefs and strategies as they mature and move into new environments. The fact that four of five VEB equations showed significant correlations between fewer prosocial anger control beliefs and violence engagement behaviors supports the need for schools to continue efforts to include in their curricula tolerance of diversity and developing interpersonal competence in handling conflict. These programs support federal health promotion objectives calling for prevention activities such as peer mediation to reduce adolescent fighting.
Lower self-efficacy was associated with more frequent violence engagement behaviors as reported by others. (10,26) Lower self-efficacy was the single predictor for Black students. While adolescents need to continue building skills in prosocial anger responses, their belief that they can effectively implement them also must mature. A marked difference exists between possessing the knowledge of subskills and being able to use them effectively under adverse circumstances. Perceived self-efficacy is a significant determinant of performance that operates partially independently from underlying skills. (32) Recent social ecology work with alcohol addiction risk factors suggests that culture and environment may prove critical to understanding racial differences in adolescent interpersonal violence as well. Wallace (33) suggests that racial practices and ideologies inherent in American society influence racial/ethnic differences in substance use outcomes. They do so both directly and indirectly through their influence on communities in which people of different racial/ethnic groups are placed, their influence on structure and process of people's interpersonal relationships, and through the impact they exert on individuals' psychology and behavior.
For the total sample, for boys, and for White students, violence avoidance contributed to the VEB model, and violence engagement contributed to the VAB model. These relationships suggest students use both avoidance and engagement behaviors when responding to interpersonal conflict. Are conflicts handled initially with avoidance measures which, if they fail, adolescents resort to violence? One possible explanation comes from social relationship models that suggest adolescent behavior varies as a function of the relationship in which conflict arises. (34) Closeness and relationship stability are characteristics presumed to interact and determine adolescent conflict behavior. In close peer relationships, adolescents seek to minimize the frequency of disagreements, and compromise in those that do arise, so as not to disrupt the relationship. When conflicts arise, less compromise and more negative outcomes are expected among individuals constrained by neither closeness nor kinship. They are more likely to resort to tactics that damage affiliations. In schools where students have fewer opportunities to know each other, engage in activities with one another, and form relationships, more conflict behavior may occur that does greater harm to a relationship than if there are more opportunities for interdependent relationships.
The SEMAIVP encourages schools and youth service agencies to consider both the adolescents and their environments in understanding interpersonal violence. Even if an individual's behavior is the source of a problem, that behavior does not occur in a vacuum. (2) It is shaped and reinforced by multiple aspects of the adolescent's physical and social surroundings. Emphasis on the environment, however, should not suggest that individuals are not actively involved in their own behavior or in modifications to their circumstances. The model assumes that appropriate changes in the social environment will produce changes in individuals, and that the support of individuals in the population is essential for implementing environmental changes. (12)
Finally, some limitations should be noted. Some of the final scales contained as few as two, three, or four items. Data analysis was based on self-report, and thus was limited by issues of recall, lack of multiple sources of data, and adolescent and parental selection to participate in the study. The study design was cross-sectional and descriptive, which was appropriate for initial field testing of a model. To learn more about causality of the risk factors mentioned, the model also must be tested longitudinally. In addition, the sample was drawn from a suburban school district with a socioeconomic level higher than some other school districts. To more generally apply the model, it must be tested among a general youth population.
Though the self-report method produces fairly reliable and valid data about the problem behavior of juveniles, (35) and evidence suggests juvenile perceptions of problems in the social environment play an important role in the development of problem behaviors, (36) a question remains concerning how the self-report from juveniles in this study group would relate to measurements obtained from other sources such as parental reports and observational ratings.
Field testing of the SEMAIVP model revealed its empirical use in understanding key intra-personal, interpersonal, school, and neighborhood factors involved in adolescents' avoidance and engagement in interpersonal violence. This study identified significant social ecology variables in violence avoidance behaviors and violence engagement behaviors among adolescents.
The model extends theoretical and research work undertaken over the past 30 years by developing a social ecology model of adolescent violence prevention. Plans for further development center on scale refinement, additional testing with adolescent populations, and continued specification of relationships among components of the model.
The model has potential use for schools seeking to develop a comprehensive approach to violence prevention. It suggests areas for targeted school-level interventions as well as areas within the school's neighborhood or community for collaborative efforts.
Table 1 Measure of the Social Ecology Model of Adolescent Interpersonal Violence Prevention Possible Number ReliabilityName Range of Items Coefficient Health Behavior Skill development opportunity 1-2 6 .67 Recognition for violence avoidance 1-5 2 .79 Activity involvement 1-2 11 .48 Environment Neighborhood Neighborhood Aesthetics 1-5 4 .80 Neighborhood fighting 1-5 2 .80 Personal Non-violence self-efficacy 1-5 6 .81 Anger control belief 1-5 4 .70 Anger control strategies 1-5 6 .69 School bonding 4 .44 Interpersonal Family climate 1-5 5 .75 School climate 1-5 8 .86 Peer influence 1-4 4 .72 Outcome Variables Violence avoidance behavior 1-5 4 .61 Violence engagement behavior 1-5 3 .63 Table 2 Violence Avoidance Behavior Partial Regression Coefficients and R Square Total B Values Sample Girls Boys Blacks Whites Skill development opportunity 0.11 0.14 -0.02 -0.60 0.14 Recognition 1.02 * 1.05 * 1.07 * 1.40 * 0.99 * Age -0.07 -0.30 0.05 0.50 -0.17 Mother's education 0.12 -0.05 0.35 0.22 0.15 Father's education -0.20 -0.28 -0.15 -0.50 -0.18 Neighborhood aesthetics -0.03 0.03 -0.09 -0.10 -0.05 Neighborhood fighting 0.01 -0.36 0.09 -0.01 -0.12 Self-efficacy 0.11 * 0.12 0.09 0.28 -0.02 Anger control beliefs 0.05 0.00 0.07 -0.04 0.22 * Anger control strategies 0.00 0.10 -0.06 0.02 -0.01 Family climate 0.03 0.17 0.05 -0.01 -0.02 School climate -0.01 -0.12 * 0.07 0.03 -0.01 Peer influence -0.14 -0.27 -0.14 0.11 -0.19 Violence engagement behaviors 0.29 * 0.23 0.35 * 0.36 0.40 * R square .48 * .49 * .56 * .67 * .46 * * indicates significance at [less than or equal to] .05 Table 3 Violence Engagement Behavior Partial Regression Coefficients and R Square Total B Values Sample Girls Boys Blacks Whites Skill development opportunity 0.21 * 0.20 * 0.34 * 0.28 0.16 Recognition -0.19 -0.03 -0.05 -0.07 -0.10 Age -0.03 -0.13 0.00 -0.24 0.16 Mother's education -0.07 -0.02 -0.19 0.12 0.02 Father's education 0.06 0.06 0.00 0.28 -0.03 Neighborhood aesthetics -0.05 -0.07 0.01 -0.08 0.10 Neighborhood fighting .27 * 0.28 * 0.19 * 0.12 0.33 * Self-efficacy -0.12 -0.17 * -0.06 -0.32 * -0.05 Anger control beliefs -0.22 * -0.21 * -0.22 * -0.18 -0.27 * Anger control strategies -0.07 * -0.06 -0.06 -0.05 -0.05 Family climate 0.07 0.13 -0.05 0.00 0.02 School climate 0.01 0.02 -0.01 0.01 -0.01 Peer influence -0.09 -0.04 -0.13 0.13 -0.01 Violence avoidance behaviors 0.14 * 0.10 0.22 * 0.17 0.18 * R square .56 * .53 * .60 * .68 * .51 * * indicates significance at [less than or equal to] .05
(1.) Stokols D. Social ecology and behavioral medicine: implications for training, practice, and policy. Behav Med. 2000;26:129-138.
(2.) Grzywacz JG, Fuqua J. The social ecology of health: leverage points and linkages. Behav Med. 2000;26:101-115.
(3.) Snyder HN, Sickmund J. Juvenile Offenders and Victims: A National Report. Washington, DC: Office of Juvenile Justice and Delinquency Prevention; 1995.
(4.) Brener ND, Simon TR, Krug EG, Lowry R. Recent trends in violence-related behaviors among high school students in the United States. JAMA. 1999;282(5):440-446.
(5.) Kumpfer KL, Turner CW. The social ecology model of adolescent substance abuse: implications for prevention. Int J Addict. 1991;25(4):435-463.
(6.) Perry C, Jessor R. The concept of health promotion and the prevention of adolescent drug abuse. Health Educ Q. 1985;12(2):169-184.
(7.) Perry CL, Williams CL, Forster JL, et al. Project Northland: outcomes of a community-wide alcohol use prevention program during early adolescence. Am J Public Health. 1993;86:956-965.
(8.) Jessor R. Risk behavior in adolescence: a psychosocial framework for understanding and action. J Adolesc Health. 1991;12:597-605.
(9.) Hawkins D, Weis JG. The social development model: an integrated approach to delinquency prevention. J Primary Prev. 1985;6(2):73-97.
(10.) Scholte EM. Prevention and treatment of juvenile problem behavior: a proposal for a socio-ecological approach. J Abnorm Child Psychol. 1992;20(3);247-263.
(11.) Huesmann LR, Maxwelll CD, Eron L, et al. Evaluating a cognitive/ecological program for the prevention of aggression among urban children. Am J Prev Med. 1996;12(5): 120-128.
(12.) Bronfenbrenner U. The Ecology of Human Development. Cambridge, Mass: Harvard University Press; 1979.
(13.) Belsky J. Child maltreatment: an ecological integration. Am Psychol. 1980;35(4):320-335.
(14.) McLeroy KR, Bibeau D, Steckler A, Glanz K. An ecological perspective on health promotion programs. Health Educ Q. 1988;15(4):351-377.
(15.) Riner ME. Social Ecology Model of Adolescent Interpersonal Violence Prevention. Indianapolis, Ind: Indiana University School of Nursing; [unpublished doctoral dissertation]. 1998.
(16.) Hirschi T. Causes of Delinquency. Berkeley, Calif: University of California Press; 1969.
(17.) Thornberry TP. Toward an interactional theory of delinquency. Criminology. 1989;25(4):863-891.
(18.) McBride CM, Curry SJ, Cheadle A, et al. School-level application of a social bonding model to adolescent risk-taking behavior. J Sch Health. 1995;65(2):63-68.
(19.) Hill SC, Drolet JC. School-related violence among high school students in the United States, 1993-1995. J Sch Health. 1999;69(7): 264-272.
(20.) Centers for Disease Control and Prevention. Behaviors related to unintentional and intentional injuries among high school students - United States, 1991. MMWR. 1992;41(6):91 - 93.
(21.) Kelley BT, Huizinga D, Thornberry TP, Loeber R. Epidemiology of Serious Violence. Washington, DC: Office of Justice Programs and Office of Juvenile Justice and Delinquency Prevention; 1997.
(22.) DuRant RH, Cadenhead C, Pendergrast RA, Slavens G, Linder CW. Factors associated with the issue of violence among urban black adolescents. Am J Public Health. 1994;84(4);612-617.
(23.) Goldstein AP, Glick B, Irwi MJ, Pack-McCartney D, Rubama I. Reducing Delinquency: Intervention in the Community. New York, NY: Pergamon Press; 1989.
(24.) Stark R. Deviant places: a theory of the ecology of crime. Criminology. 1987;25(4):893-909.
(25.) Laub J, Hindelang M. Juvenile Criminal Behavior in Urban, Suburban, and Rural Areas. Washington, DC: Office of Juvenile Justice and Delinquency Prevention; 1981.
(26.) Hausman AJ, Spiral H, Prothrow-Stith D, Roeber J. Patterns of teen exposure to a community-based violence prevention project. J Adolesc Health. 1992;13:668-675.
(27.) Gottfredson MR, Hirschi T. A General Theory of Crime. Stanford, Calif: Stanford University Press; 1990:94-97.
(28.) Gottfredson D. An evaluation of an organization development approach to reducing school disorder. Eval Rev. 1987;11(6):739-763.
(29.) Elliott DS, Huizinga D, Ageton SS. Explaining Delinquency and Drug Use. Boulder, Colo: Behavioral Research Institute; Report Number 21, 1982.
(30.) Bosworth K. Teen Conflict Survey. Bloomington, Ind: School of Education, Indiana University; 1995.
(31.) Gottfredson G. The Effective School Battery: User's Manuel. Odesa, Fla: Psychological Assessment Resources; 1991.
(32.) Bandura A. Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall; 1986:391-394.
(33.) Wallace J. The social ecology of addition: race, risk, and resilience. Pediatrics. 1999;103(5):1122-1127.
(34.) Laursen B, Collins WA. Interpersonal conflict during adolescence. Psychol Bull. 1994;115(2):187-209.
(35.) Hindelang MJ, Hirshi T, Weis JG Measuring Delinquency. London: Sage; 1981.
(36.) Jessor R. The perceived environment in psychological theory and research: problem behavior and psycho-social development. In: Magnussen D, ed. Towards a Psychology of Situations. Hillsdale, NJ: Erlbaum; 1981.
Mary E. Riner, DNSc, RN, Assistant Professor, Nursing, Dept. of Environments for Health, School of Nursing, Indiana University, 1111 Middle Drive, NU 459, Indianapolis, IN 46202-5152; (email@example.com); and Robert M. Saywell, Jr., PhD, MPH, Professor and Director of Research, Dept. of Family Medicine, School of Medicine, Indiana University, Long Hospital 247, 1110 West Michigan St., Indianapolis, IN 46202-5152; (firstname.lastname@example.org). This article was submitted March 9, 2001, and revised and accepted for publication August 27, 2001.
|Printer friendly Cite/link Email Feedback|
|Author:||Riner, Mary E.; Saywell, Robert M.|
|Publication:||Journal of School Health|
|Date:||Feb 1, 2002|
|Previous Article:||Fifth through eighth grade longitudinal predictors of tobacco use among a racially diverse cohort: CATCH. (Research Papers).|
|Next Article:||Using path analysis to examine adolescent suicide attempts, life satisfaction, and health risk behavior. (Research Papers).|