School Connectedness, Anger Behaviors, and Relationships of Violent and Nonviolent American Youth

Thomas, Sandra P

PROBLEM. Youth violence research often focuses on risk factors arising from early familial interactions rather than school-related factors.

METHODS. Via an Internet questionnaire, 282 girls and boys (ages 7-19, mean 15.3) from 47 states and Washington, DC, reported on school connectedness, interpersonal relationships, and anger behaviors.

FINDINGS. Substantial percentages of violent youth did not perceive themselves to be liked by classmates and reported loneliness. If not liked by classmates, 80% hated school. Likers and haters of school differed on seven variables (all p≤.01).

CONCLUSIONS. Insufficient attention is paid to the alienation experienced by disliked and lonely students. Mental health nurses could play a pivotal role in fostering change in the social climate of schools and helping youth to achieve better anger management and social skills.

Search terms: Anger behaviors, school connectedness, youth violence

Youth violence is a serious problem in America, commanding attention from school officials, behavioral science researchers, and juvenile justice personnel. Although recent reports suggest some decline in juvenile arrests for Violent Crime Index (VCI) offenses, Snyder and Sickmund (1999) estimate that fewer than half of violent crimes by young people are ever reported to the police. Even if arrests for VCI offenses such as aggravated assault and murder are decreasing, studies show U.S. school children remain heavily involved in fighting, weapon carrying, and bullying (which can include both psychological and physical abuse). In the national priorities outlined in Healthy People 2010 (U.S. Department of Health and Human Services, 2000), reduction of adolescent physical fighting and weapon carrying to school are specific objectives.

Out-of-control anger behavior appears to be rampant among youth, perhaps echoing the behavior modeled by adults who engage in road rage, air rage, and desk rage. A recent survey of more than 15,000 teenagers (Josephson Institute of Ethics, 2001) showed that 75% of boys and more than 60% of girls had hit someone in the past 12 months because they were angry. The high frequency of bullying behavior has been documented in several studies. More than 1 in 6 sixth- to tenth-graders say they are bullied sometimes; more than 1 in 12 say they are bullied once per week or more (Nansel et al., 2001). In an online poll conducted by Time for Kids, 27% of the respondents admitted bullying others, and 41% said they had been picked on (Labi, 2001). Middle-school students interviewed by Horowitz et al. (2004) reported being teased and bullied about their physical appearance, personality and behavior, family and environment, and school-related factors such as academic ability. "Being different in any way" was the underlying theme in focus-group discussions with these students.

Teasing and bullying escalated when students were highly sensitive, cried, or acted "odd" (Horowitz et al., 2004). The bystanding audience often supports bullying behavior, as shown in a study of more than 10,000 third- to ninth-graders (Twemlow, Fonagy, & Sacco, 2001); 10% to 20% experienced a vicarious thrill from watching other students being bullied.

Weapons are prevalent on school property despite metal detectors and other security measures. National data from the Centers for Disease Control and Prevention (CDC, 1999) showed that more than 1 in 13 students were threatened or injured with a weapon (such as a gun, knife, or club) on school property in the past year. Josephson Institute data (2001) revealed that more than 1 in 5 (21%) high-school boys and 15% of middle-school boys took a weapon to school at least once during the past year. Nearly 1 in 3 middle-school boys (31%) and 60% of high-school boys said they could get a gun if they wanted to. Highly publicized school shootings, such as those in Colorado, Mississippi, and California, have created pervasive fear (Erwin, 2002).

This article examines selected research literature on youth violence, with an emphasis on school-related factors, and presents the findings of a descriptive study of violent and nonviolent American youth.

Risk Factors for Youth Violence

While the etiology of youth violence is complex, many risk factors are well known, including family factors such as parental criminality, child maltreatment, and low levels of parental involvement (Hawkins et al., 2000). The predictive power of factors such as poverty, residence in a violent community, and neighborhood disorganization is also well established (Hawkins et al.). But the massacre at Columbine, perpetrated by boys from an affluent community, defied explanation in terms of these established risk factors, many of which were absent from the profile of the Columbine shooters. Instead, school-related factors appeared more salient. The school shooters were social outcasts who had experienced bullying and other forms of cruel treatment from classmates. Ultimately, it was the school that bore the brunt of their rage. After Columbine, the National Threat Assessment Center, run by the secret service, discovered that in more than two-thirds of 37 recent school shootings, the attackers felt "persecuted, bullied, threatened, attacked or injured" (Labi, 2001, p. 46).

Researchers began to focus on school connectedness (also called school bonding) as an important variable in reducing risk for violent behavior. Definitions of the variable usually emphasize students' experience of caring at school and a sense of closeness to school personnel and environment (Resnick et al., 1997). Attitude toward school was an important aspect of recent focus-group discussions conducted by Erwin about the social situations and problems young people face. The sample included youth in treatment for behavioral problems as well as youth who were not. Notes Erwin (2002), "The institution of school was a symbol with many meanings, but the basic attitude of liking or not liking it was the pivotal factor. . . . There was a high degree of emotion present when discussing negative experiences, including sadness, anger, and anxiety" (pp. 29-30).

Researchers have found a direct relationship between school disconnectedness and outcomes such as delinquency, truancy, drug use, and a number of physical and mental health indicators (Bonny, Britto, Klosterman, Hornung, & Slap, 2000). Students who feel close to others, fairly treated, and vested in school are less likely to engage in risky behaviors than those who do not (Resnick et al., 1997). Harsh disciplinary policies (e.g., expulsion for a first offense) produce lower school connectedness (McNeely, Nonnemaker, & Blum, 2002). Greater connectedness is promoted by a climate of positive classroom management, smaller school size, and higher student participation in extracurricular activities (McNeely et al.) This line of research is important because school connectedness is a modifiable factor: Poverty, community disorganization, and family neglect are not.

The present study contributes to the literature on school connectedness by (a) identifying differences between violent and nonviolent youth on this variable, and (b) comparing "likers" and "haters" of school on selected characteristics such as anger behaviors and interpersonal relationships. Data were collected via an Internet survey, based on prior research showing that self-report data collected from youth under conditions of anonymity are reliable (Brener, Collins, Kann, Warren, & Williams, 1995). This method enabled recruitment of a geographically diverse national sample. Specific aims of the study were to (a) extend our previous research that focused only on girls (Smith & Thomas, 2000); (b) examine differences between violent and nonviolent youth; (c) delineate differences between "likers" and "haters" of school; and (d) examine gender and racial differences in anger behaviors, school connectedness, and interpersonal relationships with classmates and family.

Related Literature on Gender and Racial Differences in Anger and School Connectedness

Some gender and racial differences in anger have been discovered in previous research. For example, boys and girls are aroused to anger by different sorts of provocations. To a female, the most anger-provoking behavior of another person is an accusation of being promiscuous; to a male, it is the accusation of being cowardly (Harris, 1996). Girls and boys studied by Stapley and Haviland (1989) differed in the causes of anger as well as the direction of their anger expression. Girls were angry because of interpersonal experiences, and directed their anger inwardly; boys were angry in situations in which their performance was evaluated, and directed the anger outwardly. Studies consistently show that boys are more likely than girls to respond to provocations by engaging in overt physical aggression, while girls tend to engage in covert relational aggression (e.g., gossip, ostracism by the clique) (Crick, Bigbee, & Howes, 1996; Crick, Grotpeter, & Bigbee, 2002; Deffenbacher & Swaim, 1999; Simmons, 2002). These differences are attributable to traditional gender role socialization in which boys are encouraged to be tough and physical, whereas girls are encouraged to be passive and indirect (Thomas, 1993).

While there has been speculation that gender role socialization is changing toward promulgation of an androgynous ideal, new studies do not demonstrate any weakening of traditional notions of masculine and feminine anger behavior. In a recent study of adolescent males conducted by Lopez and Emmer (2002), both selfpreservation crimes and vigilante crimes (i.e., violent acts on behalf of others, such as a brother) were fueled by beliefs about "manly" behavior. Likewise, adolescent girls interviewed by Simmons (2002) felt constrained from direct verbal expression of their anger by cultural injunctions regarding "fernininity" (and by fear of relationship termination). However, girls from working-class or minority communities were less constrained when angry and sometimes expressed pride in their ability to defend themselves in physical fights, pointing to the importance of community norms that vary according to socioeconomic status and ethnicity/race.

Ethnic/racial differences in anger and violent behavior of young people deserve closer attention, because studies are few. Unfortunately, Hawkins et al. (2000), in an ambitious meta-analysis of 66 studies of youth violence predictors, excluded race and gender from consideration. Deffenbacher and Swaim (1999) found Mexican-American middle- and high-school students less likely than whites to be verbally aggressive when angry. Black elementary students, whether male or female, had greater general propensity to be angry (trait anger) and higher anger-out scores than whites in a study by Hauber, Rice, Howell, and Carmon (1998). Whites scored higher on anger reflection/control (resolving conflict via a cognitive approach). In an adolescent sample of African, Hispanic, and European Americans, trait anger was higher in African Americans (Reyes, Meininger, liehr, Chan, & Mueller, 2003). Consistent with the findings of Deffenbacher and Swaim, Hispanic students scored lower on outwardly expressed anger.

Little information is available to date about school connectedness in students with different racial characteristics. It is reasonable to hypothesize, however, that minority students may feel less affinity for school. There is growing evidence that black students are more severely disciplined than whites and 2.4 times more likely to be suspended (Morse, 2002). Furthermore, black students are often disciplined for "nebulous infractions" such as being too noisy (Morse, p. 51). Black race was associated with lower school connectedness in a study of 1,959 seventh- to twelfth-graders at eight public schools (Bonny et al., 2000). Blacks (and females) felt less connected to school in an analysis conducted by McNeely et al. (2002) using the National Longitudinal Study of Adolescent Health data on more than 75,000 students in 127 schools.

Methods

Instruments

The survey instrument had been developed by the second author (Smith, 2000) for a previous Internet study assessing factors salient to youth violence. Items were based on literature review and clinical experience of the second author, a forensic psychologist. The resultant data from that survey were used for an in-depth secondary analysis focusing exclusively on the female respondents (Smith & Thomas, 2000). The survey tool was slightly modified for the present study by deleting some items about television-watching habits and adding a question about race (an inadvertent omission from the first study). Five demographic items assessed age, race, state of residence, grade in school, and sex. Respondents also were queried about juvenile justice charges (if any), and bringing weapons to school. School connectedness was assessed by the question "How do you feel about school?" Relationships with classmates were assessed by the question, "Do your classmates like you?" Relationships with family members were assessed by the question, "How do you get along with your family?" Because fair treatment is an important element of school connectedness (see Bonny et al, 2000), students were questioned about their perceptions of the fairness of school discipline. Other questions ascertained perceptions of home discipline and general perceptions of adult fairness toward kids. Respondents were also questioned about feelings of loneliness or sadness, and what they did when feeling lonely or sad. Anger behaviors were assessed by the Framingham Anger Scales (FAS), which were embedded within the questionnaire, along with an item asking respondents if they had ever felt angry enough to hit, and if so, what produced the anger. The opportunity to write a short narrative was provided. Room for writein responses was provided after several other questions and at the end of the survey. For example, if respondents believed they were disliked by classmates, they could write in an explanation of why this was so.

The Framingham Anger Scales were selected for inclusion in the survey because they include both healthy and unhealthy anger expression styles and the participant response burden is low. The FAS subscales assess four modes of anger expression: anger-in (suppression of angry feelings); anger-out (anger taken out on others in an attacking or blaming way); anger-symptoms (intense physical symptoms experienced during anger arousal, such as headache or shakiness); and anger-discussion (the healthier choice of talking about the anger with another person, such as a friend or relative). The FAS were developed for the well-known prospective study of heart disease risk in Framingham, MA, and published in an article about the instrumentation for that study (Haynes, Levine, Scotch, Feinleib, & Kannel, 1978). Using a 3-point response format, participants indicate how likely they are to behave in each of the specified ways when they are "really angry or annoyed." Items for the FAS test battery were selected from a 300-item pool by expert judges and evaluated by both item analysis and factor analysis (Haynes et al.). Internal consistency reliability (ascertained by Cronbach's alpha) of the FAS was acceptable in several large studies of young people and adults previously conducted by the first author and colleagues (Ausbrooks, Thomas, & Williams, 1995; Thomas, 1993).

In a study of African-American youth (Armstead & Clark, 2002) who were given three of the FAS subscales (anger-in, anger-out, and anger-discussion), factor analysis confirmed component structures that were virtually identical to those proposed by the tool developers (Haynes et al., 1978). The study also included comparative assessment of the psychometric properties of the FAS and the Spielberger Anger Expression Scale, and the researchers concluded that the anger-in and anger-out subscales of the FAS assessed these two dimensions of anger expression more accurately than did the Spielberger scale (Armstead & Clark). The anger-discussion subscale could not be compared to other instruments because FAS is the only tool that queries respondents about this healthy way of dealing with anger.

Procedure

Internet surveys have become commonplace because of their speed, convenience, and low cost. Such surveys can be completed at respondents' convenience, often in the privacy of their homes. More than 60% of American homes now have computers, and more than 50% have Internet access (U.S. Department of Commerce, 2002). Shaefer and Dillman (1998) found that e-mail surveys achieved response rates similar to mail surveys but produced better quality data with regard to item completion and more detailed narratives in response to open-ended questions. As noted by Fawcett and Buhle (1995), some people may find the anonymity of electronic data collection liberating. Studies consistently show that people tend to reveal more personal and/or potentially embarrassing information with a computer-administered interview than a face-to-face interaction (Joinson, 1998). For example, many of the girls surveyed by Simmons (2002) first disclosed their humiliating victimization by school bullies via e-mail communications. Internet data collection may be especially well suited to youth, who have become accustomed, through school assignments, to "surfing the net" and spend many after-school hours on the computer as well. A survey by the Kaiser Family Foundation (Jesdanun, 2001) showed that 90% of teenagers use the Internet, nearly half of whom go online at least once a day; three out of four have Internet access from their homes, while others go online from schools or libraries.

Participants accessed the survey for this study via a site on the World Wide Web that was prepared and maintained by the second author (Smith). The Webmaster hired by Smith had set up the site to ensure complete anonymity of the respondents. All participants were volunteers who sought out the Web site of their own volition and chose to answer the survey questions. Completion of the survey signified informed consent. Participants could have declined to answer any specific questions and could have stopped participating at any time. Although no incentive was offered and no therapeutic benefit was planned, many respondents made appreciative comments for the opportunity to express their thoughts and feelings. It was clear from their comments that the anonymity of the survey gave them a sense of safety and security, enabling them to talk about things they may not have shared so readily with an adult faceto-face. The Institutional Review Board of the University of Tennessee granted approval of the study.

Sample

Survey responses were received from 339 American youth. If still in high school, 18- and 19-year-olds were included, but they were excluded if enrolled in college because the "school connectedness" variable undoubtedly has different meanings and connotations for college students. Initial data screening caused 53 surveys to be deleted because the respondents were in college. One survey was deleted because of excessive missing data. Three surveys were deleted because their responses were not deemed credible (one claimed to have brought an atomic bomb to school, another claimed to have incurred juvenile justice charges for "f-ing my dog in the ass," and still another said he had not only raped his teacher but also killed the class pet and brought a gun to school). It is not uncommon to have a few prank responses in any survey of this type.

The final sample, after deletion of the 57 surveys, consisted of 282 youth from 47 states and the District of Columbia. There were 123 boys and 158 girls in the sample (one respondent failed to report sex). Age range of respondents was 7 to 19 years, with a mean age of 15.3 years. Racial breakdown was 173 white, 109 minority, with the minority portion of the sample identifying themselves as black (n = 39), Hispanic (n = 35), and "other" races (n = 35). Although all grades from 1 to 12 were represented, the preponderance of youth were in grades 8 to 12. Youth were categorized as "violent" if they had been expelled from school for fighting or bringing a weapon and/or had been charged with a violent offense by the juvenile justice system (n = 82). The remaining 200 youth were categorized as "nonviolent." Within the "violent" category were 55 boys and 27 girls, with more whites than minorities among the boys, but roughly equal numbers of white and minorities among the group of violent girls.

Findings

Comparisons of Violent and Nonviolent Youth on Categorical Variables

Violent boys and girls were compared to nonviolent boys and girls on categorical variables using chi-square analysis (Table 1). Substantial percentages of violent youth (49% of boys, 41% of girls) did not perceive themselves to be liked by classmates and reported loneliness (70% of boys, 65% of girls): Statistically significant differences between violent and nonviolent youth were found on both of these variables (χ^sup 2^ 7.60 and 9.40, respectively; p

Comparisons of Violent and Nonviolent Youth on Anger Variables

Violent and nonviolent youth differed significantly, and in the predicted direction, on three of the FAS subscales: anger suppression (t = 3.55, p = .0005; violent youth less likely to suppress their anger), anger-out (t = -3.68, p =.0004; violent youth more likely to vent anger toward others), and anger discussion (t = 2.32 p = .02) (violent youth less likely to discuss their anger in a healthy way). Violent and nonviolent youth did not differ in anger somatization (t = -0.47, n.s.). Analysis of variance revealed no significant differences among the racial groups on any of the anger variables in this study. There were only two gender differences in anger behaviors. Consistent with previous research, boys scored higher on anger-out, while girls scored higher on anger somatization.

Relationship of Anger Discussion Scores With Other Variables

Because anger discussion is the only constructive mode of dealing with anger measured by the FAS, its correlates were examined. For both boys and girls, anger discussion (the propensity to talk about angry feelings with a friend or relative) was positively correlated with being liked by classmates (r = .21, p = 001). Anger discussion was inversely correlated with being angry enough to hit (r = -.21, p = .001), loneliness (r = -.19, p = .004), anger suppression (r = -23, p = .0003), and anger symptoms (r = -.14, p = .03).

Differences Between "Likers" and "Haters" of School

Likers (n = 46) of school were those who responded "I like it" when asked about their feelings, and haters (n = 65) were those who responded "It feels like jail." These two subgroups of the sample were compared, using chi-square analysis, and found to differ significantly on seven variables (p≤.01 for all): perceptions of classmates liking them (χ^sup 2^[df = 1, n = 101] = 12.72, p = .0004), loneliness (χ^sup 2^[df = 1, n = 101] = 12.11, p = .0005, perceptions of fairness of school discipline (χ^sup 2^[df =1, n = 105] = 15.86, p = .0001), perceptions of fair treatment by adults (χ^sup 2^[df = 1, n = 102] = 6.54, p = .01), feeling anger strongly enough to hit someone (χ^sup 2^[df = 1, n = 103] = 18.05, p = .0001), likelihood of violent behavior (χ^sup 2^[df = 1, n = 103] = 23.04, p = .0001), and gender (χ^sup 2^[df = 1, n = 103] = 14.02, p = .0002). Regardless of race, fewer boys than girls liked school. A substantial percentage of boys thought school "felt like jail": 34% of white boys, 40% of black boys, and 23% of Hispanic boys (Table 2). Using the Student's t-test, no age difference was found between likers and haters of school (t = .52, p = .6013).

Regarding School Connectedness

Qualitative data, gleaned from responses to the openended questions of the survey, were subjected to content analysis. We present here only the findings pertinent to school connectedness. The students resented (a) schools' emphasis on surveillance, conformity, and regimentation; (b) schools' inequitable discipline of "jocks" and other high-status groups versus the rest of the student body; (c) schools' overreaction to trivial offenses; and (d) schools' lack of action when they report bullying and harassment. These are just a few of the student comments:

Schools' emphasis on surveillance, conformity, and regimentation.

You can't be you. You have to dress a certain way, walk in a line, don't talk, put your hand up, do this, don't do that. But they teach you to be you.

There have been times when I felt that the teachers were biased against me during grading just because of the way I dress.

Students are forced to conform to the school's way of learning. Any attempt to approach assignments differently is immediately shot down verbally or in the grading process.

A few bad people make the school board and principal neurotic.

When my parents were in high school, they never had police officers walking through the halls or video cameras watching everything they did.

All the hype about Columbine has made everyone paranoid about everything.

Schools' inequitable discipline.

The school needs jocks, so they never get punished and are allowed to treat everyone else like trash. The preps can have their mommies' and daddies' lawyers make sure they never get in trouble, and are allowed to treat everyone else like trash.

The sports players and really smart kids get away with everything and the normal kids get detentions and suspensions just for being late and uniform violations.

Schools' overreaction to trivial offenses.

I asked a girl for a pencil and got sent to the administrator.

If someone is talking, I don't get to go to the bathroom after lunch and they never let you go during the day.

I was put in suspension [because of] a rumor that was spread about me bringing chewing tobacco to school. I didn't, but they put me in suspension anyhow, though it was all hearsay.

I really do think that stuff is nasty. I would never put something like that in my body.

Schools' lack of action when bullying and harassment are reported.

They never punish the people who are harassing students. If s like we get punished more.

Some [kids] teased me for days, months. I tell the teacher, they say they'll see what they can do about it and nothing happens. I tell my parents, they talk to them, and they still don't do anything helpful. They stopped teasing me when I punched them.

If a kid is getting picked on, they turn to a teacher who says, "I'll take care of it" and never does-then [the kid] gets picked on more for telling, so that's when you take control and put matters into your own hands and deal with it your way. That's how people become violent.

As shown in the final set of quotations, victimization was a frequently cited rationale for becoming violent. Over and over, students described peer behaviors of teasing, taunting, and bullying that led to their feeling angry enough to hit. Although the group categorized as "violent" was more likely to admit being angry enough to hit than the group categorized as "nonviolent," it is notable that 75% of all youth surveyed in this study admitted being angry enough to hit someone.

Discussion

The loneliness and alienation of violent youth from their classmates, so clearly evident in this study, bring to mind the words Columbine shooter Eric Harris wrote in his journal: "I hate you people for leaving me out of so many fun things. You people had my phone number, and I asked and all, but no no no no no, don't let the weird-looking Eric kid come along" (cited in Time, December 17, 2001, p. 16). Although a host of school violence-prevention programs promote conflict resolution and problem-solving skills, perhaps there is insufficient attention to the alienation of disliked and lonely students. Findings of the study suggest that interventions are needed to increase the social competence and connectedness of alienated students. In this study, fewer boys than girls (regardless of race) liked school, indicating a need for further study. This finding is discrepant from those of Bonny et al. (2000) and McNeely et al. (2002), who found that boys were more connected to school than girls were.

Participants in this study (whether violent or nonviolent) decried the repressive and prisonlike atmosphere in their schools, commenting about the rigid rules, armed guards, metal detectors, lockdowns, and video cameras monitoring everything. They also held strongly negative views of "zero tolerance" policies and disproportionate punishments, such as expulsion for writing an e-mail message with curse words in it. Too often, suspension or expulsion is the method of dealing with students who commit relatively minor infractions. Schools today suspend twice as many students as they did in the 1970s (Morse, 2002). Nearly 40% of schools studied by McNeely et al. (2002) gave out-of-school suspension the first time students were caught smoking, a relatively minor offense. Is this really the desirable solution? While we all want American schools to be safe, the punitive atmosphere described by our study participants does not foster school connectedness and does not promote a positive climate for learning. Creation of support programs for isolated students and a positive, accepting school climate are more likely to decrease hatred of school and angry acting-out behavior than draconian disciplinary measures.

On the other side of the draconian discipline issue, students expressed frustration regarding school officials' inaction when they reported substantive problems of harassment and bullying.

Female students interviewed by Simmons (2002) also reported that school officials failed to take their complaints about bullying seriously. Relational aggression was often downplayed, and sometimes the victim herself was blamed. Students at Columbine High likewise said their teachers and staff did not seem to notice the bullying and aggression that had become part of the school culture (Greenfield & Juvonen, 1999). Fear of bullying causes more than 160,000 children to skip school every day (National Association of School Psychologists, cited in Orecklin, 2000).

While some students react to harassment and bullying by staying home from school or withdrawing from peers, others retaliate with fists or weapons. Victimization is a known predictor of future violence. In a study of Appalachian white and black girls, Lucas (2000) found that their violence occurred in response to an impending attack. They perceived themselves as "good" girls forced by others' violent acts to respond in a "bad" way. In a recent longitudinal study of inner-city youth (Fry, Jemmott, Hines, & Fong, 2002), victimization experiences at baseline were significantly associated at 3-month followup with hurting someone else in a tight, carrying a knife or razor, and stabbing or shooting at someone. Other research shows that the amount (and type) of aggression experienced as a target is positively correlated with the behavior performed as an aggressor (Harris, 1996).

Reducing victimization, then, is one way to begin reducing youth violence. Many schools are implementing the Olweus Bullying Prevention Program, which is based on extensive research (Olweus, Limber, & Mihalic, 1999). After a Colorado school principal implemented "bullyproofing" initiatives at her school, the incidence of bullying behavior decreased. Among the strategies is "Be Cool" training in which students learn to choose between a "hot" response to provocation and the more desirable "cool" response (Labi, 2001). In a Los Angeles school, students (and their parents) sign contracts stipulating that no child can be teased or ridiculed on the basis of his/her appearance, gender, family, or grades (Labi). The Ophelia Project focuses on the relational type of aggression that is more frequently employed by girls (Simmons, 2002). The project includes a schoolwide training program, guidelines for parents, and use of high-school mentors to teach younger girls how to deal with relational aggression.

Anger-management programs can help reduce students' tendencies to solve problems with physical or relational aggression. Such programs focus on arousal management (calming down with relaxation or meditation techniques) and constructive anger expression (using words, not fists, to settle disputes). As shown in this study, discussion of anger is inversely correlated with feeling angry enough to hit someone. Talking out the angry feelings with an empathie listener-friend, parent, or counselor-is a healthy choice. Anger discussion was positively correlated with being liked by classmates, suggesting that students who feel more secure in their interpersonal relationships may feel more secure in disclosing negative emotions. Greater use of anger discussion was associated with decreased suppression or somatization of anger and decreased loneliness.

Cognitive behavioral therapy has proved to be effective with a wide variety of angry clients, including aggressive children and juvenile delinquents (Beck, 1999). Most successful school programs, such as the Peaceful Conflict Resolution and Violence Prevention Curriculum (DuRant, Barkin, & Krowchuk, 2001) and the Responding in Peaceful and Positive Ways Program (Meyer, Farrell, Northup, Kung, & Plybon, 2000), are based on cognitive-behavioral concepts. Tailoring such programs to specific subgroups, based on gender and/or race, may be beneficial, although more research is necessary.

In the present study, no racial differences were observed on any of the anger expression variables. These findings are consistent with studies of college students by Harris (1996) in which there were almost no differences in anger behaviors between blacks and whites and few differences between Hispanics and whites, although whites did acknowledge having committed a greater number of physically aggressive acts (such as slapping and hitting) over their lifetime. The findings were discrepant, however, from those of Deffenbacher and Swaim (1999), Hauber et al. (1998), and Reyes et al. (2003).

Gender differences in this sample were consistent with previous research. Virtually all studies show boys scoring higher on anger-out. The tendency of girls to score higher on the FAS anger symptoms scale has been demonstrated in numerous previous studies (e.g., Thomas & Williams, 1991), although reasons are unclear. The anger symptoms scale includes both bodily concomitants of anger arousal, such as headache, and the anxiety dimension of anger experience. Girls could score higher either because anger is a more anxiety-producing experience for them or because they are more attuned to their bodily reactions. Being able to talk to someone about the anger, however, lessened the amount of anger symptoms. Not all studies find gender differences. Among African-American youths given the FAS by Armstead and Clark (2002), no gender differences were found in anger-in, anger-out, or anger discussion (the anger symptoms subscale was not administered to the participants).

Limitations

Several limitations of this study must be acknowledged. Generalizability is an issue because of the nonrandom method of data collection. The racial characteristics of the sample do not match percentages of minorities in the national population. In future studies, having a larger number of minority participants to use in statistical comparisons may permit discovery of statistically significant differences between groups. Income data were not collected in the present study, but we presume that the Internet users who participated in our survey are more affluent and probably differ in other ways from youths contacted via a random sampling procedure. According to the Kaiser Family Foundation survey (Jesdanun, 2001), young people from higher SES backgrounds are more likely to have gone online than those from lower or working-class backgrounds. Whites are more likely to have Internet access from home (85%), compared with 66% of blacks and 55% of Hispanics. On the other hand, this sample of Internet users provided us with a valuable glimpse of contemporary youth who do not fit the standard violence risk profile of poverty, parental criminality, and community disorganization.

Self-report data create another study limitation, for the researchers have no means to check the veracity of respondents' answers. However, the same limitation would have applied to a mailed questionnaire. Any tendencies toward lying or making socially desirable responses should have been minimized by the anonymity provided by the Web site; participants were assured their answers could never be linked to their identities. Given that boys and girls had to seek out the Web site to become involved in the study, participation was truly voluntary. Both violent and nonviolent youth appeared to answer the Internet survey questions freely and candidly.

The Framingham Anger Scales may not tap all relevant dimensions of anger in samples of youth. To be more specific, the FAS do not permit a fully differentiated picture of aggressive forms of anger expression. Recent factor-analytic research with a new instrument identified three distinct forms of aggressive anger expression in adolescents: abusive verbal anger, physical aggression toward people, and physical aggression toward objects (slamming doors, throwing things) (Deffenbacher & Swaim, 1999). Factors replicated across gender, ethnic, and developmental groups (i.e., middle- and high-school students). The new instrument may prove useful in expanding our understanding of aggressive youth.

A final limitation of the study is its cross-sectional design, prohibiting discovery of answers to chicken-and-egg questions such as: Does violent behavior precede or follow disconnectedness from peers and school? Does perception of unfair school discipline precede or follow dislike of school and penalties for angry acting-out? Few studies have traced the trajectory of violent youth over time, illuminating crucial turning points.

One retrospective study, involving youths (ages 15-18) already incarcerated for murder, provides some clues regarding the trajectory (Pharris, 2002). In each boy's life, there was a critical traumatic event that set him on the pathway to violent crime. The event in some cases was being bullied or assaulted, in other cases not being able to protect a mother being assaulted, being placed in foster care or special education, or witnessing a shooting. These events occurred when the boys were aged between 9 and 14. Common to all events were reactions of shame and fear, compounded by the boy's inability to talk with anyone about these feelings. All of the boys began getting in trouble in school (fighting, getting suspended) and eventually lost their connection to school, dropping out around age 13. Relates Pharris, "A point came when they could not figure out how to get back into school or how to succeed in school if they were to get in. The lost connection with school was a source of despair as they felt their time was running out and it was too late to get back on track with their lives" (p. 37). The boys felt that no one wanted them in the school anyway. Future research must examine such critical events more closely, with an eye toward swift and effective preventive and remedial interventions that can be undertaken by schools, juvenile justice officials, and mental healthcare providers.

Recommendations

Mental health nurses should advocate for adoption of the national agenda developed by the Commission for the Prevention of Youth Violence (2000). Included in the Commission's report were several recommendations that have particular relevance to this study: instituting a comprehensive, evidence-based violence-prevention curriculum in every school (K-12); implementing alternative school programs to provide a safety net for students who have been expelled; and expanding screening and support services within schools to ensure that youth at risk for violence are identified and have access to appropriate monitoring and treatment. Early identification of youth at risk for, or involved in, violence must become a national priority. However, these recommendations are only stepping stones. It is not enough to simply put in place programs to help children. The effectiveness of such programs also must be monitored; those that are not working to reduce violence must be changed or discontinued. Program content must be evaluated to determine if it increases school connectedness, promote positive interpersonal interactions, and reduces anger behaviors among those youth at risk.

With regard to fostering climate change in schools, mental health nurses could be involved in provision of both direct and indirect interventions, including consultations with teachers, school nurses, and other staff, family-centered services, and preventive strategies to promote children's mental health. Our study clearly shows how important it is to consider the school environment as a determinant of violent behavior. In the past, professionals have focused almost exclusively on the pathology of individual children in accordance with the medical model. A broader focus includes not only the child, but also his or her peers, teachers, and school administrators. Students who are being bullied need to feel that the staff is going to take definitive action on their behalf. Staff must be made aware that the popular technique of peer mediation may be ineffective in situations of bullying because there is a power imbalance between the bully and the victim (Crawford, 2002).

Let us give you a real world example of ineffective response to a bullying situation. The second author (Smith) worked as a consultant to a school system several years ago. A depressed 16-year-old girl was referred to Smith by her teacher to determine the source of her depression and provide counseling. The girl reported that she was being abused and harassed at school by a gang of girls who had even come to her home and threatened her family. Unfortunately, the school chose the path of least resistance; rather than do something about the gang of girls, their "solution" was to send the girl to an alternative school to finish out her high school years. The girl gang got off scot-free, ready to find their next victim. This solution may work in the short term but the longterm repercussions could be disastrous.

In order to take proper action, the power dynamics of the school must be understood. Mental health nurses could assist school nurses and teachers to conduct an assessment of the power dynamics of the school. Such an assessment could be an important step toward creation of a better climate (Twemlow, Fonagy, & Sacco, 2001). The following questions can be used as catalysts for discussions with students:

* Which groups or cliques can you clearly identify in your school? Is one group dominant?Are there racial/ ethnic groups who control the school?

* Are there any gangs in your school?

* Do young people plan fights during the day and talk about who will win and when or where they will fight?

* Do teachers appear intimidated at your school?

* Are there teachers or counselors you can speak to about these problems?

* Have you ever reported a student being bullied? What happened?

* What does your school tell you about how to handle bullying or what to do if you hear someone threatening to kill somebody? (Twemlow, Fonagy, & Sacco, 2001, p. 378)

It is time for Americans to pay greater attention to healthy school environments. We concur with the assertion of Reinke and Herman (2002) that "school settings can either lessen or compound risk factors (e.g., parent relationships, poverty, neighborhood effects) that children bring to the classroom" (p. 796). Every child should have an opportunity to learn, and to interact with classmates and teachers, in a safe environment that facilitates emotional intelligence and psychological well-being.

Conclusion

Results of this study indicate that there is insufficient attention to the alienation of disliked and lonely students. Mental health nurses could play a pivotal role in fostering change in the social climate of schools and helping youth to achieve better anger management and social skills.

References

Armstead, C.A., & Clark, R. (2002). Assessment of anger expression in pre- and early-adolescent African Americans: Psychometric considerations. Journal of Adolescence, 25, 365-371.

Ausbrooks, E., Thomas, S.P., & Williams, R.L. (1995). Relationships among self-efficacy, optimism, trait anger, and anger expression. Health Values, 19(4), 46-53.

Beck, A.T. (1999). Prisoners of hate: The cognitive basis of anger, hostility, and violence. New York: HarperCollins.

Bonny, A.E., Britto, M.T., Klostermann, B.K., Hornung, R.W., & Slap, G.B. (2000). School disconnectedness: Identifying adolescents at risk. Pediatrics, 106,1017-1021.

Brener, N.D., Collins, J.L., Kann, L, Warren, C.W., & Williams, B.I. (1995). Reliability of the Youth Risk Behavior Survey Questionnaire. American Journal of Epidemiology, 141, 575-580.

Centers for Disease Control and Prevention. (1999). Youth risk behavior surveillance-United States, 1999. Atlanta: Author.

Commission for the Prevention of Youth Violence. (2000). Youth and violence. Retrieved May 29,2002, from www.ama-assn.org/violence

Crawford, N. (2002). New ways to stop bullying. Monitor on Psychology, 33(9), 64-66.

Crick, N.R., Bigbee, M.A., & Howes, C. (1996). Gender differences in children's normative beliefs about aggression: How do I hurt thee? Let me count the ways. Child Development, 67,1003-1014.

Crick, N.R., Grotpeter, J.K., & Bigbee, M.A. (2002). Relationally and physically aggressive children's intent attributions and feelings of distress for relational and instrumental peer provocations. Child Development, 73,1134-1142.

Deffenbacher, J.L., & Swaim, R.C. (1999). Anger expression in Mexican American and White Non-Hispanic adolescents. Journal of Counseling Psychology, 46, 61-69.

DuRant, R.H., Barkin, S., & Krowchuk, D. (2001). Evaluation of a peaceful conflict resolution and violence prevention curriculum for sixthgrade students. Journal of Adolescent Health, 28,386-393.

Erwin, E. (2002). Adolescent perceptions of relevant social problems. Journal of Child and Adolescent Psychiatric Nursing, 15, 24-34.

Fawcett, J., & Buhle, E.L. (1995). Using the internet for data collection: An innovative electronic strategy. Computers in Nursing, 13, 273-279.

Fry, D., Jemmott, L.S., Mines, P.M., & Fong, G.T. (2002, April). Victimization and subsequent violent behavior among African American and Latino youth. Paper presented at the Society of Behavioral Medicine, Washington, DC.

Greenfield, P.M., & Juvonen, J. (1999, July/August). A developmental look at Columbine. American Psychological Association Monitor, p. 33.

Harris, M.B. (1996). Aggression, gender, and ethnicity. Aggression and Violent Behavior, 1,123-146.

Hauber, R.P., Rice, M.H., Howell, C.C., & Carmon, M. (1998). Anger and blood pressure readings in children. Applied Nursing Research, 11(1), 2-11.

Hawkins, J.D., Herrenkohl, T.I., Farrington, D.P., Brewer, D., Catalane, R.F., Harachi, T. W., & Cothern, L. (2000). Predictors of youth violence. Rockville, MD: Oifice of Juvenile Justice and Delinquency Prevention, U.S. Department of Justice.

Haynes, S., Levine, S., Scotch, N., Feinleib, M., & Kannel, W.B. (1978). The relationship of psychosocial factors to coronary heart disease in the Framingham Study: I. Methods and risk factors. American Journal of Epidemiology, 107, 362-383.

Horowitz, J.A., Vessey, J.A., Carlson, K.L., Bradley, J.F., Montoya, C., McCullough, B., & David, J. (2004). Teasing and bullying experiences of middle school students. Journal of the American Psychiatric Nurses Association, 10,165-172.

Jesdanun, A. (2001, December 13). Study: Youths research touchy health topics online. Knoxville New-Sentinel, p. C2.

Joinson, A. (1998). Causes and implications of disinhibition behaviors on the Internet. In J. Gackenbach (Ed.), Psychology and the internet (pp. 43-58). New York: Academic Press.

Josephson Institute of Ethics. (2001, April). The ethics of American youth: Violence and substance abuse: Press release, data and commentary. Retrieved July 11, 2002, from www.josephsoninstitute.org/survey2000 / violence2000-commentary.htm

Labi, N. (2001, April 2). Let bullies beware. Time, pp. 46-47.

Lopez, V.A., & Emmer, E.T. (2002). Influences of beliefs and values on male adolescents' decisions to commit violent offenses. Psychology of Men and Masculinity, 3, 28-40.

Lucas, R.L. (2000). The Southern Appalachian, adolescent female's experience of being violent. Unpublished doctoral dissertation, University of Tennessee, Knoxville.

McNeely, C.A., Nonnemaker, J.M., & Blum, R.W. (2002). Promoting school connectedness: Evidence from the National Longitudinal Study of Adolescent Health. Journal of School Health, 72,138-146.

Meyer, A.L., Farrell, A.D., Northup, W.B., Kung, E.M., & Plybon, L. (2000). Promoting nonviolence in early adolescence: Responding in peaceful and positive ways. New York: Plenum.

Morse, J. (2002, May 27). Learning while black. Time, pp. 50-52.

Nansel, T.R., Overpeck, M., Pilla, R.S., Ruan, W.J., Simons-Morton, B., & Scheldt, P. (2001). Bullying behaviors among U.S. youth: Prevalence and association with psychosocial adjustment. JAMA, 285, 2094-2100.

Olweus, D., Limber, Sv & Mihalic, S. (1999). The Bullying Prevention Program: Blueprint for violence prevention. Boulder, CO: Center for the Study and Prevention of Violence.

Orecklin, M. (2000, August 21). Beware of the in crowd. Time, p. 69.

Pharris, M.D. (2002). Coming to know ourselves as community through a nursing partnership with adolescents convicted of murder. Advances in Nursing Science, 24(3), 21-42.

Reinke, W.M., & Herman, K.C. (2002). A research agenda for school violence prevention. American Psychologist, 57, 795-797.

Resnick, M., Bearman, P., Blum, R., Bauman, K., Harris, J., Tabor, J., Beuhring, T., Sieving, R., Shew, M., Ireland, M., Bearinger, L., & Udry, R. (1997). Protecting adolescents from harm: Findings from the National Longitudinal Study on Adolescent Health. JAMA, 278, 823-832.

Reyes, L.R., Meininger, J.C., liehr, P., Chan, W., & Mueller, W.H. (2003). Anger in adolescents: Sex, ethnicity, age differences, and psychometric properties. Nursing Research, 52, 2-11.

Shaefer, D.R., & Dillman, D.A. (1998). Development of a standard email methodology. Public Opinion Quarterly, 62, 378-397.

Simmons, R. (2002). Odd girl out: The hidden culture of aggression in girls. New York: Harcourt.

Smith, H. (2000). The scarred heart: Understanding and identifying kids who kill. Knoxville, TN: Callisto Publishing.

Smith, H., & Thomas, S.P. (2000). Violent and nonviolent girls: Contrasting perceptions of anger experiences, school, and relationships. Issues in Mental Health Nursing, 21,547-575.

Snyder, H.N., & Sickmund, M. (1999). Juvenile offenders and victims: A national report. Washington, DC: U.S. Department of Justice Programs, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention.

Stapley, J., & Haviland, J. (1989). Beyond depression: Gender differences in normal adolescents' emotional experiences. Sex Roles, 20, 295-308.

Thomas, S.P. (1993). Women and anger. New York: Springer.

Thomas, S.P., & Williams, R.L. (1991). Perceived stress, trait anger, modes of anger expression and health status of college men and women. Nursing Research, 40,303-307.

Twemlow, S.T., Fonagy, P., & Sacco, F.C. (2001). An innovative psychodynamically influenced approach to reduce school violence. Journal of the American Academy of Child and Adolescent Psychiatry, 40, 377-379.

U.S. Department of Commerce. (2002). A nation on-line'. How Americans are expanding their use of the internet. Washington, DC: Author.

U.S. Department of Health and Human Services. (2000). Healthy People 2010. Washington, DC: Author.

Sandra P. Thomas, PhD, RN, FAAN, and Helen Smith, PhD

Sandra P. Thomas, PhD, RN, FAAN, is Professor and Director of the PhD program, College of Nursing, Knoxville. Helen Smith, PhD, is a psychologist in private practice, Southeastern Pyschological Services, Knoxville, TN.

Author contact: sthomas@utk.edu, with a copy to the Editor: mary@artwindows.com

Copyright Nursecom, Inc. Oct-Dec 2004
Provided by ProQuest Information and Learning Company. All rights Reserved