Substance abuse, parenting styles, and aggression: an exploratory study of weapon carrying students.

Abstract This study represents one of the first undertaken exclusively with students who brought weapons (other than guns) to school, a group often presenting special difficulty to educators, social workers, and others. The study was performed ancillary to a national study of school violence prevention programs, undertaken by the Hamilton Fish National Institute, but completed at the only site exclusively working with weapons carriers. Appended to the instrument developed for the national study (the National School Crime and Safety Survey), were questions examining measures of student and family relationships, as well as attitudes and behavior among students who had been caught with weapons in school. Subjects (N=37) were students of the Syracuse Public Schools, who had been sent to an alternative school as a result of being caught with a weapon. Substance abusing behavior by adults in the home was common and over one-quarter of respondents indicated having a parent who had been in jail. Family structure was mainly female-headed and family relationships were not reported as overly conflictual. Physical punishment was not common and most respondents reported a good relationship with their parents. Families may be better characterized as inadequate and providing poor models than violent. Family factors most strongly associated with non-aggressive strategies are those which suggest accountability or limit-setting by parents. High levels of parental alcohol and drug abuse, as well as the students' own use, suggest links to poor parenting and intergenerational processes. Findings support the need for substance abuse assessments and family interventions that strengthen disciplinary and protective functions.

Introduction

Often the setting where a problem is presented and studied influences how that problem is defined and where answers are sought. Weapon carrying in schools is one such problem. Since such potentially violent behaviors occur in, and are defined within the educational environment, specific characteristics and skills of students (e.g., anger management and conflict resolution skills) are often emphasized. The popular strategy for students who carry weapons to school is expulsion, despite critics such as Alexander and Curtis (1995) arguing that this strategy can be counterproductive.

Although educators, social workers and others have sought to gain a broader understanding of the factors that are related to students' aggressive behaviors and attitudes, the mechanism of parental influence on children's weapon carrying in school is poorly understood. We need to better understand what motivates children to bring weapons to school, what the links to substance abuse may be, and what influence their family exerts on this behavior.

Teachers' attributions of parental influence are often overly general and sometimes judgmental. Shepard and Rose (1994) found many teachers hold negative and mistrustful attitudes toward the parenting practices of their students' parents. Smith (1990) showed that teachers tended to blame the families of their students for failing to provide stimulation, skills, and resources essential to success in school. The 1993 Metropolitan Life Survey of teachers, found that most teachers felt that lack of parental supervision at home (71%) and lack of family involvement with the school (66%) were primary factors contributing to student violence. According to Kimweli (1997) parents themselves identify a lack of parental involvement as partly to blame for school violence.

Too often, though, studies of students' violence and aggression in school fail to examine the specific dynamics of parental influence and substance abuse that may be associated with these behaviors. Despite calls for action such as are made regularly by the National Education Association for schools to place safety high on their agendas, and involve parents and citizens, we still do not have an adequate understanding of the ways parents can and do impact school safety (Geiger, 1996). We need to improve our understanding of the complex relationships between parents and their children, and how the attitudes, transactions, and roles within family relationships influence the aggressive behavior of youth at school. It is this improved understanding which will support the development and implementation of effective and appropriate methods of involving parents in reducing school violence. This study examines links between selected family factors (e.g. parental substance abuse and parental criminal behavior) and the aggressive behaviors and strategies of students in alternative school placement for weapon carrying. The purpose of this study is to support the development of student and family interventions by adding to the body of research that expands the analysis of indicators of school-violence into a broader, ecological framework.

Review Of The Literature

Recent studies have demonstrated a need to gain a better understanding of the relationship between parents' and children's attitudes and behaviors associated with violence in school. Saner and Ellickson (1996) in their study examining risk factors for adolescent violence found that a lack of parental affection and support was one of the major risk factors associated with youth violence. Trusty (1996) found that parental involvement was positively correlated with positive student attitudes toward school and negatively correlated with problem behaviors such as violence and aggression. Wasserman and Miller (1998) in their review of the literature on prevention programs for violent juvenile offenders wrote "Clearly, programs that ignore the family in strict reliance on school-based interventions ... are unlikely to show powerful effects." (p.247). In their comprehensive review of predictors of youth violence Hawkins et al. (1998) found a number of factors to be significantly correlated with youth violence. These include: fathers criminality (Baker & Mednick, 1984; Farrington, 1989; Maguin et al., 1995; Moffitt, 1987); child maltreatment (Zingraff, Leiter, Myers, & Johnson, 1993) poor family management (Farrington, 1989; Maguin et al., 1995); family and marital conflict (Farrington, 1989; Maguin et al., 1995); parental attitudes favorable to violence (Maguin et al., 1995); residential mobility (Maguin et al., 1995); and separation from home and early home leaving (Farrington, 1989).

A variety of national surveys have reported on the frequency with which students are involved with violence and weapons at school. The University of Michigan "Monitoring the Future Survey" (1996) reported that 7% of male students and 3% of female students had been injured by a weapon at school. Friday (1996) reported that since 1990 the Centers for Disease Control and the National Education Goals Report have indicated rates of in-school weapon carrying ranging from 5%-36%. Consistently, students state that the reason they carry weapons to school and in school is for protection (Louis Harris and Associates, 1993; Friday, 1996). Friday (1996) in a review of the literature on students who carried weapons to school wrote "violence and weapon carrying in schools also reflect the personal attitudes of students and their families." (p.21). Forty percent of parents in the Harris survey (1993) admitted that they would tell their children to defend themselves if attacked--even if it meant using a weapon.

Although there is a general and well-documented statistical correlation between drug or alcohol use and violence among adolescents (Corvo, 1997), the nature of the relationship is more complex than it appears. Elliot's (1994) path model analysis of data from the National Youth Survey found that there was a noticeable sequence of progression toward violent behavior involving marijuana and alcohol use, though these were not identified as causal factors. Rather, substance abuse may share a common etiology with violent behavior. Drug use, in particular, has a strong association with violent juvenile crime (Office of Juvenile Justice and Delinquency Prevention 1999). Kumpfer, Olds, Alexander, Zucker, and Gary (1998) found that substance abuse and anti-social behavior may both be influenced by an interaction of genetic and family socialization factor. The National Institute on Alcohol Abuse and Alcoholism (1997) has concluded that alcohol use may increase the risk for violence only for certain individuals or sub-populations and only under certain situations or social/cultural influences. Specific links between substance abuse and violence in general are often difficult to pin down. Other than as a set of general risk factors, the mechanism of how substance abuse (student or parental) influences weapon carrying behavior, in particular, is even less clear.

The National Center for Education Statistics (NCES) (1998) in their report "Violence and Discipline Problems in U.S. Public Schools: 1996-1997" summarized survey results on the extent of school violence during that school year. Public school principals reported the number of violent incidents that occurred at their schools during the 1996-97 school year. Respondents were asked to report only those incidents for which the police or other law enforcement officials had been contacted. According to these data during 1996-97 there were about 11,000 incidents of physical attacks or fights in which weapons were used. An examination of in-school murders estimated that there were 105 school-associated violent deaths including 85 murders occurring at schools during a 2-year period from 1992 to 1994 (Kachur, 1996).

Schools have taken a variety of steps to reduce the number of weapons brought to school and the subsequent number of physical attacks and murders involving weapons. The US Department of Education (NCES, 1998) study, "Indicators of School Crime and Safety," asked respondents to indicate what disciplinary actions were taken against students for possessing weapons. Five percent (a total of 16,587 actions) of all schools reported taking one or more of these three actions against students (expulsion, transfer to alternative schools or programs, or out-of-school suspensions) for the possession or use of a firearm. Nearly half reported out-of-school suspensions. Twenty percent of students from the schools surveyed were transferred to alternative schools or programs. Thirty one percent of students were expelled for the possession or use of a firearm.

Twenty-two percent of public schools also took one or more disciplinary actions against students for possession or use of a weapon other than a firearm. About 58,000 such disciplinary actions were reported. Twenty-three percent were expelled; twenty-two percent were transferred to alternative programs or schools and fifty-five percent were suspended. Clearly the use of alternative schools and programs for students caught with weapons has become a popular option for educators.

We have a general understanding that family factors are associated with youth violence and aggression, but how do we identify specific factors associated with weapon carrying as objectives for interventions? In other words, what are the links between family factors and student attitudes and behaviors that lead to weapon carrying and alternative school referrals?

Methods

This study was performed as an independent adjunct to a national multi-site evaluation study of school violence prevention programs undertaken by the Hamilton Fish National Institute on School and Community Violence (Hamilton Fish Institute) at George Washington University. The Hamilton Fish Institute (funded by the Office of Juvenile Justice and Delinquency Prevention) subcontracted with six other universities nationwide to study the effectiveness of violence prevention methods and to develop more effective strategies. Syracuse University was one of the participating universities and an alternative school in the Syracuse Public Schools was one of the programs being evaluated. The Syracuse site was the only one working with students who had been found with weapons in school. Therefore, the authors used only these local data to gain a better understanding of the role of family and personal characteristics among weapon carriers.

Subjects were students (N=37) of the Syracuse, New York Public Schools who had been sent to an alternative school as a result of being caught with a weapon (most commonly box-cutters, knives, and bb guns) in school. Students caught with guns were sent to juvenile detention and not referred to the alternative school. Subjects completed the National School Crime and Safety Survey (NSCS) developed by the Hamilton Fish Institute. Enrollment in the school at the time the survey was administered was 68. The available sample represents those who were in attendance on the days the survey was administered (the average daily attendance rate at the school was 60%). The surveys were administered during October of 1998 during classes. Students were told that their answers would be confidential and parental consent was obtained.

Appended to the instrument developed for the national study were questions developed by the authors examining family relationships, student and familial substance abuse, and additional measures of student and family anti-social attitudes and behavior. The purpose of this study was to investigate if these added variables would expand the analysis of indicators of weapons-carrying into a broader, ecological set of associations. Although the Hamilton Fish Institute study contained several measures of familial influence, its primary focus was on school environment and student behavioral issues.

Measures

Measures concerning aggression, aggressive or non-aggressive strategies, victimization, and willingness to fight were derived from the NSCS questionnaire. In addition to the questions in the NSCS, students were asked about their relationships with their parents (e.g., "I generally have a good relatioship with my parents") and their familial context (drug use in the home, parental incarceration, eating dinner with family, telling parents where s/he goes on weekends and after school, parental attitudes toward violence and victimization) and their own use of alcohol and drugs.

Data Analysis

The analysis of the data involved exploring patterns of association among the data by assessing relevant bivariate correlations and by employing independent sample t-tests to identify significant differences between nominal level variables on key behavioral measures. Constrained by the structure of data collection required by the NSCS, data analysis was essentially secondary, though made somewhat more flexible through the addition of supplemental variables. Analysis required extracting and modifying variables associated with aggression or aggressive strategies from the NSCS data. These were primarily selected scale items from the NSCS questionnaire examining student behavior,, attitudes and beliefs regarding violence. The broader task of refining the scales and assessing overall reliability has been undertaken by the Hamilton Fish Institute. The relationship between various added independent variables was then tested as to its association with the indicators of aggression.

Findings

Characteristics of the Sample

Fourteen males (38%) and Twenty-three females (62%) completed the survey. Twenty-five (66%) were African American, three (8%) were White, two (6%) were Hispanic, one (3%) was Asian, one (3%) was Native American, and five (14%) described themselves as multi-racial or other. Students ranged in age from 12 to 19 years with the mean age of 14.6 years. Participating students were from the 7th through 12th grades with the median grade being the 9th. All students were eligible for free or reduced lunch--indicating that they met the federal and state income criteria for living below the poverty line. About a quarter (26%) of respondents reported "regular" attendance at church, temple, or mosque.

Fifty-six percent of respondents reported that they had used alcohol; 14% reported drinking at least once per week. Fifty-one percent reported having used marijuana; 17% reported using it at least once per week. Three percent reported using other illegal drugs. Lifetime prevalence use of alcohol for this sample is within the range for the national sample surveyed for the Monitoring the Future (2000) study, but lifetime prevalence use for marijuana is higher (i.e. 8th-10th grade use rates of 20-40%).

Family Characteristics and Relationships

Seventy-nine percent of students reported living with their mother or stepmother. Eighteen percent reported living with their father or stepfather. Five percent reported living with foster parents. About two-thirds responded that they had a generally good relationship with their parents. Fifty-two percent of respondents indicated that they usually ate dinner with their family, while only 40% reported telling their parents about their after school and weekend activities. Twenty-two percent reported having seen their father or stepfather drunk. Twenty-eight percent reported having seen their mother or stepmothers drunk. Thirty-nine percent reported having seen other adults in their home drunk. Eleven percent had see their mothers or stepmothers use illegal drugs. Three percent reported seeing their fathers or stepfathers use illegal drugs. Twenty-six percent of students reported being physically punished by their parents or guardians. Twenty-six percent of respondents reported that they had a parent or guardian who had been in jail. Eighty-four percent of respondents expressed a measure of agreement that their family would defend them if they got into a fight. Only 31% indicated some measure of agreement that they do not fight because their parents would not like it. When asked if they knew whether their parents or guardians thought they should get into trouble if they behaved violently at school, only 34% said "yes", while 19% said "no", the largest response category (44%) was "don't know".

Bivariate Correlations

Bivariate correlations of the previous variables with various measures endorsing aggression (scales consisting of items: 1. indicating which acts of provocation justify fighting; 2. belief in the effectiveness of aggression; and 3. skills in avoiding violence) that did not demonstrate statistical significance

Those categorical family variables which distinguished mean values for aggressive strategies are shown in Table 2.

of at least p=.05 were excluded form further analysis. Of the bivariate relationships among the variables examined, the following reached statistical significance levels of at least p=.05. Correlation coefficients are shown in Table 1.

Sample characteristics in this study describe a young, largely minority, low income, predominantly female, group with regular patterns of alcohol and marijuana use. The over-representation of females in this sample was due to poorer attendance and a higher dropout rates for males (the actual sex ratio of referred students was about 50/50). A post-hoc analysis of indicators of aggressive behaviors and attitudes did not reveal any significant differences by sex of surveyed students. Nationally, adolescent males are about 5-7 times more likely than females to carry a firearm and about 3 times more likely to carry other types of weapons (Hamilton Fish Institute, 2001). It would appear then that although males are more likely to be weapon carriers, factors associated with aggressive behaviors and attitudes may be similar regardless of sex among those already identified as weapon carriers.

Family structure was mainly female-headed and family relationships were not reported to be overly conflictual. Physical punishment was not common with most respondents reporting a good relationship with their parents. However, substance-abusing behavior by adults in the home was common and over one-quarter of respondents indicated that they had a parent or guardian who had been in jail. Their families might be better characterized as inadequate ad providing poor models rather than as conflictual and violent.

In this study, the family factors most strongly associated with non-aggressive strategies or avoiding violence were those that suggested accountability or limit-setting by parents. For example, knowing that one's parents thought you should get in trouble for behaving violently in school was associated with reluctance to fight and lower endorsement of aggressive strategies. Also important was a sense of confidence that one's family would defend one being associated with lower endorsement of aggressive strategies. Fewer individual factors (e.g., number of victimization experiences in the past 30 days reducing one's willingness to fight) were identified as being associated with aggression.

This sample of alternative school students represented those who had been identified as bringing non-firearm weapons to school. Given that the incidence of such behavior is likely to be greater than its formal identification, we can make only modest claims for generalizability for those students who have not been formally identified as weapon carriers. Also we suspect that the characteristics of firearms-carrying students would be different from this sample. However, it is reasonable to conclude that this sample represents a population that is predisposed to some moderate level of aggressive behavior. The data suggests that this sample of at-risk students is responding to an environment where parental authority is weak, models for substance-abusing behavior are common, and aggression is viewed more as instrumental to self-protection, than as a means to support other delinquent behaviors. Since this sample only included weapon carriers, respondent and family characteristics cannot be used to differentiate weapon carriers from other students. However, other efforts to profile or distinguish weapon carriers from others, based on student characteristics, have proven to be elusive. Furlong, Bates, and Smith (2001) found that a risk index of student factors (e.g. use of alcohol and drugs; fighting) produced a greater than 10:1 ratio of false to accurate predictions of weapon carrying among a large national sample.

The present study had a small sample and data analysis was exploratory. As such, findings need to be interpreted cautiously. Further, compression of scores due to the sample being drawn from a population already identified as violent may obscure other correlations perhaps observable in samples with greater range.

Implications For Practice

How do these findings contribute to improved approaches to dealing with school violence? As educators, counselors, and social workers working with youth at risk of violence in schools we need to understand the range of factors that influence young people's attitudes toward violence--particularly those youth who have brought weapons to school. From a practical standpoint, trying to identify factors which may predict weapon carrying is of limited value (Furlong, Bates, & Smith, 2001). Therefore, from a secondary prevention perspective, efforts may be better directed toward reducing risk among those already identified as weapon carriers.

As we have described, parental influences were significantly related to aggressive strategies and a willingness to be violent. Youth may be less likely to behave violently or fight if they believe that they will be held accountable by their parents or guardians for this behavior. These data, suggest a link between perceived parental concern, vulnerability, and subsequent aggression.

Although not significant as correlates of students' endorsement of aggression, the high levels of parental alcohol and drug abuse (as well as the students' own use) must be a concern for those working with these families. The role of parental substance abuse in the lives of young people at risk of violence requires further investigation. Of particular importance may be how parental substance abuse interferes with effective parenting and pro-social role modeling.

It is essential that educators and others working with at-risk students begin to think outside the box of viewing students only as students and not as members of communities and families that have great impact on behaviors. Our findings suggest that we need to better understand what motivates weapon carriers and not automatically isolate them through expulsion or separation from other students. We need to know what will work with this population instead of using a one-size-fits-all approaches. Weapon carriers and their families provide a unique challenge to schools and this study represents the first step to an improved understanding of what motivates them. School social workers often understand the importance of families when working with aggressive youth, and by better understanding what motivates children to carry weapons, and the possible role of substance abuse, they will be better able to work with these children and their families. Educators and social workers need to work together with these students and their families, making referrals where appropriate for substance abuse and family counseling. Social workers can help parents communicate with their children about their behaviors and alternatives to violence. The limit-setting and protective roles of parents can be strengthened through a variety of intervention strategies. Tolan and Guerra (1994) identified several forms of family intervention, including structural and multi systemic family therapy, that have demonstrated effectiveness in changing the violent behavior of youth. Corvo's (1997) evaluative review of various approaches to youth violence prevention and treatment concluded that, "Planners and funders should consider sound, well-implemented family interventions as one of the more robust means of addressing youth violence." (p.302).

This study makes a modest empirical contribution to an understanding of the role of substance abuse and family influences on weapon carriers, supporting the view that approaches which strengthen parents' authoritative and protective roles may hold promise for reducing the weapon carrying behavior of their children.

Table 1: Correlation Coefficients

                              Willingness to fight

Strength of belief in the effectiveness of       -.43 *
  non-aggressive strategies

# of victimization experiences in past 30 days   -.49 *

                  Strength of belief in aggressive strategies

Strength of belief that your family will
  defend you during a fight                      -.37 *

                 Ability to avoid someone who wants to fight

Do you tell your parents what you do outside
  of school and on weekends?                     .46 ***

* Pearson's r ** Speraman's rho *** Phi coefficient

Table 2: t--Tests
Do Your Parents Think You Should Get Into Trouble For Behaving
Violently at School?

                                   YES   Don't Know (1)

Strength of Belief in Aggressive   3.4        5           p=.04
Strategies*

Reluctance to Fight**              41        22           p=.03

* Scale scores range from 0-10 ** Scale scores range from 14-70

(1) In an ANOVA the small number of "no" responses eliminated
the overall statistical significance of the difference among
means. A post-hoc test using both Tukey's and LSD identified
differences between "yes" and "don't know" which were then
re-computed and are reported here.

References

Alexander, R., & Curtis, C.M. (1995). A Critical Review of Strategies to Reduce School Violence. Social Work in Education, 17(2), 73-82.

Baker, R.L.A., & Mednick, B.R. (1984). Influences on human development: A longitudinal persective. Boston: Kluwer-Nijhoff.

Corvo, K.N. (1997). Community-based youth violence prevention: A framework for planners and funders. Youth and Society, 28(3), 291-317.

Elliott, D.S. (1994). Serious violent offenders: Onset, developmental course, and termination: The American Society of Criminology 1993 Presidential Address. Criminology, 32(1), 1-21.

Farrington, D.P. (1989). Early predictors of adolescent aggression and adult violence. Violence and Victims, 4, 79-100.

Friday, J. (1996). Weapon carrying in schools. In Hoffman (ed.). Schools, Violence and Society. Connecticut: Praeger.

Furlong, M.J., Bates, M.P., and Smith, D.C. (2001). Predicting school weapon possession: A secondary analysis of the youth risk behavior surveillance survey. Psychology and the school, 38(2), 127-139.

Geiger, K. (1996). NEA's perspective and policies on violence in schools. In Hoffman (ed.). Schools, Violence and Society. Connecticut: Praeger.

Hamiltion Fish Institute (2001). http://www.hamfis.org/statistics/fedrep.php3

Hawkins, J.D., Herrenkohl, T., Farrington, D.P., Brewer, D., Catalano, R.F., & Harachi, T. (1998). In Loeber and Farrington (eds.). A review of predictors of youth violence. Serious and Violent Juvenile Offenders: Risk Factors and Successful Interventions. California: Sage Publications.

Kachur, S.P. (1996). School Associated Violent Deaths in the United States, 1992 to 1994, Journal of the American Medical Association, June 12, 1996: 275(22): 1729-2733.

Kimweli, D. (1997). Parents' versus students' perception of predictors of violence and substance abuse in schools: Psychological and contextual factors, Paper Presented at the Annual Meeting of the American Psychological Association (105th, Chicago, IL, August 15-19).

Kumpfer, K.L., Olds, D.L., Alexander, J.F., Zucker, R.A., and Gary, L.E. (1998). Family Etiology of Youth Problems. In NIDA Research Monograph 177, USDHHS, NIH, Rockville, MD.

Louis Hams and Associates. (1993). A Survey of the American People on Guns as a Children's Health Issue. Boston: Harvard School of Public Health.

Maguin, E., Hawkins, J.D., Catalano, R.F., Hill, K., Abbott, R., & Herrenkohl, T. (1995, November). Risk factors measured at three ages for violence at age 17-18. Paper presented at the American Society of Criminology, Boston.

Metropolitan Life Insurance Company (1993). Violence in America's Public Schools. New York: Metropolitan Life Survey of the American Teacher 35-38.

Moffitt, T.E., (1987). Parental mental disorder and offspring criminal behavior: An adoption study. Psychiatry, 50, 346-360.

Monitoring the Future (2000). www.monitoringthefuture.org

National Center for Education Statistics (1998). Indicators of School Crime and Safety. http://nces.ed.gov/pubs98/safety/index.html.

National Institute on Alcohol Abuse and Alcoholism (1997). Alcohol, Violence, and Aggression. www.niaaa.gov/publications/aa38htm.

National School Crime and Safety Survey (1998). Hamilton Fish National Institute on School and Community Violence. www.hamfish.org.

Office of Juvenile Justice and Delinquency Prevention (1999). Juvenile Offenders and Victims. Washington, D.C. U.S. Government Printing Office.

Saner, H. & Ellickson, P. (1996). Concurrent risk factors for adolescent violence. Journal of Adolescent Health, 19(2), 94-103.

Shepard, R. & Rose, H. (1994). The power of parents: An empowerment model for increasing parental involvement. Education, 115(3), 373-377.

Smith, D. (1990). Women's work as mothers: A new look at the relation of class, family, and school achievement. In Forman, O'Brian, Haddad, Hamilton, and Mater (eds) Feminism and Education. Toronto, OISE Press.

Tolan, P.H., & Guerra, N.G. (1994). What works in reducing adolescent violence: An empirical review of the field. Monograph prepared for the Center for the Study and Prevention of Youth Violence, Boulder, CO.

Trusty, J. (1996). Relationship of parental involvement in teens' career development to trends, attitudes, perceptions, and behavior. Journal of Research and Developement in Educaiton, 30(1), 63-69.

University of Michigan (1996) Monitoring the Future Survey. http://monitoringthefuture.org/

Wasserman, G.A., & Miller, L.S. (1998). The prevention of serious and violent juvenile offending. In Loeber and Farrington (eds.) Serious and Violent Juvenile Offenders: Risk Factors and Successful Interventions. California: Sage Publications.

Zingraff, M.T., Leiter, J., Myers, K.A., & Johnson, M.C. (1993). Child maltreatment and youthful problem behavior. Criminology, 31, 173-202.

JADE's website: www.unomaha.edu/~healthed/JADE.html

KENNETH CORVO
KIMBERLY WILLIAMS
Syracuse University

Kenneth Corvo, Please send all correspondence to first author at: Syracuse University, School of Social Work Syracuse, New York 13244 e-mail: [e-mail]