A Relational Framework for Understanding Bullying
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
A Relational Framework for Understanding Bullying
Developmental Antecedents and Outcomes
Philip C. Rodkin and Dorothy L. Espelage University of Illinois, Urbana–Champaign
Laura D. Hanish Arizona State University
This article reviews current research on the relational problems (Dodge, Coie, & Lynam, 2006). What makes
processes involved in peer bullying, considering develop- aggression adaptive is also well documented (Rodkin &
mental antecedents and long-term consequences. The fol- Wilson, 2007). Aggression is often successful in changing
lowing themes are highlighted: (a) aggression can be both other’s behavior and can be used to acquire resources and
adaptive and maladaptive, and this distinction has impli- maintain group boundaries.
cations for bullies’ functioning within peer social ecolo- The bifurcated adaptive/maladaptive nature of aggres-
gies; (b) developmental antecedents and long-term conse- sion makes its way into two basic issues regarding bullying
quences of bullying have not been well-distinguished from perpetration: psychological processes, or what drives some
the extant research on aggressive behavior; (c) bullying is children to bully others, and concurrent correlates, namely
aggression that operates within relationships of power and the social and personality characteristics of youth who
abuse. Power asymmetry and repetition elements of tradi- bully. Descriptions of why youth bully, or of the kinds of
tional bullying definitions have been hard to operational- youth who bully, often include either or both of two dis-
ize, but without these specifications and more dyadic mea- cordant elements: a functional element, because the bully is
surement approaches there may be little rationale for a in a superior position of dominance with the ability to
distinct literature on bullying—separate from aggression. coerce, and a dysfunctional element, because bullying is
Applications of a relational approach to bullying are pro- morally outrageous and chronic aggression a serious risk
vided using gender as an example. Implications for future factor for maladjustment.
research are drawn from the study of relationships and
interpersonal theories of developmental psychopathology. Psychological Processes: Why Bully?
Keywords: bully, aggression, bully prevention Why would anyone want to bully others? Guerra, Williams,
and Sadek (2011) posed this question to 14 focus groups of
W hich youth bully others? What are the factors
that lead them to behave negatively and ag-
gressively toward others, picking on the vul-
nerable for their own advantage? What are the develop-
mental consequences for youth who engage in bullying
115 elementary, middle, and high school students. Across
this wide age span, half of the focus groups said that youth
who bully were integrated into the school environment,
with high self-esteem and a desire to demonstrate social
prowess. However, just as many mentioned that youth who
behavior? In this review, we consider psychological corre- bully were marginalized and had emotional problems, such
lates of bullying behavior, the developmental antecedents as low self-esteem and prior victimization. We will exam-
that predict bullying, and the adjustment outcomes that ine these perspectives in turn, one echoing a theme of
flow from it. Throughout, we consider perspectives that social accomplishment and the other a theme of relational
stress the functionality of aggressive behavior, and the difficulties.
distinction between aggression and bullying. Ultimately,
that distinction might best be considered from a relational
perspective, because bullying is aggression within the con- Editor’s note. This article is one of six in the “School Bullying and
Victimization” special issue of the American Psychologist (May–June
text of a relationship of abuse. We consider some ramifi-
2015). Susan M. Swearer and Shelley Hymel provided the scholarly lead
cations of this relational view of bullying perpetration and for the special issue.
propose a relational approach to understanding the rich
social context which supports and influences bullying be-
Authors’ note. Philip C. Rodkin and Dorothy L. Espelage, Department
haviors. of Educational Psychology, University of Illinois, Urbana-Champaign;
Laura D. Hanish, T. Denny Sanford School of Social and Family Dynam-
The Duality of Aggression and ics, Arizona State University.
Bullying: A Relational Perspective on Philip C. Rodkin died on May 6, 2014, during the final stages of the
Adaptation and Maladaptation editorial process of this article. We acknowledge his major contribution to
the article and his compassionate leadership in advancing understanding
The paradox of aggressive behavior is that it is both adap- of the social dynamics of adolescence.
Correspondence concerning this article should be addressed to Dor-
tive and maladaptive. What makes aggression maladaptive othy L. Espelage, College of Education, University of Illinois at Urbana–
is all too obvious: Aggressive children endanger others and Champaign, 1310 South 6th Street, Champaign, IL 61820. E-mail:
are themselves at risk for a host of serious adjustment espelage@illinois.edu
May–June 2015 ● American Psychologist 311
© 2015 American Psychological Association 0003-066X/15/$12.00
Vol. 70, No. 4, 311–321 http://dx.doi.org/10.1037/a0038658from poor self-regulation and dysregulation between ap-
proach and avoidance motivational systems (Carver, John-
son, & Joormann, 2008). Youth with an overactive ap-
proach orientation and an underactive avoidance
orientation might disregard social norms or potential con-
sequences of their actions to pursue their own self-interest,
increasing the likelihood of aggressive behavior. This sug-
gests, as with Moffitt’s (1993) early starter (life-course
persistent) pathway to aggression, that temperamental def-
icits associated with attention problems, hyperactivity, and
irritability can be expressed through negative peer interac-
tions. These negative temperamental qualities promote ag-
gressive behavior which in turn may lead to the develop-
ment of social– cognitive biases that include viewing
aggressive behavior as effective, endorsing positive social
norms for aggression, and moral disengagement. All of
these allow for greater likelihood of aggression (Hymel,
Schonert-Reichl, Bonanno, Vaillancourt, & Henderson,
2010). The result is a developmental cascade reflecting the
negative cumulative continuity entailed in repeatedly ag-
Philip C. gressing and being aggressed against (Rudolph, Lansford,
Rodkin & Rodkin, in press).
Finally, it is not possible to fully understand the psy-
chological mechanisms that lead a child to bully without
considering who it is that the bully is targeting, and why.
Aggressive behavior has deep phylogenetic roots and, Bullying and aggression are dyadic or group phenomena
therefore, an aspect of functionality (Ellis et al., 2012). The (Card & Hodges, 2010; Espelage, Holt, & Henkel, 2003).
functionality of aggression is part of the allure of bullying. Goals and related cognitions guide behavior within specific
Youth bully to get what they want and to control the contexts, such as peers who are liked versus disliked.
behavior of others. Peer social structures exist in which Hostile attributional biases, for example, are situationally
some members are more popular, more powerful, and more primed toward enemies, and are less likely to occur toward
likely to control resources than other members. Children friends (Peets, Hodges, & Salmivalli, 2010). This can trans-
who prioritize public demonstrations of their social power, late into a bully’s choice of target. To illustrate, Veenstra,
popularity, and status increase their aggressive behavior Lindenberg, Munniksma, and Dijkstra (2010) find that pre-
over time (Rodkin, Ryan, Jamison, & Wilson, 2013). Chil- adolescent bullies target rejected, unpopular peers whom
dren who bully encourage group members to rally around classmates will support harassing. Cognitive-affective pro-
them, to define themselves by scapegoating their targets, cessing structures also help explain individual differences
and to maintain cohesion and identity for their clique (Faris in why some youth target gender-atypical peers to bully
& Felmlee, 2011; Juvonen & Galván, 2009). Individuals (Pauletti, Cooper, & Perry, 2014; Poteat & Espelage,
may gain and maintain status by using aggression, alone or 2005). We will elaborate these points later when discussing
in combination with more pro-social behavior (Hawley, the value of an etiological relational approach to bullying.
Stump, & Ratliff, 2011). Children with moderate positions
on the social status system and children who are attempting Concurrent Correlates: Two Types of Bullies?
to increase their status may be motivated to proactively A person-oriented version of the distinction between more
aggress against others to establish their social position and and less adaptive patterns of bullying was captured by
to push others down the social hierarchy. Olweus (1978), who distinguished children who perpetrate
Yet, aggressive behavior is perhaps the clearest risk bullying (“bullies”) from children who both perpetrate and
factor for poor adjustment and psychopathology over time are victims of harassment (“bully victims”; see Swearer &
(Dodge et al., 2006). Aggressive behaviors trigger peers to Hymel, 2015, this issue). According to Olweus (1978),
respond with aggression of their own (Hanish, Sallquist, neither bullies nor bully victims were completely free of
DiDonato, Fabes, & Martin, 2012). While those who are mental health problems, but bullies were generally more
most effective in using aggression to gain power may tend functional, more likely to use proactive aggression, and
to rise in the hierarchy, others fail. Aggressive children, more likely to have an extensive social network than bully
particularly those who are also rejected and harassed, have victims, who were more likely to reactively aggress and
a unique motivational stance to interpersonal relationships, show troubling risk patterns across virtually all adjustment
characterized by frustration, hostile attributional biases in indicators.
the face of ambiguity (Crick & Dodge, 1994), and retalia- Besides the co-occurring presence of bullying perpetra-
tion goals in the face of social obstacles (Troop-Gordon & tion and victimization, there are other ways to capture distinc-
Asher, 2005). Impulsive, reactive aggression may stem tions between aggressors with higher and lower levels of
312 May–June 2015 ● American Psychologistmay have lower physiological reactivity to stressful situa-
tions than aggressive children who are less well integrated
in the social sphere (Kliewer, Dibble, Goodman, & Sulli-
van, 2012). Socially integrated bullies are both underrec-
ognized as seriously aggressive, and popularized in the
media. Vaillancourt, McDougall, Hymel, and Sunderani
(2010) go so far as to call these socially connected youth
“Machiavellian”: “popular, socially skilled and competent
. . . [with] high self-esteem . . . low on psychopathology . . .
[and] many assets” (p. 218; see also Hawley et al., 2011).
Hawley et al.’s (2011) own view seems to suggest that
bistrategic controllers (who employ both prosocial and
coercive strategies to gain resources) are aggressive, but
may not be bullies given that they aggress upon their “inner
circle” more than they aggress upon low status peers.
However, not all youth who bully are socially inte-
grated; instead, some are socially marginalized. For these
children, the source of their power is more difficult to
identify. It may come from their unsuccessful attempts to
raise their own social status or to attempt to dominate
Dorothy L. others through coercive strategies (Hawley et al., 2011). It
Espelage may also come from their efforts in bringing together a
small group of equally marginalized youth to support their
attacks. A peer rejection framework is a compelling ex-
planatory framework for marginalized youth who bully,
social functioning. The extant literature on bullying varies who are in conflict with others in the world. These children,
dramatically in how bullies are portrayed: as adaptive Machi- mostly boys, tend to be characterized by a clear pattern of
avellians (e.g., Ellis et al., 2012; Faris & Felmlee, 2011; deficits in broad domains of developmental functioning.
Hawley et al., 2011) or as maladjusted and prone to conduct Their aggression is impulsive and overly reactive to real or
and mental health problems (e.g., Cook, Williams, Guerra, perceived slights; they are consistently identified as “at-
Kim, & Sadek, 2010; Farrington, Ttofi, & Lösel, 2011). From risk” students (Cook et al., 2010). Cook et al.’s meta-
the standpoint of a more relational approach to bullying, a analysis concluded that this type of bully “has comorbid
compelling taxonomy is found in Farmer and colleagues’ externalizing and internalizing problems, holds signifi-
(2010) description of “two social worlds” of bullying: social cantly negative attitudes and beliefs about himself or her-
integration on the one hand and social marginalization on the self and others, is low in social competence, does not have
other. Socially integrated bullies “may use aggression to con- adequate social problem-solving skills, performs poorly
trol” others, while socially marginalized bullies “may be fight- academically, and is not only rejected and isolated by peers
ing against a social system that keeps them on the periphery” but also negatively influenced by the peers with whom he
(p. 386). Whatever the specific distinction used, there is typ- or she interacts” (p. 76). Farmer et al. (2010) reported that
ically one type of youth who bully that seems to have average marginalized, unpopular youth who bully, whether girls or
adjustment outcomes, and another type that seems to suffer boys, are often shunted into peer groups with other bullies,
from very poor adjustment outcomes. We now discuss the and sometimes even with the children they harass. In sum,
personality and behavioral characteristics of these bullying marginalized bully victims have a host of problems of
types and conclude with a brief comment about the psycho- which bullying behavior is but one manifestation. Their
logical validity of this taxonomic structure. bullying might stem from an inability to control their
Some youth who bully are well-integrated into their hostile actions, or from a desire to gain a preferred status
peer culture and do not lack for peer social support (Farmer that generally eludes them. These youth would benefit from
et al., 2010); they have surprisingly high levels of popu- services that go beyond bullying-reduction programs such
larity among their peers (Olweus, 1978). These youth are as more general aggression reduction therapies (e.g., Gold-
relatively evenly split between boys and girls. They have a stein et al., 1986) and social skills training (Cook et al.,
variety of friends that engage in varying degrees of bully- 2010).
ing and possess strengths that are easy to recognize, in- The distinction between bullies and bully victims, or
cluding social skills, athleticism, and/or attractiveness. between socially integrated and socially marginalized bul-
These youth tend to use proactive and goal-directed ag- lies, is an informal heuristic, useful for communicating the
gression (for more, see Rodkin, 2011). Some incorporate considerable variability in social functioning among youth
prosocial strategies into their behavioral repertoire, for who display aggressive and bullying tendencies. Bullying
example reconciling with their targets after conflict, or classifications have not been subjected to the level of
becoming less aggressive once a clear dominance relation- statistical scrutiny as have taxonomic classifications such
ship has been established (Pellegrini et al., 2010). They as Moffitt’s (1993) distinction between adolescence-lim-
May–June 2015 ● American Psychologist 313amination of developmental antecedents of bullying. There
are many antecedents to aggressive behavior and its phys-
ical, verbal, and social or relational forms, and these have
been extensively reviewed (e.g., Dodge et al., 2006). Ag-
gressive behavior has a trait-like quality, as aggression is
among the most temporally stable of all human character-
istics (Dodge et al., 2006; Huesmann, Eron, Lefkowitz, &
Walder, 1984; Olweus, 1978). At the same time, aggression
is quite responsive to situational factors such as being
reared in isolation, or settings in which violence is already
high or valued, or, of course, acute conditions of perceived
threat. Aggressive and bullying behaviors operate accord-
ing to some of the cardinal principles of developmental
psychopathology. One such principle, concerning the in-
tersection of normal and atypical behavior, reminds us that
the aggressive behavioral system has deep phylogenetic
roots, is highly evolved, and has functional properties (Ellis
et al., 2012). Another principle, equifinality, suggests that
multiple pathways exist to problematic patterns of bullying
and aggressive behavior (for more, see Rudolph et al., in
Laura D. press).
Hanish There may be as many developmental antecedents to
bullying as there are to aggressive and antisocial behavior.
Yet, the literature focusing explicitly on antecedents to
ited and life-course persistent antisocial behavior, which bullying (net of aggression more generally) is incredibly
indeed may show extensive overlap with bully and bully sparse. We will focus on relational antecedents, including
victim (or socially integrated and marginalized) categories. (a) poor home life and (b) less widely considered, the peer
Future research will have to be devoted to the issue of social ecology. Hawley et al.’s (2011) jingle fallacy, where
whether bullying behavior is better described in terms of the constructs of aggression and bullying are used inter-
person-oriented configurations versus a relational frame- changeably, unfortunately suffuses all these cases. Studies
work. are lacking that distinguish developmental pathways of
Relational Antecedents and bullying behavior from aggressive behavior, or which iden-
tify developmental antecedents of bullying that do not also
Consequences of Bullying Behavior hold for aggression. Although we do not devote sustained
All bullying is aggressive, but not all aggression is bully- attention to the distinctive subtypes of physical, verbal, and
ing. The overlap between aggression and bullying invites social or relational bullying, we would expect antecedents
the critical question of distinctiveness: What is new in to be very similar to those identified in discussions of
bullying research that has not already been concluded in the physical, verbal, and social or relational aggression (Crick
voluminous literature on child and adolescent aggression? & Zahn-Waxler, 2003) and to overlap with aggression
What new principles or methods are needed to account for more generally.
what makes bullying unique and distinct from aggressive Antecedents in family history. The psycho-
behavior? Without conceptual and empirical differentia- logically impoverished home life found for many children
tion, there is little reason other than semantics and faddism who bully is concordant with the more general literature on
to tailor antibullying interventions anew when more gen- aggressive behavior, which points to risk factors of insen-
eral aggression-reduction programs are on the shelf (see sitive parenting and coercive cycles of parent– child inter-
also Cook et al., 2010). What Hawley et al. (2011) refer to action (Dishion & Tipsord, 2011; Patterson, Littman, &
as the jingle fallacy is worth considering: Perhaps bullying Bricker, 1967), along with other family hardships such as
and aggression really are different, but current methodol- low socioeconomic status, family instability, family dys-
ogies simply fail to measure and take account of those function, and child maltreatment (Copeland, Wolke, An-
differences. We frame our discussion of developmental gold, & Costello, 2013). Coercive exchanges co-occur with
antecedents and consequences of bullying behaviors from harsh parental discipline and conflictual family dynamics,
the value-added approach vis á vis aggression more gen- and all are associated with later bullying (Espelage, Low,
erally, and where available, we note evidence that would Rao, Hong, & Little, 2013). In coercive and hostile rela-
distinguish developmental pathways for socially integrated tionships with caregivers and exposure to family violence,
and marginalized youth who engage in bullying behaviors. children who are maltreated may come to expect that
aggression is fundamental to interpersonal relationships,
Antecedents of Aggression and Bullying affecting children’s expectations (or possibly, internal
The challenge of differentiating bullying from aggressive working models) of new social interactions and guiding
behavior more generally is immediately evident upon ex- their behavioral responses (Rudolph et al., in press). To
314 May–June 2015 ● American Psychologistillustrate, Baldry (2003) found that Italian elementary and beyond. Unfortunately, there was no independent measure
middle schoolchildren, particularly girls, who were ex- of bullying in the Study of Early Child Care and Youth
posed to violence at home were involved in bullying. Girls Development to determine the extent of overlap between
who witnessed a father’s violence against the mother or a popular-aggressive and bully, and nonpopular-aggressive
mother’s violence against the father were among the most and bully victim categories, and neither was there the
likely to bully others, compared to girls who did not wit- ability to control for concurrent aggression and bullying.
ness interparental violence. Summary. The research literature in the area of
Thus, as is well known, conflict and violence in the developmental antecedents is at once voluminous (when
home is a training ground for bullying and aggression with considering aggression) and sparse (when considering bul-
peers (Dishion & Tipsord, 2011; Dodge et al., 2006). lying). Current study designs on developmental anteced-
Although progress has been made in establishing relations ents of bullying are subject to the jingle fallacy (Hawley et
between family dysfunction (e.g., maltreatment, exposure al., 2011): The discriminant validity of bullying is lacking
to domestic violence, sibling aggression) and bullying, with respect to aggressive behavior or any of its subtypes,
there is still a need for closer, empirical investigation of and as a result, findings are ambiguous. Until such research
mechanisms that link the two. Further, research needs to accumulates, conclusions regarding the social functioning
consider bullying as a mediator of other risky behaviors. of youth who bully must be qualified as suggesting that
For example, Espelage and colleagues (2013) found that bullying perpetration may be incidental to underlying is-
both bullying and fighting mediated the association be- sues with aggressive and antisocial behavior.
tween family violence and alcohol and drug use in a lon-
Long-Term Consequences of
gitudinal study, but only for boys, not girls. This research
Bullying Behaviors
calls attention to the importance of examining bullying
behaviors among youth as mediators or even antecedents to Incidental and causal models of later ad-
other adverse, yet malleable, outcomes, apart from aggres- justment outcomes. The distinction between aggres-
sion more generally. sion and bullying is also central to the question of long-
Antecedents in the peer social ecology. term consequences for youth who bully. Are long-term
Rodkin and Roisman (2010) investigated developmental effects of bullying to be ascertained relative to aggressive
antecedents of popular-aggressive children in the National children who do not bully (if there are such individuals) or
Institute of Child Health and Human Development Study of to nonaggressive, nonbullying children? The latter compar-
Early Child Care and Youth Development. This study ison is not so interesting, because it is well-known that
examined whether children identified as popular-aggres- sustained aggressive behavior increases risk and reduces
sive over third to sixth grades could be differentiated from life opportunities (Dodge et al., 2006). Possibly the most
unpopular-aggressive and other children on the basis of useful way to consider long-term adjustment outcomes for
early maternal sensitivity, high cognitive functioning, or bullying perpetration is through the distinction between
quality and quantity of early child care. The hypothesis, incidental and causal models that Parker and Asher (1987)
following Bronfenbrenner’s (1979, p. 203) Hypothesis 23, put forth in the context of peer rejection. Parker and Asher
was that extensive time within early childhood peer ecol- (1987) asked, does being rejected make a child more likely
ogies would promote the social status of aggression, and to become depressed, aggressive, or to flounder in school,
that once established, linkages between popularity and ag- or is rejection simply a marker for preexisting third vari-
gression would endure over the elementary years. Indeed, ables (e.g., the qualities that make one rejected in the first
controlling for many other factors, including child care place)?
quality, popular-aggressive third to sixth graders had eight Adapting the Parker and Asher (1987) framework to the
to 11 more hr per week of child care up to age 4.5 years experience of bullying perpetration, an incidental model
than all other children. Popular-aggressive children were would frame bullying as epiphenomenal or tangential to early
viewed as socially skilled and aggressive by disinterested risk and later emotional and behavioral maladjustment. In an
observers who focused on peer interaction quality, but were incidental model, bullying would be indicative of risk, but
viewed as not socially skilled by adults, such as parents and there would be no prediction that children who bully for
teachers, who may have a frustrating history of trying to reasons other than some underlying disorder would suffer
guide these children toward pro-social behavior. Regard- from later maladjustment. In contrast, causal models presume
less of popularity levels, aggressive children had very low that early peer interactions, such as bullying perpetration,
levels of maternal sensitivity and relatively poor cognitive make active contributions to subsequent psychopathology,
functioning. perhaps as part of a larger developmental cascade. According
One implication of Rodkin and Roisman (2010) is that to a causal model, just as healthy peer relationships serve as
popular-aggressive children have a history of exposure to positive socialization forces that enhance normative develop-
peer ecologies in which aggression is effective or valued. ment, unhealthy peer relationships—such as bullying some-
From this perspective, aggression becomes popular when it body else—serve as negative socialization forces that foster
is normative, and once normative, can have enduring ef- psychopathology (Rudolph et al., in press).
fects. Some children with an extensive child care back- Empirical evidence of long-term conseq-
ground may find that their early expressions of aggression uences. Studies have examined longitudinal associa-
have social value within the peer ecology of child care and tions between bullying during childhood or adolescence
May–June 2015 ● American Psychologist 315and adult outcomes such as substance use, offending, and tors, or for personological indicators, such as being a mean
job status (Farrington et al., 2011). However, extant studies boss or corporate billionaire. Some individuals might feel
are limited by only providing evidence for incidental models long-lasting regret and remorse for their participation in
of bullying perpetration. For example, Sourander et al. (2011) bullying as youth, without this translating into detectable
reported that bullying at age eight was a strong predictor of differences on standard adjustment outcomes. In some
adult criminality among males in their early 20s, echoing ways this may be just as well. Once “being a bully” is
Huesmann et al.’s (1984) finding that peer-nominated ag- reified as a risk factor, attempts follow to predict who “will
gression at age eight predicts, for both boys and girls, be a bully,” just as attempts were made in the past to
criminal convictions at age 30. Unfortunately, Sourander et identify the individual characteristics that predict being a
al. (2011) did not control for peer-nominated, or any other school shooter (Mulvey & Cauffman, 2001). These steps
measure, of aggression. Hemphill et al. (2011) found that may be necessary and unavoidable, but they fall short. As
greater bullying perpetration among Australian youth in we will now highlight, it is also critical to identify changes
Year 7 of school was associated with a twofold increase in in life circumstances and relationships that motivate some
binge drinking and marijuana use when these students were youth to bully—a recent humiliation, a new enemy, losing
in Year 10. Similarly, from the U.S. Raising Healthy Chil- or gaining friends. As Mulvey and Cauffmann (2001)
dren project, self-reported childhood bullying in Grade 5 write, “peers and teachers who talk with problem students
was associated with self-reported heavy drinking and mar- can often provide the most useful information about when
ijuana use at age 21 (Kim, Catalano, Haggerty, & Abbott, such students are in trouble” (p. 800). The question of the
2011). Although there were significant associations re- long-term incidental or causal consequences of bullying is
ported by Kim et al. (2011) between childhood bullying a side issue: For the sake of the broader peer ecology,
and later adverse outcomes, these relations were attenuated bullying is just as important to stop whether or not future
when a host of risky individual (hyperactivity) and contex- research reveals negative long-term sequelae for children
tual (family) variables were included. Thus, bullying might and youth who bully.
increase the likelihood of these later outcomes, but even so,
with no independent measure of aggressive behavior, bul- A Relational Approach to Bullying
lying perpetration would likely be incidental to more global This review has had two main points thus far. First, ag-
antisocial tendencies, influenced by multiple contextual gressive and bullying behavior can potentially have func-
factors. tional and dysfunctional components. Not all, or even most,
It is possible that for some outcomes, bullying perpe- aggression and bullying is in equal parts functional and
tration is not even incidental to later, long-term conse- dysfunctional, but whatever might be functional in any
quences of psychopathology. Sourander et al. (2009) re- particular aggressive act is especially difficult to commu-
ported that once childhood history of psychopathology was nicate because it risks implying that aggression can be
controlled, bullying in childhood did not significantly pre- good, or justifiable. This scientific paradox and moral quan-
dict the use of psychotropic medication or psychiatric hos- dary complicates descriptions of psychological mecha-
pitalization in early adulthood. The nine to 16-year-old nisms and concurrent correlates involved in bullying be-
participants of the Great Smoky Mountain Study were later havior. Second, aggressive and bullying behavior have not
assessed for psychiatric disorders between the ages of 19 been adequately distinguished from one another, and it is
and 26 – depression, suicidality, anxiety, panic disorder, not clear that distinct developmental antecedents and long-
agoraphobia, antisocial personality disorder, and alcohol term consequences are entailed for different kinds of bul-
and marijuana disorders (Copeland et al., 2013). Youth lies, or for bullying as opposed to aggressive behavior more
who bullied during childhood were no different than chil- generally. Future research must clarify the taxonomic
dren not involved in bullying on any of the nine long-term structures of bullying and better distinguish bullying from
outcomes examined save antisocial personality disorder aggression.
(not surprisingly; however, also no Type I error adjustment There is a third point embedded in this review that we
across the nine outcomes). Bully victims, however, were will now highlight, namely that bullying is relationally
susceptible to a wide range of psychiatric disorders in oriented (e.g., Pepler et al., 2006; Pepler & Craig, 2011;
adulthood, even after controlling for childhood variables Veenstra et al., 2007). In this section we consider distinc-
and hardships such as low socioeconomic status, maltreat- tions between aggression and bullying more carefully,
ment, and family instability and dysfunction. This is con- highlight why bullying should be considered a relational
sistent with the fact that bully victims tended to be more phenomenon in addition to individual or behavioral char-
severely maladjusted across more domains than other chil- acteristics, and illustrate how a relational approach can be
dren, similar to the profile of rejected-aggressive children. applied by considering questions about the role of gender in
However, extant research designs make it difficult to parse bullying.
the extent to which findings for bully victims are attribut- Distinguishing bullying from aggression.
able to bullying others, being bullied, or possible co-oc- Olweus (2010) had a response to the question of the dis-
curring risk factors such as rejection. tinctiveness of bullying from aggression, writing that bul-
Summary. There may be little beyond our faith in lying is “aggressive behavior with certain special charac-
a just world to demonstrate that bullying perpetration has teristics such as repetitiveness and an asymmetric power
causal, long-term consequences for criminological indica- relationship” (p. 11). The repetition element draws atten-
316 May–June 2015 ● American Psychologisttion to the fact that bullying is not a one-time interaction, or 2% of 8- to 18-year-old participants spontaneously men-
a passing conflict, but a relationship with temporal ex- tioned that bullying is intentional, 6% spontaneously men-
panse. To the extent that there are multiple interactions tioned that bullying is repetitive, and only 26% (but up to
between a child who bullies and another child who is a third of adolescent respondents) noted that bullying en-
harassed by the bullying, a relationship has emerged, even tails a power imbalance.
if it is one that is unwanted by the victim. The term For these reasons, some bullying prevention efforts,
“relationship” should not imply that the tie between bully particularly those from a positive behavioral support per-
and victim is, or ever was, positive, but that “bullying” is spective, address the problem of bullying from a behavioral
intrinsically dynamic, in a relationship that unfolds over level by avoiding the use of the term bully, redefining the
time and that potentially entangles others, such as bystand- bullying construct so as to focus on concrete behaviors
ers. underlying bullying (e.g., hitting, threatening, name-call-
What kind of relationship exists between bullies and ing), without bothering with amorphous and subjective
victims? The ugly heart of bullying and its distinctive requirements of assessing power differentials, and without
element is an asymmetric power relationship. It is power considering the interpersonal dynamic existing between
asymmetry and the accompanying specter of abuse that children who bully and the peers they harass (e.g., Ross &
elicits outrage. Unequal, coercive power, in which a more Horner, 2009). It is indeed possible that the most successful
powerful aggressor attacks a less powerful victim, is what interventions will be those designed to reduce aggression
distinguishes bullying from other forms of aggression. The and antisocial behavior more generally (e.g., Goldstein,
broader construct of aggression encompasses antagonistic Glick, & Gibbs, 1986), with decreases in bullying neces-
behaviors that are emitted within a wider array of relation- sarily following.
ships. Aggressive behaviors that are directed toward one or Our perspective is that there exists unique relational
more individuals of relatively equal power might be con- features of bullying that can be usefully synthesized within
ceptualized as a game of the dozens, a fight, or even as individual, clinical and behavioral perspectives. This
mean-spirited and harmful, but they are less likely to pro- would give the aggression field greater balance, provide
voke moral offense. Situations that involve a weaker indi- more comprehensive information about children who bully
vidual attacking a more powerful individual can evoke and their peer relationships, and allow a conceptual bridge
images of the underdog and may therefore be heralded as to research in relationship science (Reiss, Collins, & Ber-
morally justified (i.e., David and Goliath). In both of these
scheid, 2000) and interpersonal theories of developmental
cases, the nature and impact of the aggressive behavior
psychopathology (Rudolph et al., in press). For example,
may be less serious, as serious, or even more serious than
the bully victim classification itself is useful inasmuch as it
the nature and impact of aggressive behaviors that occur
points to a real profile of children with extensive needs, but
during bullying. Yet, the meaning typically promulgated is
it leaves open basic relational questions such as, “Who is
often less morally repugnant.
With this in mind, what kind of power does a bully the bully victim bullying?” “Who is the bully victim being
really have? A bully can have physical power over his or harassed by?” and the temporal processes behind these
her victim. “His poor neighbor is bully’d by his big ap- complex interactions.
pearance,” wrote Samuel Palmer in 1710, according to the Applying the relational approach to bully-
Oxford English Dictionary, which cites the first known ing: A gendered example. A relational perspec-
written use of the term “bully.” A bully may also have tive on bullying is particularly important when dealing with
psychological power relative to the target of harassment— issues of gender (Pepler et al.,2006). Gender issues in
more friends, greater status and prestige, greater access to bullying include, but go beyond, such contrastive questions
resources, or in the case of bullying, a child may strive to of whether boys or girls are more likely to be bullies, or
derive or gain power by constructing weakness in the whether female bullying is best captured through attention
children being targeted. Power asymmetries may exist to social or relationally aggressive forms (Crick & Zahn-
along a broad array of peer-valued characteristics: physical, Waxler, 2003). Bullying is a gendered phenomenon in
material, psychological, and social (Vaillancourt & Hymel, many ways, whether it be in children’s targeting of, and
2006). attempts to gain status among, same- and other-sex peers
The difficulty of operationalizing an asymmetric (Faris & Felmlee, 2011; Hanish et al., 2012; Rodkin &
power relationship, and of accurately assessing repetition, Berger, 2008), cross-gender bullying as an immature at-
have led to questions of whether there is any meaningful tempt to enter into the “push-and-poke courtship” of early
difference between bullying and aggression, and to at- adolescence (Pellegrini et al., 2010), developmental link-
tempts to do away with or revise definitions of bullying. ages to intimate relationships of coercion and control (Es-
Indeed, researchers are not the only ones who jingle to- pelage, Low, Anderson, & De La Rue, 2013; Rodkin &
gether aggression and bullying (Hawley et al., 2011). Chil- Fischer, 2003; Pepler et al., 2006), the targeting of youth
dren pay little heed to the components of bullying outlined based on real or perceived sexual orientation (Espelage,
by Olweus (1993) when asked to complete the sentence, “A Aragon, Birkett & Koenig, 2008; Pauletti et al., 2014;
bully is . . .,” emphasizing instead global negative behav- Robinson & Espelage, 2011), or using sexually harassing
iors, such as teasing or mean behaviors (Vaillancourt et al., behaviors and homophobic epithets as a means of bullying
2008). According to Vaillancourt et al. (2008), less than (Espelage, Basile, & Hamburger, 2012). Gender often un-
May–June 2015 ● American Psychologist 317derlies decisions about whom to harass, why to harass, and risk factors for aggression are aggregated, such as when
how to harass. arousable, dysregulated, and aggressive boys spend time
One finding that becomes evident once bullies and with other similar boys, the likelihood for the inculcation of
victims are considered in a dynamic relationship is that aggressive and bullying behaviors is enhanced (Martin &
bullying is not limited to same-sex peers. Indeed, once Fabes, 2001). These examples highlight the importance of
children’s interactional preference for same-sex partners is the larger social network in which bully/victim relation-
controlled, rates of same- and other-sex aggression are ships are embedded. Importantly, the social embedding of
relatively similar, although there may be a preference for bullying begins early, as children first come together in
boys (but not girls) to direct physical aggression (but not groups of same-age peers in preschools and daycares (An-
relational aggression) to other boys (Hanish et al., 2012; drews, Hanish, Fabes, & Martin, 2014; Rodkin & Roisman,
Ostrov & Keating, 2004). Sexual selection theory provides 2010), and continues through childhood and adolescence
a conceptual framework for understanding same-sex ag- and into adulthood (Hanish et al., 2011).
gression, as same-sex peers vie for dominance and resource While we have focused on applying a relational ap-
control. However, dominance motives can underlie other- proach to issues of gender, similar questions regarding who
sex aggression as well (Hanish et al., 2012). Even before bullies whom can be applied to race and ethnicity (Garan-
adolescence, empirical reports suggest that there are a deau, Wilson, & Rodkin, 2010; Graham & Juvonen, 2002;
disturbing number of cases, possibly half, where aggressive Tolsma, van Duerzen, Stark, & Veenstra, 2013). From a
boys are harassing girls (Rodkin & Berger, 2008; Veenstra relational perspective, any notable difference between peo-
et al., 2007). Olweus (1993) first reported this finding, ple that can be associated with power differentials, such as
writing that “boys carried out a large part of the bullying religion, disability, or ethnicity, has the potential to be
to which girls were subjected” (p. 18, italics original). That seized upon as an object of harassment in a way that cannot
is, 60% of fifth through seventh grade girls who were be captured by strictly individual-level taxonomies or ap-
reportedly harassed said that they were bullied by boys. proaches that focus exclusively on bullying-like behaviors.
Socially marginalized boys, who are unpopular and re- From a measurement standpoint, one goal for future re-
jected, are most likely to harass girls (Rodkin & Berger, search will be to empirically assess repetition and power
2008). In contrast, socially integrated bullies tend to dem- asymmetries within the bully/victim relationship and social
onstrate within-sex bullying and dominance against unpop- networks, as opposed to simply providing (or not) children
ular targets (Pellegrini et al., 2010; Veenstra et al., 2010). and youth with definitions of bullying that may or may not
Such cross-sex bullying is uniquely problematic in be relevant to their construction and experiences of bully-
that it likely diminishes children’s understanding of, appre- ing (Rodkin, Hanish, Wang, & Logis, 2014; Vaillancourt et
ciation for, and efficacy in interacting with cross-sex peers. al., 2008).
As a result, we may see less adaptive and more maladaptive
interactions with cross-sex peers, the consequences of Conclusions
which can play out seriously in such behaviors as sexual
harassment and relationship violence becoming an ac- After decades in which interest in bullying was minimal,
cepted part of peer culture and the culture at large (Espe- contemporary scholarship has grown in quantity and ex-
lage et al., 2012; Rodkin & Fischer, 2003; Pepler et al., tent, capturing the ear of educators, policymakers, parents,
2006). Peer sexual harassment is often seen as a purely and researchers around the globe. This growing body of
adolescent phenomenon, related to adolescents’ attempts to research has helped draw attention to some important find-
position themselves as desirable mates (Guerra et al., ings about the nature of bullying, and aggression more
2011). However, according to the American Association of generally. Bullying can be fostered by both the presence of
University Women Educational Foundation (2001), sexual a social network and its absence. At least some youth who
harassment begins as early as elementary school, with 38% bully are well-integrated into peer social ecologies, and
of girls who experience sexual harassment reporting first may derive at least short-term benefit from bullying other
occurrence even before entering middle school. Moreover, youth. Other children who bully are socially marginalized
the longitudinal associations among bullying, homophobic and enmeshed in reoccurring cycles of abuse. Among some
name-calling, and sexual harassment in youth are evident children who engage in high levels of bullying, bullying is
as young as age 10 (Espelage et al., 2012). Thus, the origins an indicator, the tip of an iceberg, for a larger profile of
of peer sexual harassment may be linked to how boys and antisocial problems. These distinct profiles and develop-
girls get along with another in early and middle childhood mental pathways could be better established, and related to
(Hanish et al., 2012). other person-oriented classifications of aggression in future
Children’s interactions with male and female peers research.
have the potential to exacerbate or mitigate aggressive An existential problem with the literature on bullying
tendencies, depending on children’s own sex and tenden- is that it may not be distinguishable from the larger body of
cies to display gendered characteristics (Hanish, Hill, Gos- work on childhood aggression. What is there about anti-
ney, Fabes, & Martin, 2011). For example, peers can so- bullying interventions that could not also be addressed by
cialize aggressive behaviors in one another, as evident in a more general violence-reduction program? When reading
interactions with both same- and other-sex peers (Hanish, a journal article, or a popular book for that matter, what
Martin, Fabes, Leonard, & Herzog, 2005). However, when would be lost in the understanding by substituting “aggres-
318 May–June 2015 ● American Psychologistsor” or “aggression” for “bully” or “bullying”? What is the use homophobic language or engage in other forms of
heuristic value of having a literature on bullying when one gender-based harassment as a way to enforce traditional
on aggression already exists? This challenge was especially masculinity norms (Birkett & Espelage, 2014; Darwich,
clear in our discussion of developmental antecedents and Hymel, & Waterhouse, 2012).
long-term consequences of bullying. We would venture to There are translational implications of a relational
predict that meaningful, robust distinctions between bullies approach to bullying and peer harassment. Initiatives like
and other aggressive children are unlikely to be found in the Promoting Relationships and Eliminating Violence
antecedent personality or dispositional features. Similarly, Network (PREVNet), advocate that “promoting relation-
we would predict that genetic correlates to aggression, such ships is essential to . . . eliminating violence” and that “the
as those found in twin studies (e.g., Brendgen et al., 2011), use of power and aggression underlies most forms of in-
would also be found were measures of bullying substituted terpersonal violence throughout the life span . . .. Healthy
for measures of aggression— but not necessarily when relationships [are] trusting, supportive, and devoid of the
more general aggressive tendencies were controlled for. In use of power and aggression” (Pepler & Craig, 2011, p.
short, readers interested in the question of antecedents or 389). Long-term outcomes that are most causally associ-
consequences of bullying might be well-served by consult- ated with bullying may be those that are domain-specific,
ing what is known about aggression more generally. focused on aggression within relationships, such as inti-
In order to better distinguish bullying from aggres- mate partner violence or intergroup hostility. In the years
sion, we have stressed that, along with being a feature of ahead, researchers interested in bullying should consider
individual personality and behavior, bullying is perpetrated the unique insights bullying provides on aggression as it
within a relationship, albeit a coercive, unequal, asymmet- operates, however destructively, within diverse relation-
ric relationship, characterized by aggression. Because bul- ships and through the social networks of childhood and
lying is aggression directed from at least one person to adolescence.
another, research on bullying might benefit from a more
explicitly relational or interpersonal perspective that in- REFERENCES
cludes information about the bully victim dynamic, by-
American Association of University Women Educational Foundation.
standers, and related social networks. It has been overdue (2001). Hostile hallways: Bullying, teasing, and sexual harassment in
for the field of social development to accept what Olweus school. Washington, DC: Author.
(1978) pointed out 35 years ago: Aggression can bring Andrews, N. C. Z., Hanish, L. D., Fabes, R. A., & Martin, C. L. (2014).
social capital, some bullies may hide in the plain sight of With whom and where you play: Preschoolers’ social context predicts
peer victimization. Social Development, 23, 357–375. http://dx.doi.org/
social acceptance, while other, more socially marginalized 10.1111/sode.12051
youth are mired in what may seem like a chronic cycle of Baldry, A. C. (2003). Bullying in schools and exposure to domestic
bullying and being bullied. violence. Child Abuse & Neglect, 27, 713–732. http://dx.doi.org/
While understanding differences between socially in- 10.1016/S0145-2134(03)00114-5
tegrated bullies and socially marginalized bully victims is Birkett, M., & Espelage, D. L. (2014). Homophobic name-calling, peer-
groups, and masculinity: The socialization of homophobic behavior in
important, and distinguishing bullying from aggression is adolescents. Social Development. Advance online publication.
essential, research would be well-served by moving beyond Brendgen, M., Boivin, M., Dionne, G., Barker, E. D., Vitaro, F., Girard,
categorical schemes to relational, situational analyses of A., . . . Pérusse, D. (2011). Gene-environment processes linking ag-
the extent to which bullying behaviors are dysfunctional gression, peer victimization, and the teacher-child relationship. Child
Development, 82, 2021–2036. http://dx.doi.org/10.1111/j.1467-8624
and maladaptive versus functional and adaptive with re- .2011.01644.x
spect to social dynamics that are operating at any given Bronfenbrenner, U. (1979). The ecology of human development: Experi-
point in time. This includes knowing not just who is a bully ments by nature and design. Cambridge, MA: Harvard University
and who is a victim, but who bullies whom and how bullies Press.
and the children they harass are situated within peer social Card, N., & Hodges, E. V. E. (2010). It takes two to fight in school, too:
A social relations model of the psychometric properties and relative
ecologies. Which bullies are harassing which victims? variance of dyadic aggression and victimization in middle school.
Does bullying occur between youth of different ethnicities, SocialDevelopment,19,447– 469.http://dx.doi.org/10.1111/j.1467-9507
genders, or ages? Not much is known about the relationship .2009.00562.x
between a bully and the child whom he or she targets—for Carver, C. S., Johnson, S. L., & Joormann, J. (2008). Serotonergic
function, two-mode models of self-regulation, and vulnerability to
instance, were they friends, former friends, enemies, or depression: What depression has in common with impulsive aggression.
strangers (for an exception, see Rodkin et al., 2014)? Ad- Psychological Bulletin, 134, 912–943. http://dx.doi.org/10.1037/
ditionally, we need to know more about the nuanced di- a0013740
mensions upon which power differences are expressed. Cook, C. R., Williams, K. R., Guerra, N. G., Kim, T. E., & Sadek, S.
We considered the complexity that gender brings to a (2010). Predictors of bullying and victimization in childhood and ado-
lescence: A meta-analytic investigation. School Psychology Quarterly,
relational analysis of bullying, beyond issues of what may 25, 65– 83. http://dx.doi.org/10.1037/a0020149
be gender-normative subtypes of overt or social aggression Copeland, W. E., Wolke, D., Angold, A., & Costello, E. J. (2013). Adult
or prevalence rates of male and female bullying. Gender psychiatric outcomes of bullying and being bullied by peers in child-
and sexuality is a hidden underbelly of much bullying and hood and adolescence. Journal of the American Medical Association
Psychiatry,70,419 – 426.http://dx.doi.org/10.1001/jamapsychiatry.2013
harassment. Bullying may be as likely to occur between .504
boys and girls as within gendered peer groups, and to start Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social
well before the transition to adolescence. In addition, youth information processing mechanisms in children’s social adjustment.
May–June 2015 ● American Psychologist 319Psychological Bulletin, 115, 74 –101. http://dx.doi.org/10.1037/0033- Hanish, L. D., Hill, A., Gosney, S., Fabes, R. A., & Martin, C. L. (2011). 2909.115.1.74 Girls, boys, and bullying in preschool: The role of gender in the Crick, N. R., & Zahn-Waxler, C. (2003). The development of psychopa- development of bullying. In D. L. Espelage & S. Swearer (Eds.), thology in females and males: Current progress and future challenges. Bullying in North American schools (2nd ed., pp. 132–146). New York, Development and Psychopathology, 15, 719 –742. http://dx.doi.org/ NY: Routledge. 10.1017/S095457940300035X Hanish, L. D., Martin, C. L., Fabes, R. A., Leonard, S., & Herzog, M. Darwich, L., Hymel, S., & Waterhouse, T. (2012). School avoidance and (2005). Exposure to externalizing peers in early childhood: Homophily substance use among Lesbian, Gay, Bisexual, and Questioning youths: and peer contagion processes. Journal of Abnormal Child Psychology, The impact of peer victimization and adult support. Journal of Educa- 33, 267–281. http://dx.doi.org/10.1007/s10802-005-3564-6 tional Psychology, 104, 381–392. http://dx.doi.org/10.1037/a0026684 Hanish, L. D., Sallquist, J., DiDonato, M., Fabes, R. A., & Martin, C. L. Dishion, T. J., & Tipsord, J. M. (2011). Peer contagion in child and (2012, September). Aggression by whom-aggression toward whom: adolescent social and emotional development. Annual Review of Psy- Behavioral predictors of same- and other-gender aggression in early chology, 62, 189 –214. http://dx.doi.org/10.1146/annurev.psych.093008 childhood. Developmental Psychology, 48, 1450 –1462. http://dx.doi .100412 .org/10.1037/a0027510 Dodge, K. A., Coie, J. D., & Lynam, D. (2006). Aggression and antisocial Hawley, P. H., Stump, K. N., & Ratliff, J. (2011). Sidestepping the jingle behavior in youth. In W. Damon (Series Ed.) & N. Eisenberg (Vol. fallacy: Bullying, aggression, and the importance of knowing the dif- Ed.), Handbook of child psychology: Social, emotional, and personality ference. In D. L. Espelage & S. Swearer (Eds.), Bullying in North development (6th ed., Vol. 3, pp. 719 –788). New York, NY: Wiley. American schools (2nd ed., pp. 101–115). New York, NY: Routledge. Ellis, B. J., Del Giudice, M., Dishion, T. J., Figueredo, A. J., Gray, P., Hemphill, S. A., Kotevski, A., Herrenkohl, T. I., Bond, L., Kim, M. J., Griskevicius, V., . . . Wilson, D. S. (2012). The evolutionary basis of Toumbourou, J. W., & Catalano, R. F. (2011). Longitudinal conse- risky adolescent behavior: Implications for science, policy, and prac- quences of adolescent bullying perpetration and victimisation: A study tice.DevelopmentalPsychology,48,598 – 623.http://dx.doi.org/10.1037/ of students in Victoria, Australia. Criminal Behaviour and Mental a0026220 Health, 21, 107–116. http://dx.doi.org/10.1002/cbm.802 Espelage, D. L., Aragon, S. R., Birkett, M., & Koenig, B. (2008). Ho- Huesmann, L. R., Eron, L. D., Lefkowitz, M. M., & Walder, L. O. (1984). mophobic teasing, psychological outcomes, and sexual orientation Stability of aggression over time and generations. Developmental Psy- among high school students: What influence do parents and schools chology, 20, 1120 –1134. http://dx.doi.org/10.1037/0012-1649.20.6 have? School Psychology Review, 37, 202–216. .1120 Espelage, D. L., Basile, K. C., & Hamburger, M. E. (2012). Bullying Hymel, S., Schonert-Reichl, K. A., Bonanno, R. A., Vaillancourt, T., & experiences and co-occurring sexual violence perpetration among mid- Henderson, N. R. (2010). Bullying and morality: Understanding how dle school students: Shared and unique risk factors. Journal of Adoles- good kids can behave badly. In S. R. Jimerson, S. M. Swearer, & D. L. cent Health, 50, 60 – 65. http://dx.doi.org/10.1016/j.jadohealth.2011.07 Espelage (Eds.), Handbook of bullying in schools: An international .015 perspective (pp. 101–118). New York, NY: Routledge. Espelage, D. L., Holt, M. K., & Henkel, R. R. (2003). Examination of Juvonen, J., & Galván, A. (2009). Bullying as a means to foster compli- peer-group contextual effects on aggression during early adolescence. ance. In M. J. Harris (Ed.), Bullying, rejection, and peer victimization: Child Development, 74, 205–220. http://dx.doi.org/10.1111/1467-8624 A social cognitive neuroscience perspective (pp. 299 –318). New York, .00531 NY: Springer. Espelage, D. L., Low, S. K., Anderson, C., & De La Rue, L. (2013). Kim, M. J., Catalano, R. F., Haggerty, K. P., & Abbott, R. D. (2011). Bullying, sexual, and dating violence trajectories from early to late Bullying at elementary school and problem behaviour in young adult- adolescence. Report submitted to the National Institute of Justice Grant hood: A study of bullying, violence and substance use from age 11 to #2011-MU-FX-0022. age 21. Criminal Behaviour and Mental Health, 21, 136 –144. http:// Espelage, D. L., Low, S., Rao, M. A., Hong, J. S., & Little, T. (2013). dx.doi.org/10.1002/cbm.804 Family violence, bullying, fighting, and substance use among adoles- Kliewer, W., Dibble, A. E., Goodman, K. L., & Sullivan, T. N. (2012). cents: A longitudinal transactional model. Journal of Research on Physiological correlates of peer victimization and aggression in African Adolescence. Advance online publication. American urban adolescents. Development and Psychopathology, 24, Faris, R., & Felmlee, D. (2011). Status struggles: Network centrality 637– 650. http://dx.doi.org/10.1017/S0954579412000211 and gender segregation in same- and cross-gender aggression. Amer- Martin, C. L., & Fabes, R. A. (2001). The stability and consequences of ican Sociological Review, 76, 48 –73. http://dx.doi.org/10.1177/ young children’s same-sex peer interactions. Developmental Psychol- 0003122410396196 ogy, 37, 431– 446. http://dx.doi.org/10.1037/0012-1649.37.3.431 Farmer, T. W., Petrin, R. A., Robertson, D. L., Fraser, M. W., Hall, C. M., Moffitt, T. E. (1993). “Life-course-persistent” and “adolescence-limited” Day, S. H., & Dadisman, K. (2010). Peer relations of bullies, bully- antisocial behavior: A developmental taxonomy. Psychological Review, victims, and victims: The two social worlds of bullying in second-grade 100, 674 –701. http://dx.doi.org/10.1037/0033-295X.100.4.674 classrooms. The Elementary School Journal, 110, 364 –392. http://dx Mulvey, E. P., & Cauffman, E. (2001). The inherent limits of predicting .doi.org/10.1086/648983 school violence. American Psychologist, 56, 797– 802. http://dx.doi.org/ Farrington, D. P., Ttofi, M. M., & Lösel, F. (2011). School bullying and 10.1037/0003-066X.56.10.797 later criminal offending. Criminal Behaviour and Mental Health, 21, Olweus, D. (1978). Aggression in school: Bullies and whipping boys. 77–79. http://dx.doi.org/10.1002/cbm.807 Washington, DC: Hemisphere. Garandeau, C. F., Wilson, T., & Rodkin, P. C. (2010). The popularity of Olweus, D. (1993). Bullying at school. Oxford, UK: Blackwell. elementary school bullies in gender and racial context. In S. R. Olweus, D. (2010). Understanding and researching bullying: Some critical Jimerson, S. M. Swearer, & D. L. Espelage (Eds.), Handbook of issues. In S. R. Jimerson, S. M. Swearer, & D. L. Espelage (Eds.), bullying in schools: An international perspective (pp. 119 –136). New Handbook of bullying in schools: An international perspective (pp. York, NY: Routledge. 9 –34). New York, NY: Routledge. Goldstein, A. P., Glick, B., & Gibbs, J. C. (1986). Aggression replacement Ostrov, J. M., & Keating, C. F. (2004). Gender differences in preschool training: A comprehensive intervention for aggressive youth. Cham- aggression during free play and structured interactions: An observa- paign, IL: Research Press. tional study. Social Development, 13, 255–277. http://dx.doi.org/ Graham, S., & Juvonen, J. (2002). Ethnicity, peer harassment, and adjustment 10.1111/j.1467-9507.2004.000266.x in middle school: An exploratory study. The Journal of Early Adolescence, Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal 22, 173–199. http://dx.doi.org/10.1177/0272431602022002003 adjustment: Are low-accepted children at risk? Psychological Bulletin, Guerra, N. G., Williams, K. R., & Sadek, S. (2011). Understanding 102, 357–389. http://dx.doi.org/10.1037/0033-2909.102.3.357 bullying and victimization during childhood and adolescence: A mixed Patterson, G. R., Littman, R. A., & Bricker, W. (1967). Assertive behavior methods study. Child Development, 82, 295–310. http://dx.doi.org/ in children: A step toward a theory of aggression. Monographs of the 10.1111/j.1467-8624.2010.01556.x Society for Research in Child Development, 32 (5, Serial No. 113). 320 May–June 2015 ● American Psychologist
You can also read
