Attachment and empathy as predictors of roles as defenders or outsiders in bullying interactions,☆☆

https://doi.org/10.1016/j.jsp.2008.06.002Get rights and content

Abstract

Bullying almost always occurs in the presence of peers, most of whom do not intervene to stop it. This study examined the role of attachment, empathy, and gender in predicting the behavior of 105 middle school students who reported being either (a) defenders who actively intervened to stop bullying, or (b) outsiders who were either not involved or stood by passively. It was hypothesized that gender, attachment to mother, attachment to father, and empathy would predict whether children nominated themselves as defenders in bullying situations. A hierarchical logistic regression analysis indicated that these variables together significantly predicted whether children nominated themselves as defenders or outsiders. Gender was not a unique significant predictor. Attachment to mother contributed significant unique variance to the model, but attachment to father did not after accounting for gender and attachment to mother. Empathy contributed significant unique variance to the model even after accounting for all other variables. Results suggest that middle school personnel should openly discuss with students their roles in bullying interactions. In addition, teaching empathy towards others and involving parents may be critical components in comprehensive prevention programs.

Introduction

Bullying in schools has become a problem of increasing concern. Children who bully are at risk for experiencing legal and social problems as adults (Oliver, Hoover, & Hazler, 1994), and victims of bullying also experience difficulties, including high levels of emotional distress, loneliness, peer rejection, and a desire to avoid school (Crick and Bigbee, 1998, Kochenderfer and Ladd, 1996). Although bullying has typically been viewed as an interaction between two individuals (i.e., the bully and the victim), it almost always occurs in the presence of bystanding peers (Lodge & Frydenberg, 2005). Researchers have identified and studied several roles that these bystanding peers play in bullying interactions (Camodeca and Goossens, 2005, Goossens et al., 2006, Salmivalli et al., 1997, Salmivalli et al., 1996). For example, Salmivalli et al. (1996) identified the following roles: (a) reinforcer of the bully (e.g., laughs at the victim, encourages the bully to keep teasing the victim), (b) assistant to the bully (e.g., joins in the bullying when someone else starts it), (c) defender of the victim (e.g., stops the bullying by telling a teacher, comforts or makes friends with the victim), and (d) outsider (e.g., does not know about the bullying or ignores it).

This study explored the extent to which attachment to mothers and fathers and empathy predicted whether a child assumes the role of a defender of the victim or an outsider in bullying interactions. According to the above definition of outsider, the child could either be unaware of the bullying or be present but fail to intervene. Given research indicating that peers witness more than 80% of peer victimization episodes but intervene less than 20% of the time (Atlas and Pepler, 1998, Hawkins et al., 2001), it is likely that many outsiders are present but do not intervene.

The bystander effect has been studied by social psychologists for decades (see Latane & Nida, 1981, for a review), and recent attention has been given to this phenomenon in bullying interactions. Although children often express disgust at witnessing bullying and state that they are interested in helping victims of bullying, they seldom intervene to stop it (Lodge & Frydenberg, 2005; O'Connell, Pepler, & Craig, 1999). This disconnect may be due to fear of becoming the next victim or not having effective strategies to counteract bullying (Lodge & Frydenberg, 2005; O'Connell et al., 1999).

Research on bystanding peers in bullying interactions indicates that reactions depend on individual and social factors. Personal characteristics of bystanders, such as competence and self-restraint, affect their willingness to help (Warden and Mckinnon, 2003, Lodge and Frydenberg, 2005; O'Connell et al., 1999). In addition, children who defend victims of bullying are more likely than bullies, victims, reinforcers, assistants, and outsiders to display the Big Five Personality Trait of agreeableness, a prosocial orientation including trust, cooperation, and altruism (Tani, Greenman, Schneider, & Fregoso, 2003). Children who defend also have higher self-esteem, productive coping styles, and express altruistic feelings (Lodge & Frydenberg, 2005). Other studies have not found differences between defenders and outsiders. For example, Salmivalli and Voeten (2004) found that although anti-bullying attitudes differentiate children who bully others and those who do not, there were no attitudinal differences between defenders and outsiders. Camodeca and Goossens (2005) also found no differences between defenders' and outsiders' opinions about the best way to deal with bullying. Defenders and outsiders both reported favoring strategies to resolve the conflict (e.g., assertiveness), in contrast to bullies, who favored retaliation.

Children's tendency to ignore, help, or join in bullying interactions also depends on how members of their social network react (Salmivalli et al., 1997). One-fifth of children and adolescents reported that they might join bullying when their friends bully others (Whitney & Smith, 1993), which may result from peer pressure (Espelage & Asidao, 2002), a desire to be accepted by peers (Rogers & Tisak, 1996), or avoidance of being teased themselves (Cowie & Sharp, 1994). Bystanders are more likely to offer help to victims who are in-group members (Levine, Cassidy, Brazier, & Reicher, 2002) or friends (Lodge & Frydenberg, 2005). Unfortunately, victims of bullying tend to form their social networks with other victims (Salmivalli et al., 1997), decreasing the chance that bystanding peers will intervene.

Although both individual and relational factors have been associated with children's roles in bullying interactions, scant research exists on the role of attachment to parents and empathy in children's tendency to be defenders or outsiders. Theorists and researchers have posited that parental responsiveness and children's empathy are integral components in the development of caring and empathy-related responding (Chase-Lansdale et al., 1995, Zhou et al., 2002).

There is a growing recognition that bullying interactions should be viewed from an ecological perspective that considers familial, peer, school, and community factors (Espelage & Swearer, 2003). Because parents are the primary agents of socialization, there is great interest in the role of parents in children's interactions with peers (Christie-Mizell, 2003, Zahn-Waxler and Radke-Yarrow, 1990).

Attachment theory is a well-accepted, empirically supported developmental theory that explains the nature of the parent–child bond (Bowlby, 1958, Hazan and Shaver, 1994). Originally applied to infants and young children, attachment behaviors include seeking contact with the primary caregiver (typically the mother), turning to the caregiver for comfort, support, and reassurance, and using the caregiver as a base from which to explore (Bowlby, 1958). Young children with secure attachment patterns are more likely than anxiously attached children to display competent, self-confident, and socially skilled behavior years later (Elicker, Englund, & Sroufe, 1992). Attachment theory has been applied to relationships throughout the lifespan (Hazan & Shaver, 1994), but recent emphasis has been placed on the importance of attachment in late childhood and early adolescence (Nickerson & Nagle, 2005). Indeed, attachment security is related to friendship quality (Lieberman, Doyle, & Markewicz, 1999) and intimacy with friends (Updegraff, Madden-Derdich, Estrada, Sales, & Leonard, 2002) in early adolescence. The proportion of variance in these indexes of peer relationships that is explained by attachment security, however, tends to be small (Updegraff et al., 2002). In addition, attachment security has not been shown to relate to children's popularity (Lieberman et al., 1999) or to the numbers of their peer enemies (Card & Hodges, 2003). These findings raise the question of whether the link between attachment security and peer relationships is restricted only to close friends as opposed to interactions with the larger peer group.

Traditionally, attachment has been conceptualized as the mother–child bond, but there is a growing interest in the father–child relationship and its influence on peer relationships. A number of factors are associated with security in the father–infant attachment, such as personality characteristics of fathers (e.g., extraversion and agreeableness), marriage quality, and harmony across work and home (Belsky, 1996). In addition, fathers' interactions with infants, their attitudes about making it a priority to spend time with children, and their describing their infant in positive terms predicted later attachment security (Cox, Owen, Henderson, & Margand, 1992). Although children and adolescents tend to report being more securely attached to mothers than fathers (Williams & Kelly, 2005), a secure father–child relationship is predictive of positive peer relationships (Youngblade & Belsky, 1992) and fewer behavioral problems (Williams & Kelly, 2005). Some findings suggest that a secure attachment to fathers, but not mothers, has been related to children's prosocial behavior (Kerns & Barth, 1995) and decreased conflict with peers (Ducharme et al., 2002, Lieberman et al., 1999).

Attachment has been found to relate to children's roles as victims and bullies (Nickerson, Mele, & Osborne, in press). Insecure attachment, characterized by emotionally intense, positive interactions, and overprotection, is associated with increased risk for victimization in studies using diverse methodologies, such as observations (Ladd & Ladd, 1998), child self-report (Bowers et al., 1994, Finnegan et al., 1998), and parent interviews (Olweus, 1991). The father–child relationship appears to play a critical role in victimization. Perceived paternal support is related to reduced victimization and rejection (Rubin et al., 2004), and a positive father–son relationship is a protective buffer for adolescents experiencing peer victimization (Flouri & Buchanan, 2002). Research has indicated that the large majority of children who bully peers are insecurely attached, often displaying an avoidant attachment style (Monks et al., 2005, Troy and Sroufe, 1987). Bullies tend to have an absence of a father at home, lower family cohesion, and more disengaged relationships with parents than do victims and children who are neither bullies nor victims (Berdondini and Smith, 1996, Bowers et al., 1994). Research has also linked parental conflict to bullying, although children's self-concept mediates this relationship (Christie-Mizell, 2003).

Empathy has been defined in many ways, but it refers broadly to the reactions of an individual to the experiences of another (Davis, 1983). Empathy is commonly described as consisting of perspective taking, empathic concern, and personal distress (Davis, 1983, Olweus and Endresen, 1998). Empathic concern, the affective response of feeling concern for another and wanting to alleviate distress, is related to altruistic responding in adults (Batson et al., 1987, Coke et al., 1978).

It has been argued that the connection between affective empathy and the positive actions that children show in response to victims in distress (e.g., helping and comforting) is an important area of study (Zahn-Waxler & Radke-Yarrow, 1990). Empathy has been related to social functioning (Zhou et al., 2002) and prosocial behavior (Eisenberg and Miller, 1987, Strayer and Roberts, 2004) in children, and this relationship becomes more evident during early adolescence (Espelage, Mebane, & Adams, 2004). Interestingly, early adolescence is also the time when bullying peaks (Olweus, 1993). Because bullying can result in such a high degree of emotional distress in victims (Boivin, Hymel, & Bukowski, 1995), it is possible that a peer experiencing empathy towards a child in distress may be motivated to do something to stop the bullying.

There is a surprising lack of research on the relationship between empathy and bullying, although teaching empathy is a critical component of many social cognitive interventions to reduce bullying and other forms of aggressive behavior (Manger, Eikeland, & Asbjornsen, 2001). Empathy has been shown to be inversely related to antisocial behavior (see Miller & Eisenberg, 1988, for a review). Children with conduct disorder score lower than comparison youth on measures of affective and cognitive empathy (Cohen & Strayer, 1996). In a preliminary investigation, Espelage et al. (2004) found an inverse association between empathy and bullying. In addition, they found that victims demonstrated more caring behaviors than bullies and students who were not involved as bullies or victims. In a study of empathy in prosocial children, bullies, and victims, Warden and Mckinnon (2003) found that children nominated as prosocial by peers were more likely than peer-nominated bullies or victims to report empathic awareness, although when gender was covaried in the analysis, the effect of role was not significant.

Some interesting gender differences have emerged in the study of empathy. Although girls have consistently been found to demonstrate more empathy than boys (Olweus and Endresen, 1998, Roberts and Strayer, 1996, Warden and Mckinnon, 2003), both genders reported more empathic concern for a girl in distress than for a boy in distress (Olweus & Endresen, 1998). Interestingly, Olweus and Endresen (1998) found boys' empathy towards girls to increase with age, whereas empathy towards boys decreased with age for both genders. Research has yielded contradictory results with regard to gender differences in empathy and prosocial behavior. Roberts and Strayer (1996) found that the relation between empathy and prosocial behavior was stronger for boys than girls (Roberts & Strayer, 1996), whereas Warden and Mckinnon (2003) found empathy and prosocial behavior to be correlated for girls, but not for boys. It has been suggested that females are less likely than males to engage in violent behavior due to empathy; in adolescence, males engage in higher levels of offending while females have higher levels of empathy (Broidy, Cauffman, Espelage, Mazerolle, & Piquero, 2003). Clearly, the role of empathy in children's interactions with peers is an area worthy of study. Seeking greater clarity about the role of empathy in children's defending victims of bullying is important to understanding the possible motivators behind this behavior, which may have implications for intervention.

The primary objective of this study was to examine the role of attachment and empathy for early adolescents who defend victims of bullying compared to those who are outsiders. Because gender differences have been found in bullying roles (Warden & Mckinnon, 2003) and empathy (Olweus and Endresen, 1998, Roberts and Strayer, 1996), this variable was also included in the analyses. Sixth, seventh, and eighth grade students were studied because bullying occurs most frequently in middle school (Olweus, 1993), and the link between empathy and prosocial behavior is hypothesized to become more evident in early adolescence (Espelage et al., 2004). It was hypothesized that attachment to mother, attachment to father, and empathy would predict whether children nominated themselves as defenders in bullying situations.

Section snippets

Participants, setting, and recruitment

Parents of all students (approximately 498) in a middle school in a suburban area in the northeastern United States were sent consent forms via their children's social studies classes. The school was in a predominantly middle-class area, with approximately 10% of children from economically disadvantaged backgrounds. Although no formal bullying programs were in place, the school had a demonstrated commitment to bullying prevention and intervention, evidenced by a recent in-service training on

Kerns' Attachment Security Scale (Kerns, Klepac, & Cole, 1996)

This 15-item self-report scale is designed to obtain children's perceptions of a particular attachment relationship. One version is completed to represent attachment to mother, and another version is completed to represent attachment to father. Items assess a child's belief that a parent is responsive and available, open to communication, and a reliable source of help and comfort when needed. It was developed for children 10 to 12 years of age, but it has also been used with children as young

Procedure

Upon obtaining consent and assent, a trained research assistant administered the mother and father versions of the Kerns' Attachment Security Scale, the ERQ, and the PRD to small groups (approximately 10–20 students) in an unused classroom during their resource period. The research assistant read the assent form and the instructions for each assessment instrument, including the sample item for the Kerns' Attachment Security Scale. The order of the instruments was counterbalanced to minimize

Results

Less than 5% of the cases had incomplete data that resulted from omitting an item (or items) on an instrument. For these individuals, we computed a scale score by averaging their available items. For example, if 1 item was missing on the 15-item Kerns' Attachment Security Scale, the individual's other 14 responses were summed and divided by 14 to arrive at the mean. Prior to conducting the main analyses, data for all 105 participants were screened for accuracy of data entry, missing values,

Discussion

Just over half of the children in this study (i.e., 52%) reported that they most often defended victims in bullying interactions. This finding is higher than that found in a similar study by Jeffrey, Miller, and Linn (2001), where 23% of boys and 32% of girls reported being defenders, and 41% of males and 48% of females reported being outsiders. In the Jeffrey et al. study, these terms were not defined, which may have led some children to not consider themselves as defenders if they engaged in

References (83)

  • BowlbyJ.

    The nature of the child's tie to his mother

    International Journal of Psychoanalysis

    (1958)
  • BrockS.E. et al.

    Understanding children victimized by their peers

    Journal of School Violence

    (2006)
  • BroidyL. et al.

    Sex differences in empathy and its relations to juvenile offending

    Violence and Victims

    (2003)
  • CamodecaM. et al.

    Children's opinions on effective strategies to cope with bullying: The importance of bullying role and perspective

    Educational Research

    (2005)
  • CardN.A. et al.

    Parent–child relationships and enmity with peers: The role of avoidant and preoccupied attachment

    New Directions for Child and Adolescent Development

    (2003)
  • CedarR.B. et al.

    A meta-analysis of the effects of Parent Effectiveness Training

    American Journal of Family Therapy

    (1990)
  • Christie-MizellC.A.

    Bullying: The consequences of interparental discord and child's self-concept

    Family Process

    (2003)
  • CohenD. et al.

    Empathy in conduct disordered and comparison youth

    Developmental Psychology

    (1996)
  • CokeJ.S. et al.

    Empathic mediation of helping: A two-stage model

    Journal of Personality and Social Psychology

    (1978)
  • Conduct Problems Prevention Research Group

    Initial impact of the fast track prevention trial for conduct problems: II. Classroom effects

    Journal of Consulting and Clinical Psychology

    (1999)
  • CornellD.G. et al.

    Assessment of bullying

  • CowieH. et al.

    Tackling bullying through the curriculum

  • CoxM.J. et al.

    Prediction of infant–father and infant–mother attachment

    Developmental Psychology

    (1992)
  • CrickN.R. et al.

    Relational and overt forms of peer victimization: A multiinformant approach

    Journal of Consulting and Clinical Psychology

    (1998)
  • DavisM.H.

    Measuring individual differences in empathy: Evidence for a multidimensional approach

    Journal of Personality and Social Psychology

    (1983)
  • DucharmeJ. et al.

    Attachment security with mother and father: Associations with adolescents' reports of interpersonal behavior with parents and peers

    Journal of Social and Personal Relationships

    (2002)
  • EisenbergN. et al.

    The relation of empathy to prosocial and related behaviors

    Psychological Bulletin

    (1987)
  • ElickerJ. et al.

    Predicting peer competence and peer relationships in childhood from early parent–child relationships

  • EspelageD.L. et al.

    Conversations with middle school students about bullying and victimization: Should we be concerned?

  • EspelageD.L. et al.

    Research on school bullying and victimization: What have we learned and where do we go from here?

    School Psychology Review

    (2003)
  • EspelageD.L. et al.

    Empathy, caring, and bullying: Toward an understanding of complex associations

  • FeshbachN.D.

    Learning to care: A positive approach to child training and discipline

    Journal of Clinical Child Psychology

    (1983)
  • FinneganR.A. et al.

    Victimization by peers: Associations with children's reports of mother–child interaction

    Journal of Personality and Social Psychology

    (1998)
  • FletcherA.C. et al.

    Strategies for obtaining parental consent to participate in research

    Family Relations

    (2003)
  • FlouriE. et al.

    Life satisfaction in teenage boys: The moderating role of father involvement and bullying

    Aggressive Behavior

    (2002)
  • GoossensF.A. et al.

    New participant role scales: Comparison between various criteria for assigning roles and indications for their validity

    Aggressive Behavior

    (2006)
  • HawkinsD. et al.

    Naturalistic observations of peer interventions in bullying

    Social Development

    (2001)
  • HazanC. et al.

    Attachment as an organizational framework for research on close relationships

    Psychological Inquiry

    (1994)
  • HosmerD.W. et al.

    Applied logistic regression

    (2000)
  • JeffreyL.R. et al.

    Middle school bullying as a context for the development of passive observers to the victimization of others

    Journal of Emotional Abuse

    (2001)
  • KazdinA.E.

    Problem-solving skills training and parent management training for conduct disorder

  • Cited by (192)

    • School climate and bullying bystander responses in middle and high school

      2022, Journal of Applied Developmental Psychology
    View all citing articles on Scopus

    We are grateful to the students, parents, and school whose participation made this work possible. We also thank Josh Blumkin, Carlo Cuccaro, and Evan Slater for their assistance with data collection.

    ☆☆

    Action Editor: Randy Floyd.

    View full text