Peer Victimization the Nature and Prevalence of Bullying among Children and Youth

Susan Limber. Handbook of Children, Culture, and Violence. Editor: Nancy E Dowd, Dorothy G Singer, Robin Fretwell Wilson. Sage Publication. 2008.

Although bullying is an age-old phenomenon, it has only recently been recognized as a serious and pervasive problem among children and youth in the United States. Research regarding peer bullying in Scandinavia has been active for more than three decades, and wide-scale public attention to the problem was aroused there in the early 1980s, after the suicides of several young victims of bullying were brought to light (Olweus, 1993a). In the United States, such wide-scale interest in bullying was not triggered until the spring of 1999, when media accounts of the tragic shootings at Columbine High School identified the perpetrators as victims of bullying by classmates. Research on the nature and extent of bullying among children and youth ballooned after Columbine as we rushed (belatedly) to better understand and prevent the phenomenon.

Bullying is most commonly defined as repeated, aggressive behavior, in which there is an imbalance of power or strength between the parties (Nansel, Overpeck, Pilla, Rúan, Simmons-Morton, et al., 2001; Olweus, 993a). Bullying behaviors may be direct (e.g., hitting, kicking, taunting, malicious teasing, name-calling) or indirect (e.g., rumor-spreading, social exclusion, friendship manipulation, cyber-bullying) (Olweus, 1993a; Rigby, 1996).

Research in recent years confirms that bullying is a complex phenomenon with no single cause. Rather, bullying among children and youth is best understood as the result of a dynamic interaction between an individual and his or her social ecology—his or her family, peer group, school, and broader community (Espelage & Swearer, 2004; Olweus, Limber, & Mihalic, 1999). This chapter summarizes current research on the nature and prevalence of bullying among children and youth, and identifies aspects of a child’s social ecology that may increase or decrease the risk of bullying.

Prevalence of Bullying

The earliest systematic studies assessing the prevalence of bullying were conducted by Olweus (1993a) with 150,000 Norwegian and Swedish children aged 8 to 15. Using anonymous, self-report measures, Olweus found that 15% of children and youth reported being involved in bully-victim problems 2 to 3 times per month or more often. Nine percent reported having been bullied by their peers, 7% had bullied others, and approximately 2% had both bullied others and been bullied with this frequency.

Surveys of children and youth in the United States typically have found higher rates of bullying than those in Scandinavia (Melton Limber, Cunningham, Osgood, Chambers, et al, 1998; Nansel et al, 2001). In a nationally representative sample of more than 15,000 students in Grades 6 through 10, Nansel and colleagues (2001) found that 17% of children and youth reported having been bullied “sometimes” or more often during the school term and 19% had bullied others “sometimes” or more frequently. Of five specific types of bullying that were examined by the authors (being “belittled about religion or race”; being “belittled about looks or speech;” being “hit, slapped, or pushed;” being “subjects of rumors;” or being “subjects of sexual comments or gestures”), being belittled about one’s looks or speech was most common. Most recently, Finkelhor and colleagues (2005) conducted telephone interviews with children and parents to assess a wide range of victimization among children and youth aged 2 to 17. In this nationally representative sample, the researchers observed that 22% of children and youth had been physically bullied in the last year (resulting in a national estimate of 13.7 million children and youth), and 25% had been teased or emotionally bullied within the past year (an estimated 15.7 million children and youth nationally).

Demographic Influences on Bully Victimization

The prevalence of bullying among children and youth is correlated with a variety of demographic characteristics, including age, gender, race, and ethnicity. Despite popular mythology, it is not more prevalent in urban as opposed to rural or suburban settings.

Age and Developmental Stage

Somewhat different developmental trends have been observed depending upon whether the focus is on victims or perpetrators. Although few studies have systematically examined bullying across broad age groups of children, researchers who have focused on victimization experiences have found the highest rates of bully victimization among elementary-aged children, somewhat lower rates among middle school students, and lower rates still among high school students (Finkelhor et al, 2005; Limber, 2002; Nansel et al, 2001; Olweus, 1991). Finkelhor and colleagues (2005) compared victimization experiences of children aged 2 to 5, 6 to 12, and 13 to 17, and observed that children aged 6 to 12 experienced the highest rates of physical bullying. They also experienced the highest rates of teasing or emotional bullying. Preschool-aged children experienced the second-highest rates of physical bullying, and children and youth aged 13 to 17 experienced the least amount of physical bullying. There were no significant differences in rates of teasing or emotional bullying among children in the 2- to 5-year-old group compared to the 13- to 17-year-old group.

Nansel et al. (2001) found that rates of occasional (“sometimes”) or weekly bully victimization decreased steadily among sixth through tenth graders in the United States. Although 24% of children in Grade 6 reported being bullied “sometimes” or more often, only 9.4% of students in tenth grade reported being bullied with this frequency. Similarly, in his study of more than 150,000 Norwegian and Swedish students age 8 to 15, Olweus (1991) observed steadily decreasing rates of victimization from children aged 8 through 15. Rates of bully victimization were twice as high for 8-year-old boys (17.5%) compared with 12-year-old boys (8.4%), and nearly three times as high for 8-year-old girls (16%) compared with 12-year-old girls (5.5%). By age 15, rates of bully victimization had decreased to 6.4% among boys and 3% among girls.

Children and youth typically report being bullied either by same-aged peers or by older children and youth (Olweus, 1993b). This finding may explain why somewhat different age trends are found when focusing on rates of bullying others versus rates of bully victimization. Most researchers have found peaks of self-reported bullying during early to mid-adolescence (Eisenberg & Aalsma, 2005; Espelage & Swearer, 2003; Nansel etal, 2001; Olweus, 1991). Olweus (1991) found the highest rates of self-reported bullying among 14- and 15-year-olds. Similarly, Nansel et al. (2001) noted in their sample that the highest rates of self-reported bullying were among eighth graders. Because of the heightened prevalence of bullying among elementary and middle school children, most bullying prevention and intervention efforts have appropriately focused at these grade levels. Although some researchers have noted a decrease in physical bullying and an increase in verbal bullying with age (Craig, 1998; Rivers & Smith, 1994), more research is needed to examine age trends in various types of bullying experienced by children at different ages.

Gender Differences

Although both girls and boys are frequently engaged in bullying problems, researchers have debated the relative frequency with which they engage in and experience bullying. Studies relying on self-report measures typically have found that boys are more likely than girls to bully (Duncan, 1999; Haynie, Nansel, Eitel, Crump, Saylor, et al., 2001; Nansel et al., 2001; Olweus, 1993a; Seals & Young, 2003). Findings are less consistent when examining gender differences in peer victimization. Some studies (Boulton & Underwood, 1992; Haynie et al, 2001; Nansel et al, 2001; Olweus, 1993a; Perry, Kusel, & Perry, 1988; Rigby & Slee, 1991; Ronning, Handergaard, & Sourander, 2004; Whitney & Smith, 1993) have found that boys report higher rates of victimization than girls. Other studies, however, have found either no gender differences or only marginal differences (Boulton & Smith, 1994; Charach, Pepler, & Ziegler, 1995; Duncan, 1999; Hoover, Oliver, & Hazier, 1992; Melton et al, 1998). What is clear is that girls are bullied by both boys and girls, while boys are most often bullied by other boys (Limber, 2002; Melton et al., 1998; Olweus, 1993a).

Perhaps more important than the relative frequency with which boys and girls are bullied is the different types of bullying that they engage in and experience. Boys are more likely than girls to experience physical bullying by their peers (Finkelhor et al., 2005; Harris, Petne, & Willoughby, 2002; Nansel et al, 2001; Ronning et al, 2004). The picture is somewhat less clear when examining gender differences for other forms of bullying. Ronning and colleagues (2004) found that boys in grades 6 and 10 were more likely than girls to be verbally bullied, but they found no such gender differences in Grades 7 to 9. Finkelhor and colleagues (2005) found no gender differences in rates with which boys and girls aged 2 to 17 had been teased or emotionally bullied. However, Nansel and colleagues (2001) found that girls were more likely than boys to be bullied through rumor-spreading or being the subjects of sexual comments or gestures. Olweus (cited in Limber, 2002) studied the nature of same-gender bullying (e.g., bullying of girls by girls) and found that girls are more likely than boys to be bullied through social exclusion.

There are inconsistent findings of gender differences in children’s use of indirect or relational forms of bullying (Crick & Grotpeter, 1995; Espelage, Holt, & Henkel, 2003; Espelage & Swearer, 2003; Rys & Bear, 1997). For example, although Crick & Grotpeter (1995) found that third- through sixth-grade girls were much more likely than their male peers to be rated as relationally aggressive (i.e., behaviors intended to “damage another child’s friendships or feelings of inclusion by the peer group,” others have failed to replicate these findings (Espelage et al, 2003; Rys & Bear, 1997).

Racial and Ethnic Differences

Although relatively few large-scale studies have systematically examined race and ethnic factors in bullying, several researchers have observed significant differences in rates of peer victimization among children and youth from different racial and ethnic groups (Graham & Juvonen, 2002; Hanish & Guerra, 2000; Nansel et al, 2001). For example, in their nationally-representative survey of more than 15,000 U.S. students in Grades 6 to 10, Nansel and colleagues (2001) found that Black youth reported being bullied less frequently than White or Hispanic children, and Hispanic youth reported slightly higher involvement in moderate and frequent bullying of others compared with White or Black students. A different pattern emerged in Graham and Juvonen’s (2002) study of students from a multiethnic middle school. Black students were more likely than Latino and multiethnic students to be named by their peers as aggressive, and less likely to be nominated as victims of peer victimization. Additional research is needed to further explore how numerical majority/minority status within schools and communities, status hierarchy (e.g., economic and social mobility of different racial and ethnic groups), and cultural views of or experiences with aggression influence rates of bullying among peers (Graham & Juvonen, 2002).

Bullying in Urban, Suburban, and Rural Communities

Although bullying has often been viewed as a problem primarily of urban schools, there is no evidence to support this claim (Nansel et al, 2001; Seals & Young, 2003). To the contrary, bullying has been documented in widely diverse communities. In their nationally representative study of sixth through tenth graders in the United States, Nansel and colleagues (2001) found no significant differences in rates of bully victimization among youth from urban, suburban, town, and rural areas. They observed only very small differences in students’ reports of bullying others. By self-report, suburban youth were 2 to 3% less likely to report participating in bullying with moderate frequency (i.e., “sometimes” or more often), and rural youth were 3 to 5% more likely to ever bully their peers.

Behavioral Characteristics of Children Involved in Bullying

In order to understand how to best intervene to prevent and address bullying among children and youth, it is useful to understand the behavioral characteristics of children who bully and those who are bullied. Although researchers have documented some common behavioral patterns, it is also important to note that (a) not all children fit these patterns, and (b) these patterns may not be static.

Children Who Bully

Researchers have identified a number of general characteristics of children who bully their peers regularly. These children tend to have impulsive, dominant personalities; have difficulty conforming to rules; and view violence in a positive light (Limber, 2002; Olweus, 1993a; Olweus et al, 1999). These children engage in both proactive aggression (i.e., goal-directed, deliberate aggression) and reactive aggression (defensive responses to provocation) (Camodeca & Goossens, 2005). Contrary to the common myth that children who bully are “loners,” these children are not socially isolated (Cairnes, Cairnes, Neckerman, Gest, & Gariepy, 1988; Nansel et al, 2001; Olweus, 1978; Juvonen, Graham, & Schuster, 2003). For example, Nansel and colleagues (2001) noted that among the sixth to tenth graders in their study, children who bullied their peers had an easier time making friends than other children.

In their study of sixth-grade students from low-socioeconomic status urban communities, Juvonen and colleagues (2003) observed that children who bully were less depressed, socially anxious, and lonely than their peers (according to self-report measures of psychological distress). Their classmates rated them as having high social status, and teachers confirmed that children who bully were the most popular of all students. Recent research by Sutton and colleagues (Sutton, Smith, & Swettenham, 1999a, 1999b) reveals that children who bully (and particularly those who are “ringleaders”) have high scores on tests of social cognition. As Sutton and colleagues (1999b) note, “In contrast to the popular stereotype and research tradition of the ‘oafish— bully lacking in social skills and understanding, the bully may be a cold, manipulative expert in social situations, organizing gangs and using subtle, indirect methods” (p. 435).

Children who bully are more likely than their peers to be engaged in a variety of antisocial, violent, or troubling behaviors (Haynie et al., 2001). They are more likely to be involved in vandalism (Olweus, 1993a), fighting (Nansel, Overpeck, Haynie, Ruan, & Scheldt, 2003; Olweus, 1993a), theft (Olweus, 1993a), and weapon-carrying (Nansel et al., 2003), and are more likely than non-bullying peers to consume alcohol (Nansel et al., 2001; Olweus, 1993a) and smoke (Nansel et al., 2001). They may exhibit poorer school adjustment in terms of academic achievement, assessments of school climate (Nansel et al., 2001), and school dropout rate (Byrne, 1994). In a study of fifth to seventh graders in rural southern communities, and colleagues (Cunningham, Henggeler, Limber, Melton, & Nation, 2000) observed that rates of self-reported bullying were related to students’ reasons for gun ownership. Those students who owned guns in order to gain respect or frighten others (high-risk gun owners) reported higher rates of bullying others than did their peers who owned guns for reasons of protection or sport (low-risk gun owners) or those who did not own guns at all.

Bullying also may be an indicator that boys are at risk for engaging in later troublesome and criminal behavior (Limber, 2002; Loeber & Dishion, 1983; Olweus, 1993a; Pellegrini, 2001, 2002). Preliminary findings from a longitudinal study by Pellegrini (2001) suggest that male perpetrators of sexual harassment at the end of middle school had been identified as bullies in elementary school and at the beginning of middle school. In a longitudinal study conducted by Olweus (1993a) in Norway, 60% of boys identified as bullies in middle school had at least one criminal conviction by age 24, and 35 to 40% had three or more convictions. To date, no such longitudinal studies have focused on girls who bully their peers.

Children Who are Bullied

Children who are bullied by their peers are commonly characterized as being either “passive victims” or “bully-victims” (also referred to as “aggressive victims” or “provocative victims”) (Limber, 2002; Olweus, 1993a). Care should be taken to ensure that these terms are not used pejoratively or to blame children for being bullied. Although estimates vary somewhat from study to study, passive victims comprise a larger proportion of victimized children than do bully-victims (Limber, 2002; Nansel et al., 2001; Olweus, 2001). For example, Nansel and colleagues (2001) observed that 11% of their nationally representative sample were passive victims of bullying, whereas only 6% were characterized as bully-victims. Passive victims of bullying have been described in the literature as being distanced from and submissive to their peers. In a study by Juvonen and colleagues (2003), teachers rated victims of bullying as being unpopular, and their peers rated them as having very low social status. Frequently, victims of bullying are socially isolated from their peers (Nansel et al., 2001; Olweus, 1993a) and report feeling lonely (Hawker & Boulton, 2000; Juvonen, Graham, & Schuster, 2003; Kochenderfer & Ladd, 1996; Nansel et al, 2001).

Victims of bullying are also more likely than their non-bullied peers to report feeling anxious (Craig, 1998; Fekkes, Pijpers, & Verloove-Vanhorick, 2004; Hawker & Boulton, 2000; Hodges & Perry, 1996; Juvonen et al., 2003; Olweus, 1978). Anxiety among bullied children may be viewed both as a precursor to and a consequence of bully victimization (Olweus, 2001; Swearer, Grills, Haye, & Cary, 2004). For example, anxious behaviors may signal to others that a child is an easy target for victimization. On the other hand, anxious symptoms may develop among bullied children as a result of their experience, as they react to their threatening environment with hypervigilance (Roth, Coles, & Heimberg, 2002; Swearer et al, 2004).

Depression is another internalizing behavior that is more common among bullied versus non-bullied children (Craig, 1998; Hawker & Boulton, 2000; Hodges & Perry, 1996; Juvonen et al., 2003; Kumpulainen, Rasanen, & Puura, 2001; Olweus, 1978; Rigby & Slee, 1993; Seals & Young, 2003; Slee, 1995; van der Wal, de Wit, & Hirasing, 2003). In a study of 2,700 Dutch children aged 9 to 12, Fekkes and colleagues (2004) observed that bullied children were four times as likely as their non-bullied peers to report feeling unhappy. They also were three times as likely to show moderate signs of depression, and eight times as likely to exhibit strong indications of depression. Several studies also have linked bully victimization with suicidal ideation (van der Wal, et al, 2003; Rigby, 1996; Roland, 2002). For example, Rigby (1996) observed that children who are frequently bullied (e.g., at least once a week) were twice as likely as their peers to wish they were dead or to admit to having recurring thoughts of suicide. Van der Wal and colleagues (2003) found that the associations between bully victimization and depression and suicidal ideation are stronger for indirect (e.g., “They pretend they don’t see me”) than direct (e.g., “They hit me”) forms of bullying.

Bullied children are more likely than their peers to suffer from low self-esteem (Eagan & Perry, 1998; Hawker & Boulton, 2000; Hodges & Perry, 1996; Olweus, 1978; Rigby & Slee, 1993). Although children’s self-esteem may be lowered as a consequence of being bullied, Eagan and Perry (1998) observed that low self-regard, especially when assessed in terms of children’s self-perceptions of low social competence within their peer group, also contributes to peer victimization. The authors speculate that there are likely a number of explanations for this finding. A child’s confidence in his or her standing within the peer group may protect him or her from being victimized. Thus, although most children are occasionally tested by threats from their peers, those with high self-regard refuse to tolerate these attacks and defend themselves more assertively than others. Children who have low self-perceived peer social competence also may exhibit behaviors that “signal” to their peers that they may be “easy targets” for bullying (Eagan & Perry, 1998; Limber, 2002).

Research also reveals common health effects of bullying. In a recent study of 2,766 Dutch elementary school children aged 9 to 12, Fekkes and colleagues (2004) compared self-reports of psychosomatic complaints among bullied children, children who bullied, and children not directly involved in bullying incidents. The researchers observed a consistent association between bully victimization and all measured health symptoms. For example, compared with children who were not involved in bullying, bullied children were approximately three times as likely to experience headaches, feel listless, and wet their beds. They were about twice as likely to experience problems sleeping, have abdominal pain, feel tense, be tired, and have a poor appetite.

Finally, there is some indication that bullying may affect not only the health but also the academic work of bullied children. Bullied children are more likely than their peers to report wanting to avoid attending school (Kochenderfer & Ladd, 1996) and have higher absenteeism rates (Rigby, 1996; Smith, Talamelli, Cowie, Naylor, & Chauhan, 2004). In a cross-sectional study of students in Grades 7 to 12, Eisenberg, Neumark-Sztainer, and Perry (2003) found that those students who were most frequently bullied by peers were more likely than others to report disliking school. They also received the lowest grades. Although these findings do not necessarily imply a causal relationship between bullying and academic achievement, the authors speculate that “young people mistreated by peers may not want to be in school and may thereby miss out on the benefits of school connectedness as well as educational advancement” (Eisenberg et al., 2003, p. 315).

As noted above, children who are anxious and who have low self-esteem are at increased risk of being bullied by their peers. In addition, research suggests that bully victimization is higher among boys who are physically weaker than their peers (Olweus, 1993b), and among children and adolescents who are gay, lesbian, or bisexual (Eisenberg & Aalsma, 2005; Garofalo, Wolf, Kessel, Palfrey, & DuRant, 1998). There also is a small but growing literature on the bullying of children with disabilities and special needs in mainstream settings (Martlew & Hodson, 1991; Mishna, 2003; Nabuzoka & Smith, 1993; Thompson, Whitney, & Smith, 1993). Children with chronic diseases or special health care needs may be at increased risk of being bullied or harassed by their peers. Researchers have observed higher rates of victimization among children with medical conditions that affect their appearance (e.g., cerebral palsy, muscular dystrophy, spina bifida) (Dawkins, 1996), children with hemi-plagia (paralysis on one side of the body) (Yude, Goodman, & McConachie, 1998), and children with diabetes (Storch, Lewin, & Silverstein, 2004a; Storch, Lewin, Silverstein, Heidgerken, Strawser, et al., 2004b).

There also is some indication that overweight and obese school children are more likely than their normal-weight peers to be bullied (Janssen, Craig, Boyce, & Pickett (2004). In a representative study involving more than 5,700 Canadian children aged 11 to 16, researchers found a direct and significant relationship between Body Mass Index (BMI) and victimization among all ages of girls studied, and among the youngest group of boys (aged 11-12). Specifically, overweight and obese girls and boys were more likely than normal-weight peers to experience verbal victimization (be teased or made fun of) and relational bullying (e.g., be socially excluded). Moreover, overweight or obese girls were more likely than their peers to be physically bullied.

Lasting Effects of Bullying

Compared to the extensive research that has been conducted on the immediate effects of peer victimization, there has been a dearth of research on its possible long-term effects (Roth et al., 2002). In a retrospective study of young adults, Olweus (1993b) found that former victims of bullying were more likely than their non-bullied peers to be depressed and have low self-esteem, even though they were no longer experiencing bullying. Similarly, Roth and colleagues (2002) found a relationship between teasing during childhood and depression and trait anxiety in young adulthood. They speculate that, “Children who are repeatedly teased may develop beliefs that the world is a dangerous place and that they have little control over outcomes in their lives” (p. 161). These are characteristic thinking patterns of individuals suffering from anxiety and depression. In a study assessing possible long-term effects of childhood bullying on lesbian, gay, and bisexual, and transgendered adults, Rivers (2001) noted that lesbian, gay, and bisexual individuals who were former victims of bullying were more likely than comparison groups (e.g., lesbians, gays and bisexuals who had not been bullied, or heterosexuals who had been bullied) to be prone to depression but not anxiety.

Aggressive Victims of Bullying (Bully-Victims)

As noted previously, some children are bullied with regularity but also bully other children. Referred to as aggressive victims, bully-victims, or provocative victims, these children may tend to be hyperactive (Kumpulainen & Rasanen, 2000), restless, and have difficulty concentrating (Olweus, 1993a, 2001). They may be more clumsy and immature than other peers their age, and they may have difficulty reading the social cues of their peers. These children tend to be quicktempered and may try to fight back when they feel that they have been insulted or attacked. Not only are they bullied by their peers, but they also may try to bully weaker children (Olweus, 1993a, 2001).

Research suggests that there is particular reason to be concerned about aggressive victims of bullying, as they may display some of the social and emotional problems of victimized children and the behavioral problems of children who bully (Haynie et al., 2001; Limber, 2002; Nansel et al, 2001). For example, Nansel and colleagues (2001) observed that bully-victims in Grades 6 to 10 were more likely to report poor relationships with classmates and be lonely. They also were at higher risk of smoking, fighting, and lower academic achievement. Similarly, Haynie and colleagues (2001) found that bully-victims rated more poorly than other children (victims, bullies, or children not involved in bullying) across a wide spectrum of variables, including problem behaviors, behavioral misconduct, self-control, deviant peer influences, social competence, school adjustment and bonding, and depressive symptoms.

In a recent study involving 1,985 mostly Latino and Black sixth graders, Juvonen and colleagues (2003) assessed self-reports of psychological distress, peer ratings, and teacher ratings of students who were categorized based on peer nominations as “bullies,” “bully-victims,” “borderline,” or “uninvolved.” The bully-victims were the most troubled group of the four. They were identified by their peers as the children who were most avoided. Teachers rated bully-victims as being low in popularity, high in conduct problems, and high in school disengagement.

Kim and colleagues (2005) studied suicidal and self-injurious behaviors among a sample of 1,700 seventh- and eighth-grade students in Korea. Compared with bullies, victims, and students not involved in bullying, those students who were identified as bully-victims were most likely to report suicidal or self-injurious behavior (“I deliberately try to hurt or kill myself”) and suicidal ideation (“I think about killing myself”) (Kim et al, 2005, p. 358).

Finally, authors of at least two recent retrospective studies of violent acts at school (Anderson, Kaufman, Somon, Barrios, Paulozzi, et al., 2001; Fein, Vossekuil, Pollack, Borum, Modzeliski, et al., 2002) have observed that many of the violent youth in their studies had also been bullied. Anderson et al. (2001) speculated that these children “may represent the ‘provocative— or ‘aggressive— victims described in recent studies on bullying behavior, who often retaliate in an aggressive manner in response to being bullied” (p. 2702). The literature clearly identifies this group of children to be at very high risk of a range of troublesome behaviors, which suggests that educators and practitioners should pay particular attention to them in prevention and intervention efforts.

Family Influences on Bullying

Not only do certain characteristics of individual children place them at risk of being involved in bullying, but there also are family characteristics that may increase a child’s likelihood of bullying or of being bullied (Duncan, 2004; Espelage, Bosworth, & Simon, 2000; Olweus, 1993a; Olweus et al, 1999).

Family Characteristics of Children Who Bully

Common characteristics of parents of children who bully include a lack of warmth and involvement, a lack of supervision, and inconsistent but corporal discipline (Duncan, 2004; Olweus, 1993a; Olweus et al, 1999). Children who bully report high negative affect within their families (Rigby, 1993), poor relationships with their parents (Rigby, 1993), and little emotional support (Rigby, 1994). Children who bully their peers are also more likely than other children to be engaged in or exposed to violence within the home. They are, for example, more likely to bully their siblings (Duncan, 1999). In a study involving middle school students in the United States, Duncan (1999) found that 57% of children who bullied their peers and 77% of children who were bully-victims also bullied their siblings, whereas only 38% of bullied children and 32% of children not involved in peer bullying indicated that they bullied their siblings. Children who bully also are nearly twice as likely to have been exposed to domestic violence (Baldry, 2003). Research by Shields and Cicchetti (2001) also reveals that children who were maltreated by a parent are more likely to bully their peers.

Family Characteristics of Children Who Are Bullied

Several researchers have examined familial characteristics of bullied children (Duncan, 2004; Fosse & Holen, 2002; Ladd & Ladd, 1998; Olweus, 1993a; Rigby, 1993, 1994). Although care must be taken not to blame bullied children or their families for abuse that their peers inflict, recognition of familial risk factors may be helpful in prevention and intervention efforts. Somewhat different familial characteristics emerge for bullied boys and girls. Male victims of bullying tend to have close relationships with their mothers (Olweus, 1994; Duncan, 2004). Mothers of bullied boys have been described as being overprotective (Hodges & Perry, 1996; Olweus, 1993a), intense, and overly involved in their sons’ lives (Bowers, Smith, & Binney, 1994). Bullied girls, on the other hand, are more likely to report having negative attitudes toward their mothers (Duncan, 2004; Rigby, 1993), poor family functioning and communication, and low family affect (Rigby, 1994). Mothers of bullied girls have been described in the research literature as being “emotionally abus[ive], hostile, and rejecting of their daughters” (Duncan, 2004, p. 235). Both boys and girls who have been maltreated by parents are more likely to be bullied by their peers (Shields & Cicchetti, 2001). Moreover, boys and girls exposed to domestic violence are more likely to be bullied than those who are not exposed to violence between their parents (Baldry, 2003). Finally, children who are bullied by their peers may be slightly more likely than non-involved children to also be bullied by their siblings. Duncan (1999) found that 36% of victims of peer abuse were also bullied by their siblings, compared with 24% of children who were not involved in peer bullying.

Family Characteristics of Aggressive Victims

Very little research has focused on the family characteristics of bully-victims. Bowers and colleagues (1994) observed that bully-victims report troubled parent-child relationships, low parental warmth, abusive and inconsistent monitoring and discipline, and neglect. Duncan (1999) found that sibling bullying was most common among children who were characterized as bully-victims at school. As Duncan notes, “the majority of bully/victims reported that they were at the receiving and the giving end of bullying with their siblings” (p. 882).

The Peer Context and Bullying

Although adults may tend to view bullying as an aggressive exchange between two individuals (a child who bullies and his or her victim), it is more accurately understood as a group phenomenon (Hanish, Kochenderfer-Ladd, Fabes, Martin, & Denning, 2004; Olweus et al, 1999; Pellegrini, 2002; Salmiavelli, Lagerspetz, Björkqvist, Öster-man, & Kaukianen, 1996) in which children may play a variety of roles as aggressors, victims, observers, and defenders. Olweus et al. (1999) identified at least six distinct roles that children may play in acute bullying situations: bullies (who initiate the bullying), followers (who may take an active part in the bullying but do not initiate it), supporters (who support the bullying but do not take an active role), passive supporters (who like the bullying but do not show open support), disengaged onlookers, possible defenders (who dislike the bullying but do not take action to help), defenders (who try to help the child being bullied), and victims. In a study of sixth graders in Finland, Salmiavelli and colleagues (1996) observed that the vast majority of students (87%) assumed participant roles in bullying situations as rein-forcers, assistants, defenders of the victim, or outsiders. Unfortunately, the important roles of these various bystanders has tended to be somewhat overlooked by researchers and educators (Hazier, 1996).

Students’ Attitudes toward and Responses to Bullying

Bullying appears to thrive in school environments in which there are positive (or at least accepting) attitudes on the part of peers and adults toward aggressive behavior. Therefore, it is important to understand how children view bullying. Most studies have found that children have generally negative views toward bullying and positive or sympathetic responses toward victims of bullying (Baldry, 2004; Rigby & Slee, 1993). For example, Unnever and Cornell (2003) found that 80% of the students in their middle school sample reported feeling sorry for victims of bullying. Sympathy, however, did not translate into action. Most students, 64%, also noted that other students actually try to prevent bullying only “once in a while” or “almost never.” Hazier (1996) notes that bystanders frequently comprise a “silent majority” who may remain on the sidelines because they may be uncertain about how to respond to bullying, are afraid that they might make the situation worse for the victim, or are fearful of becoming targets themselves (p. 19).

Not only do children often fail to take action to stop or prevent bullying, but a disturbing minority of children display attitudes that are supportive of bullying or blaming of the victims of bullying (Baldry, 2004; Unnever & Cornell, 2003). When asked, “When you see a student your age being bullied at school, what do you think or feel?,” 20% of the students surveyed by Unnerver and Cornell admitted either that the bullying did not bother them much or that the victim deserved the bullying. These attitudes may contribute to a culture supportive of bullying at schools and within communities. Prevention and intervention efforts must, therefore, target bystanders as potential agents for change, as deLara argues in this volume.

Risk and Protective Factors Associated with Peer Relations

Not only do peers play a role in influencing a school culture that is generally supportive of or opposed to bullying, but they may also make it more or less likely that individual children are bullied or bully others. Research has confirmed that the number and quality of friends is important in protecting children from possible bullying by peers (Pellegrini, 2002; Pellegrini & Long, 2004). If children affiliate with large numbers of peers, it is less likely that they will be bullied, as bullies may risk disapproval and possible retaliation from the target’s friends (Pellegrini, 2002). Moreover, it appears that affiliation with particular types of peers (those who are rated as popular and strong) may be particularly helpful in preventing bullying (Pelligrini & Long, 2004). Finally, there also are peer-related risk factors associated with bullying behavior among children and youth. Children who bully are more likely to associate with other aggressive or bullying children (Olweus, 1993a).

Characteristics of Schools that Influence Bullying

Not only are bullying rates influenced by characteristics associated with individual children, family units, and peer groups, but they also may be affected by characteristics of the child’s school. One particular variable of interest is the transition from elementary to middle school. Drawing upon data from studies of bullying in five countries, Pellegrini (2002) observed that bullying rates tend to decline across the primary school years, increase with the transition to secondary school, and then resume their decline. He notes that this pattern changes, however, if students remain in the same schools during this age period, in which case rates of bullying decline steadily through elementary, middle, and high school. Pellegrini notes that there are a number of characteristics of middle and junior high schools that may be responsible for this increase in bullying, including their size and impersonal nature, a relative lack of supervision, increased focus on competition and social comparisons among peers, teacher attitudes, and a loss of a sense of community.

Studying students in Norway, Olweus (1993a) did not observe that rates of bullying were explained by competition among students or school size. However, it is not certain that the negative findings related to school size would translate to U.S. schools, which typically are significantly larger and structured differently than corresponding schools in Norway. The structure of U.S. middle and junior high schools is also such that they tend to lack the sense of community, familiarity, and closeness that more commonly characterize elementary school settings (and corresponding schools) in Norway (Limber, 2004). When children transition from elementary school, they typically trade the closeness of a single classroom setting (in which they have a primary teacher and interact regularly with 20-30 peers) for the comparative anonymity of a middle school setting in which they frequently change classrooms, may have five or more teachers, and interact with different cohorts of peers. Whereas the closeness and familiarity of the elementary setting may foster cooperation and minimize aggression among students (Pellegrini, 2002), its absence may contribute to increased bullying. Several bullying prevention efforts in the United States (e.g., Olweus et al., 1999) have encouraged school staff to hold regular classroom meetings in an effort to increase familiarity and cohesion among students. Additional research in the United States is needed to better understand how the structure of middle school settings affects children’s sense of closeness and familiarity, and in turn, rates of bullying.

Not only may middle school settings lack the close sense of community that characterizes elementary schools, but they also may have less direct supervision from adults. Olweus (1993a) and others (Boulton, 1994; Pellegrini & Bartini, 2000; Smith & Sharp, 1994) have documented that rates of bullying are significantly affected by the quality and amount of adult supervision of students, particularly during break times. Boulton (1994) found, for example, that the training of playground supervisors reduced rates of various forms of bullying by 40-50%. The U.S. Department of Education (1998) and the Health Resources and Services Administration (Maternal and Child Health Bureau, 2005) also have emphasized the role of adult supervision in reducing bullying in schools.

In a two-and-one-half-year longitudinal study with more than 500 youth from 250 schools in the United States, Kasen and colleagues (Kasen, Berenson, Cohen, & Johnson, 2004) examined a number of aspects of school climate that may affect rates of bullying in school: (a) conflict (i.e., the extent to which teachers are ineffective in maintaining control and students discount authority); (b) learning focus (i.e., the extent to which schools are well-organized to make learning a priority); (c) social facilitation (“which reflects an open and informal atmosphere where personal concerns may be voiced and social ties encouraged”; and (d) autonomy (in which students are encouraged to have a voice in school politics and decisions). Kasen and colleagues (2004) found that adolescents who attended schools high in conflict and social facilitation issues increased their bullying behaviors during the course of the study. Conversely, schools that were high in learning focus and in granting autonomy appeared to have a positive influence on youth; researchers saw a decrease in bullying-related behaviors among youth who attended schools with these characteristics.

The attitudes and behaviors of teachers and other staff members with regard to bullying also may significantly affect rates of bullying within a school (Limber, 2002; Olweus, 1993a; Pellegrini, 2002). The majority of studies find that most teachers have generally negative attitudes about bullying, are sympathetic toward children who are bullied, and feel a responsibility to prevent bullying (Boulton, 1997). For example, in a study of nearly 800 elementary through high school teachers, Holt and Keyes (2004) noted that 97% agreed that adults in the school “usually stop hurtful comments from students toward other students” (p. 127).

However, teachers fairly consistently perceive less bullying at school than do students (Holt & Keyes, 2004). They also are less likely to accurately identify children as bullies or victims in middle school as opposed to elementary school (Leff, Kupersmidt, Patterson, & Power, 1999). There may be several reasons why teachers are not as aware of bullying as students are. A small minority may discount its seriousness or view bullying as a rite of passage. Others may hold somewhat limited definitions of what constitutes bullying (Holt & Keyes, 2004). For example, Boulton (1997) observed that although 96%of teachers considered hitting, pushing, and kicking to be forms of bullying, one quarter of teachers did not consider name-calling, rumor-spreading, or intimidation (by staring) to be bullying. Approximately half did not view social exclusion as a form of bullying. Pellegrini (2002) also noted that teachers may be especially likely to ignore or fail to detect indirect forms of bullying.

Teachers also may perceive less bullying at school because it often goes unreported by students (Holt & Keyes, 2004; Limber, 2002; O’Moore, 2000; Pellegrini, 2002). For example, studies in England revealed that less than one quarter of those students who had been bullied with some frequency reported the incidents to teachers or other staff at school (Boulton & Underwood, 1992; Whitney & Smith, 1993). Many students likely fail to report bullying because they fear retaliation. Others may not feel that their concerns will be taken seriously by adults or be dealt with effectively. In their survey of middle school youth, for example, Unnever and Cornell (2003) found that 42% of the students believed that their teachers had done fairly little, little, or nothing to counteract bullying. Those students who had been bullied were even less positive about their teachers” actions. Similarly, in a survey of high school students, two thirds of those who had been bullied believed that school personnel had responded poorly to bullying instances, and fewer than 1 in 10 believed that staff handled these problems very well (Hoover et al., 1992). Not only may some teachers fail to recognize or appropriately deal with bullying, but a small minority may also model inappropriate behavior for students. In their survey of school teachers, Holt and Keyes (2004) found that 26% had observed other teachers making fun of the way students talk, look, or act.

Only a handful of researchers have explored teacher characteristics that may be linked to more positive or negative attitudes toward bullying. Disturbingly, Boulton (1997) observed that teachers “sympathy toward victims of bullying decreased with their length of service. Kallestad and Olweus (2003) studied teacher characteristics that predicted teachers” use of bullying prevention strategies in the classroom. They observed that those teachers who viewed themselves, their colleagues, and their schools as important agents for change did more to prevent bullying in their classrooms. Moreover, those who perceived more bullying in their classrooms, who had been bullied themselves as children, or who reported feeling upset and uncomfortable about bullying were more likely to implement components of a bullying prevention program.

Community and Societal Influences on Bullying

Just as a wide range of community and societal influences have been found to affect rates of violence among children and youth (e.g., exposure to community violence and media violence), so too may community and societal factors affect rates of bullying. To date, there has been little research on this topic. However, a recent study by Lee and Kim (2004) explored connections between media violence and bullying among a sample of 648 Korean junior high school students aged 12 to 16 years. These researchers found a direct, positive relationship between exposure to media violence and rates of bullying at school. The more media violence that students viewed, the more bullying the adolescents were involved in. They also found that two variables—anger and contact with delinquent friends—mediated these effects. They speculated that viewing violent scenes primes aggressive thoughts, which increases the probability that a child will respond aggressively to real or imagined provocation. Moreover, the more children are exposed to media violence, the more likely they are to take part in a delinquent group, which increases the likelihood of engaging in bullying.

Conclusions

This chapter began with the recognition that bullying is a common yet complex phenomenon that affects millions of children each year, either directly or indirectly. Although much attention has focused on understanding individual characteristics of children involved in bullying, a growing research literature confirms that bullying is best understood within a social-ecological framework. Future research must further illuminate those aspects of the family, peer, and school environments that both contribute to and protect children from being involved in bullying. Evaluation and dissemination of bullying prevention and intervention strategies that address the multiple known risk and protective factors for bullying must be encouraged. And finally, an understanding of the broader societal or cultural influences on bullying behavior is needed. Although being bullied, harassed, and excluded are common experiences for many school children, we need not and should not accept that they are inevitable experiences.