This article was downloaded by: [University of California Santa Cruz]On: 19 November 2014, At: 05:19Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Journal of Clinical Child & Adolescent PsychologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/hcap20

Do Actions Speak Louder Than Words? ClassroomAttitudes and Behavior in Relation to Bullying in EarlyAdolescenceRon Scholte a , Miranda Sentse a & Isabela Granic aa Behavioural Science Institute, Radboud UniversityPublished online: 06 Nov 2010.

To cite this article: Ron Scholte , Miranda Sentse & Isabela Granic (2010) Do Actions Speak Louder Than Words? ClassroomAttitudes and Behavior in Relation to Bullying in Early Adolescence, Journal of Clinical Child & Adolescent Psychology, 39:6,789-799, DOI: 10.1080/15374416.2010.517161

To link to this article: http://dx.doi.org/10.1080/15374416.2010.517161

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Do Actions Speak Louder Than Words? Classroom Attitudesand Behavior in Relation to Bullying in Early Adolescence

Ron Scholte, Miranda Sentse, and Isabela Granic

Behavioural Science Institute, Radboud University

The aim of the present study was to examine to what extent classroom factors (i.e.,classroom antibullying attitudes and behavioral norms) contributed to individualbullying, after controlling for individual difference characteristics. Participants were2,547 early adolescents (M¼ 13.4 years, SD¼ .63) from 109 middle school classes. Self-and peer reports were used to answer the research questions. It was found that adoles-cents in classrooms that held permissive attitudes toward bullying were more likely tobully themselves, even after controlling for individual attitude, gender, social preference,and number of reciprocal friends. However, the association of classroom attitudes withindividual bullying decreased substantially when classroom bullying behavior was takeninto account. Our study suggests that the effects of classroom antibullying attitudesmight be partly mediated by classroom behaviors. It implies that research that hasnot included classroom behavior might have overestimated the effects of classroomattitudes on bullying.

Recent studies show that bullying is a serious andcommon problem, with many children and adolescentsin both Western and non-Western countries (e.g., Dueet al., 2005; Eslea et al., 2003; Nansel, Craig, Overpeck,Saluja, & Ruan, 2004). Bullying is usually defined asrepeated, negative actions by one or more peers towarda person who cannot defend himself or herself (Olweus,1993). To diminish the prevalence of bullying, numerousprevention and intervention programs have beendeveloped and implemented. Despite some promisingprograms that have shown good outcomes (Olweus,1994; Salmivalli, Kaukiainen, & Voeten, 2005), mostof these initiatives have been modest at best in theirsuccess (see, e.g., Samara & Smith, 2008; Smith, Pepler,& Rigby, 2004) and have produced moderate to loweffect sizes. Part of the reason for these small effects isthat an adequately comprehensive model of the develop-ment and maintenance of bullying behavior is lackingand important factors may have been overlooked.

One promising theoretical model is the social ecologi-cal model of Swearer and Espelage (2004). According to

this model, bullying does not occur in isolation, butrather is the result of interactions between individualfactors, peers, family, classroom, and community fac-tors. Until recently, most research on bullying focusedon individual factors. The last decade, however, haswitnessed an increase in studies examining the socialcontext in which bullying episodes are embedded. Theclassroom context, both in terms of the collective stu-dents’ attitudes and bullying behaviors, may have stronginfluences on individual adolescents’ bullying propensi-ties. To develop more effective and more comprehensiveprevention programs, we might need a conceptual modelof bullying behavior that takes into account individualdifference characteristics as well as social factors, includ-ing both classroom attitudes and overt behavioralnorms. The shift from investigating individual differencecharacteristics to classroom factors is relevant becausethe attitudes and behaviors of the class as a whole canbe an explanation for why the same child (with stableindividual characteristics) may be bullied in one classbut not in another.

Previous research on bullying has usually focused onindividual difference characteristics of bullies and hasshown that boys are more likely to be bullies than girls(e.g., Boulton & Smith, 1994; Nansel et al., 2001;

Correspondence should be addressed to Ron Scholte, Behavioural

Science Institute, Radboud University, PO Box 9104, 6500 HE

Nijmegen, the Netherlands. E-mail: r.scholte@bsi.ru.nl

Journal of Clinical Child & Adolescent Psychology, 39(6), 789–799, 2010

Copyright # Taylor & Francis Group, LLC

ISSN: 1537-4416 print=1537-4424 online

DOI: 10.1080/15374416.2010.517161

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

Scholte, Engels, Overbeek, de Kemp, & Haselager,2007) and that Machiavellialism (Andreou, 2004),empathy (Caravita, Di Blasio, & Salmivalli, 2009) andsocial skills (Sutton, Smith, & Swettenham, 1999) arerelated to bullying. Furthermore, bullying has beenrelated to social acceptance and peer relations. Ingeneral bullies are less liked and more disliked by peersand have fewer friends (Boulton & Smith, 1994;Pellegrini, 1998; Salmivalli, Huttunen, & Lagerspetz,1997). Valuable as these studies have been, focusingsolely on individual difference characteristics may notbe adequate to fully understand bullying.

Research also suggests that bullying is a groupphenomenon, taking place in social groups such asschool classes (Ladd, 1983). The work on participantroles in bullying (Goossens, Olthof, & Dekker, 2006;Salmivalli, Lagerspetz, Bjorkqvist, Osterman, &Kaukiainen, 1996; Schafer, Korn, Brodbeck, Wolke, &Schultz, 2005; Sutton & Smith, 1999) has revealed thatup to 85% of all class members may in one way oranother be involved in bullying. It is clear that a com-prehensive model of bullying should include the socialcontext in which bullying takes place, in addition toindividual factors. Because bullying predominantlytakes place in school classes (but see, e.g., Coyne, Craig,& Chong, 2004; Smith, Singer, Hoel, & Cooper, 2003,for bullying in the workplace), in this study we focusedon individual difference factors and on factors related tothe classroom context.

Two classroom factors may be especially relevantwith respect to bullying. First, classroom attitudestoward bullying can be expected to predict individualinvolvement in bullying. Existing research on attitudeshas been restricted to students’ individual attitudes. Ingeneral, children and adolescents report a negativeattitude toward bullying (Andreou, Vlachou, &Didaskalou, 2005; Boulton, Bucci, & Hawker, 1999;Menesini et al., 1997; Rigby & Slee, 1991). Given theassumed association between individual attitudes andbullying, various intervention programs aim at changingbullying attitudes in classrooms. For example, theprogram developed by Olweus (1994), which has beenimplemented in more or less adapted versions in variouscountries including Belgium (Stevens, de Bourdeaudhuij,& Oost, 2000) Canada (Pepler, Craig, Ziegler, &Charach, 1994), Germany (Hanewinkel & Eichler,1999), Great Britain (Smith & Sharp, 1994), and theUnited States (Bauer, Lozano, & Rivara, 2007; Olweus& Limber, 1999) explicitly focuses on developing antibul-lying or pro-victim attitudes in order to reduce individualbullying in classrooms (see also Smith & Ananiadou,2003). Other programs similarly target students’ bullyingattitudes (e.g., Salmivalli et al., 2005).

Still, the extent to which individual attitudes predictactual bullying behavior is not yet clear. Whereas some

studies report a significant association (Andreou et al.,2005; Boulton et al., 1999; Eslea & Smith, 2000), otherssuggest that individual bullying attitudes are only minorto moderate predictors of bullying (e.g., Boulton,Trueman, & Flemington, 2002; Rigby, 2004; Salmivalli& Voeten, 2004).

Of interest, whereas many of the children and adoles-cents hold a negative attitude toward bullying, they arenevertheless directly or indirectly involved in bullying(Salmivalli & Voeten, 2004). One explanation for thisparadox is that children and adolescents’ behavior notonly is guided by their individual attitudes but may besubstantially affected by the social context (i.e., theclassroom) in which they are embedded. Attitudes atthe classroom level may be important and may partlypredict whether individuals are more likely to becomeinvolved in bullying. That is, when a class as a wholeholds a permissive attitude toward bullying, a studentmay be more inclined to become involved in bullyingcompared to a student within a class that is character-ized by a nonpermissive attitude.

We examined bullying in a sample of early adoles-cents just after their transition to secondary school. Thisis a crucial period to investigate because the transitionfrom primary school to secondary school is a time ofabrupt changes. Children enter a new social environ-ment and have to (re)establish social relationships withclassmates (Eslea et al., 2003). A commonly used strat-egy to establish dominance status with peers, especiallyfor boys, is the use of aggression and bullying (Pellegrini& Long, 2002). Attitudes toward bullying change overthe period of early to middle adolescence. Specifically,attitudes become less negative about bullying until theage of 14 or 15, after which these attitudes shift onceagain and become less permissive (Olweus & Endresen,1998). Studies on attitudes on bullying and aggressionat a classroom level have revealed that classmates’ atti-tudes appeared to predict changes in children’s individ-ual behavior (Henry et al., 2000; Salmivalli & Voeten,2004). Other studies, however, have failed to find anysignificant long-term associations (e.g., Stevens, Oost,& De Bourdeaudhuij, 2000).

One reason why classroom attitudes may not alwaysbe strongly related to individual bullying behavior isthat not only the classroom attitudes but also thegeneral level of bullying inside a class (i.e., descriptivenorm; cf. Cialdini, Kallgren, & Reno, 1991; Henryet al., 2000) may influence an individual’s behavior.Studying classroom behavioral norms could be animportant extension of previous studies on classroomattitudes. It is possible that even in classroomscharacterized by a nonpermissive attitude toward bully-ing, some class members will bully. In line withsocial-learning theory, adolescents may observe thatclassmates who bully, despite the dominant antibullying

790 SCHOLTE, SENTSE, GRANIC

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

attitudes in class, are rewarded for that behavior, suchas by an increase in social status or dominance(Pellegrini & Long, 2002; Sentse, Scholte, Salmivalli, &Voeten, 2007). Observing these rewards may serve as areinforcer for other adolescents to also engage in thisbehavior. In this respect, the classroom behavior canmediate the link between classroom attitudes and indi-vidual bullying. Studies on the broader area ofaggression, as well as on bullying, reveal that groupnorms are related to individuals’ intentions to aggressor bully (Nesdale, Durkin, Maass, Kiesner, & Griffiths,2008; but see also Nesdale, Milliner, Duffy, & Griffiths,2009) and to actual aggression and bullying (Duffy &Nesdale, 2009; Espelage, Holt, & Henkel, 2003; Henry,2001; Kuppens, Grietens, Onghena, Michiels, &Subramanian, 2008). Other studies show that groupnorms influence the extent to which children or adoles-cents find bullying or aggression acceptable (Ojala &Nesdale, 2004; Sentse et al., 2007; Stormshak, Bierman,Bruschi, Dodge, & Coie, 1999). Although both normsand attitudes are important factors in relation to indi-vidual bullying or aggression, studies that examinedbullying attitudes at the classroom level have usuallynot accounted for classroom behavioral norms. Thus,it is not yet clear to what extent classroom attitudesare related to individual levels of bullying, whencontrolling for classroom behavior.

THE PRESENT STUDY

The overall aim of the present study was to examine towhat extent classroom factors (i.e., class-level antibully-ing attitudes and behavioral norms) would contribute toindividual bullying, after controlling for social charac-teristics of the individual. These characteristics includedindividual antibullying attitudes and also gender, socialpreference, and the number of friends, as these factorshave been shown to be associated with bullying in pre-vious research (e.g., Hodges & Perry, 1999; Pellegrini,Bartini, & Brooks, 1999; Scholte et al., 2009; Sentseet al., 2007). From a social perspective, having friendsand being liked by peers can be regarded as social fac-tors. From a statistical point of view, however, theyare treated as individual-level factors in multilevel analy-ses because each child attains an individual score onthese variables, which can differ from the scores of otherchildren. We also expected that classroom behavioralnorms (i.e., the mean behavior of all classmates in aclassroom) would be associated with individual bullyinginvolvement over and beyond classroom attitudes. Find-ing significant associations between classroom factorsand individual bullying would have important implica-tions, both for building a more adequate conceptual

model of bullying and for refining prevention andintervention programs.

Two main hypotheses were investigated: First, class-room antibullying attitudes were expected to be relatedto adolescents’ bullying behavior, even after controllingfor individual antibullying attitudes and importantindividual difference characteristics such as gender,social preference, and number of reciprocal friends.Second, classroom bullying behavior was hypothesizedto mediate, or partially mediate, the association betweenclassroom attitudes and individual bullying behavior.

METHOD

Participants

Participants were 2,547 early adolescents (M¼ 13.4years, SD¼ .63) from 109 classes from 43 schools inthe Netherlands. The adolescents were all in the firstgrade of secondary school. Of our total sample ofadolescents, 51% were boys and 49% were girls. Ethnicbackground was recorded, showing that 2,051 of theparticipants were of Dutch origin. The otherparticipants were ethnic minorities, originating fromSurinam and the Dutch Antilles (17), Turkey andMorocco (141), or elsewhere (338).

Procedure

Forty-seven secondary schools were sent a letter,directed to the headmaster of the school, in which weexplained our study on bullying and in which westressed the importance of examining young adolescentsjust after their transition to secondary school. Of the 47schools we contacted, only 4 schools (9%) did not wantto participate in our study because of a busy educationalprogram (n¼ 1), because of their participation in othercurrent studies (n¼ 2), or because they did not thinkthe present study was a useful one, as bullying wouldnot be taking place in their school (n¼ 1). We thus wereable to collect data in 43 secondary schools with 109classes. Passive assent and consent was obtained fromthe parents and adolescents, respectively. Schools weresent letters that all adolescents took home to informtheir parents about the study. In these letters, parentswere asked to contact the research team when they didnot want their child to participate. None of the parentsrefused to consent. The study was approved by theInstitutional Review Board.

The participants were asked to fill out questionnairesduring classroom sessions of 50min, for each schoolclass separately. The questionnaires were administeredby trained master-students, in the presence of the tea-cher of the class. Before handing out the questionnaires,

BULLYING ATTITUDES AND BEHAVIOR 791

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

the examiner gave an introduction and classinstructions. The participants were told that the studywas on well-being and social interactions. They wereguaranteed confidentiality and were reminded that theirparticipation was not obligatory, but no one refused toparticipate. After the classroom assessment, the parti-cipants were thanked for participation. After the wholedata collection was finished, schools received a shortresearch report.

Measures

Self-reports of involvement in bullying. Involve-ment in bullying was measured using a Dutch adap-tation of the Olweus Bully=victim Questionnaire(Olweus, 1989). This measure has been frequently usedin previous studies and has been correlated with socialfactors such as preference and number of friends, andto personal factors including personality, self-esteem,and loneliness in large-scale adolescent samples (e.g.,Scholte et al., 2009; Sentse et al., 2007). The BullyingOthers subscale consisted of five items—(a) How oftendid you bully others this term; (b) How often did youbully others in the last five days; (c) How often doyou say mean things to other kids; (d) How often doyou hit, kick, or threaten others; (e) How often didyou bully others last term—and had a reliability ofa¼ .70. Answers were made on a 5-point scale ranging1 (never), 2 (sometimes), 3 (regularly), 4 (once a week),and 5 (more times a week). For the second question,the answers were never, once, two times, three or fourtimes, and five times or more. For each participant,the scores on the five items were averaged to create ascore on bullying. In the questionnaire we defined theconcept of ‘‘bullying’’ as follows:

We can speak of bullying when a child or a group ofpeers say mean or hurtful things to another child. Thesame is true when a child is being hit, kicked, threatened,or is being excluded from the group. These things can beclassified as bullying when they happen frequently orregularly, and when it’s difficult for the child being bull-ied to defend him- or herself. It’s NOT bullying whentwo or more children who are equally strong tease eachother or fight with each other.

Social preference. Adolescents were handed out aroster containing the names and numbers of their class-mates. Each participant was asked to nominate up tofive classmates in response to the questions ‘‘Whichclassmates do you like most?’’ and ‘‘Which classmatesdo you like least?’’ Adolescents could nominate eachclassmate as well as classmates who were not presentat the time of the data collection. Self-nominations were

not allowed. For each participant the number ofnominations received on the ‘‘like most’’ were sub-tracted from the ‘‘like least’’ questions to obtain thesocial preference score (Coie & Dodge, 1983). Thesevalues were then standardized within the classroom toaccount for differences in class size, a well-establishedmethodology for assessing social preferences (Newcomb& Bukowski, 1983).

Reciprocal friends. Reciprocal friends are friendswho mutually nominate one another as friends (Hartup,1996). In line with this conceptualization, the number ofreciprocal friends an adolescent had was measured witha sociometric question: ‘‘Which classmates are your bestfriends?’’ Friendships were labeled reciprocal if both anadolescent and a classmate reciprocally nominated eachother, consistent with past research (e.g., Scholte et al.,2009; Swenson & Rose, 2009). The concept ‘‘friends’’was defined as follows: ‘‘Friends are classmates whoyou see and meet with on a regular basis and who wouldnominate you as a friend as well.’’ Adolescents couldnominate up to five classmates. For each participantthe number of reciprocal friend nominations wascalculated, which could vary from zero to five.

Antibullying attitudes. To measure antibullying atti-tudes we adapted the measure developed by Salmivalliand Voeten (2004). The validity of that measure wassupported by findings showing that the attitudesassessed by this scale significantly predicted bullyingbehavior. In our study, participants were asked to evalu-ate the extent to which they agreed or disagreed with 10statements about bullying. Items with (r) were reversecoded.

1. One should try to help the bullied victims.2. Bullying may be fun sometimes (r).3. It is the victims’ own fault that they are bullied (r).4. Bullying is stupid.5. Joining in bullying is a wrong thing to do.6. It is not that bad when you laugh with others when

someone is being bullied (r).7. One should report bullying to the teacher.8. Making friends with the bullied victim is the right

thing to do.9. It is funny, when someone ridicules a classmate

over and over again (r).10. Bullying makes the victim feel bad.

Answers were made on a 5-point scale, ranging from0 (strongly disagree) to 4 (strongly agree). As in Salmi-valli and Voeten’s study, factor analyses revealed oneattitude factor. Reliability of the scale in our studywas a¼ .68. After recoding and averaging the students’

792 SCHOLTE, SENTSE, GRANIC

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

scores on this scale, a high score on this scale indicated anegative attitude toward bullying.

Classroom antibullying attitudes (attitudinal norm).The self-reported antibullying attitude scores wereaggregated to the classroom level by averaging adoles-cents’ scores within each classroom, following the sameprocedure for creating group-level variables as was donein previous research (Bellmore, Witkow, Graham, &Juvonen, 2004; Espelage et al., 2003; Sentse et al.,2007). We left out the target child score in calculatingclassroom averages. Classroom antibullying attitudesin this study are thus referring to an attitudinal norm,conceptualized as the average antibullying attitude inclass, which can be seen as normative and from whichindividual attitudes can deviate.

Classroom bullying behavior (descriptive,behavioral norm). We computed a classroom-level(aggregated) score of bullying based on the self-reportsof the variables on the Olweus Bully=victim Question-naire (Olweus, 1989; see the Self-Reports of Involvementin Bullying section). We averaged the adolescents’ scoreswithin each classroom to create a classroom level ofbullying. The target child’s score was left out of theclassroom average. This procedure has been shown tobe valid in prior studies that revealed that classroombullying behavioral norms moderated the associationbetween bullying involvement and social preference(e.g., Sentse et al., 2007).

Data Analyses

First, descriptive statistics are presented for bullying, thenumber of reciprocal friends, social status, and antibul-lying attitude (individual-level effects), and for class-room levels of antibullying attitudes and classroombullying behavior (classroom-level effects). Second,multilevel models, which are equivalent to hierarchicallinear models, are presented. First, we modeled the vari-ance on both the individual level and the classroom levelin adolescents’ bullying scores (the empty model).Second, the individual-level factors were included inthe model (the individual-level model). Finally, theclassroom-level effects were entered into the equation(the full model).

RESULTS

Descriptive Statistics

Means and standard deviations of the individual andclassroom level variables are given in Table 1 for the

total sample, as well as for girls and boys separately.Bullying scores were higher for boys, t(2,545)¼�10.70, p< .01, whereas girls had higher scores on thenumber of reciprocal friends, t(2,545)¼ 6.22, p< .01;social preference, t(2,545)¼ 3.32, p< .01; and a morenegative attitude toward bullying, t(2,545)¼ 10.95,p< .01. We subsequently tested gender differences byincluding gender and interactions with gender as predic-tors in our multilevel models.

Table 2 shows the correlations between the variablesunder study. These correlations regard the total sampleand do no account for possible classroom differences.Bullying was strongly but not perfectly correlated withantibullying attitudes, both at the individual and grouplevel. Social preference was negatively related to bully-ing at the individual level but was unrelated to eitherbullying or antibullying attitudes at the group level.The opposite pattern can be observed for number ofreciprocal friends, which was unrelated to bullying atthe individual level whereas it was negatively related tobullying and permissive attitudes at the group level.

Multilevel Analysis

To test the hypothesized effects we constructed multi-level models for bullying, using MLwiN 2.02 (Rasbash

TABLE 1

Distribution of Main Study Variables

Girlsa Boysb Totalc

Variable M SD M SD M SD

Bullying 1.33 .40 1.53 .50 1.43 .46

N of Reciprocal Friends 2.35 1.43 1.99 1.46 2.17 1.46

Social Preference .09 .90 �.04 1.02 .02 .97

Individual Antibullying Attitudes 4.28 .54 4.03 .61 4.15 .59

Classroom Antibullying Attitude 3.97 .20 3.97 .21 3.97 .20

Classroom Bullying Behavior 1.37 .15 1.37 .16 1.37 .15

Note. Social Preference is z standardized.an¼ 1,237.bn¼ 1,310.cn¼ 2,547.

TABLE 2

Zero-Order Correlations Among Main Study Variables

(Uncorrected for Classroom-Level Differences)

Variables 1 2 3 4 5 6

1. Bullying —

2. N of Reciprocal Friends .01 —

3. Social Preference �.04� .41� —

4. Individual Antibullying Attitudes �.52� .03 .01 —

5. Classroom Antibullying Attitude �.22� .05� �.01 .21� —

6. Classroom Bullying Behavior .23� �.06� .00 �.20 �.76� —

�p< .05.

BULLYING ATTITUDES AND BEHAVIOR 793

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

et al., 2000). The reason for using multilevel analysis inthe present study was our nested data (students withinclassrooms) and variables measured on different levels.We defined two levels in our data: an individual level(student level) and a classroom level (Lee, 2000). The rea-son that we did not use the school as a third level is thatwe did not have specific theoretical hypotheses aboutschool-level factors. Another reason for not examiningschool-level factors was that the 109 classroom camefrom 43 different schools and, in a number of schools,we had data from only one classroom. A school-level fac-tor would in those cases represent the same informationand variability as the classroom-level factor.

In our multilevel models, individual bullying was thedependent variable. The independent variables at theindividual level were gender, social preference, numberof reciprocal friends, and antibullying attitudes. The inde-pendent variables at the classroom level were classroomlevels of bullying and antibullying attitudes. We centeredour independent classroom-level variables around theirgrand mean before they were entered in the multilevelprediction (i.e., for every participant the grand mean wassubtracted from the raw scores). Gender was entered inthe multilevel prediction as a dummy variable, with code0 referring to girls and code 1 referring to boys.

Empty model. The empty model contains noexplanatory variables and gives an estimation of thevariances to be explained at the individual and at theclassroom level. For bullying, the variance betweenadolescents within classrooms was estimated to be .20and the variance between classrooms was .02. Therewas a statistically significant variability betweenclassrooms, as indicated by the variance coefficientand its standard error (p< .05). The intraclasscorrelation was .09, meaning that 9% of the variancein adolescents’ bullying scores consisted of variancebetween classrooms.

Individual-level model. Next, we tested theindividual-level predictors of bullying. These variableswere gender, number of reciprocal friends, social prefer-ence, and antibullying attitudes and were entered in theprediction simultaneously. Regression coefficients andstandard errors are reported in Table 3. For exploratoryreasons, interactions with gender were tested but failedto reach statistical significance. We also tested forrandom slopes, that is, differences between classroomsin the effect of a predictor on individual bullying. Forall predictors, the random slopes were not significantand are therefore not reported here.

All individual predictors were statistically significant(p< .05), as indicated by the t test. As expected, boysbullied more than girls. Moreover, having reciprocal

friends was associated positively with bullying, whereassocial preference was negatively related to bullying, indi-cating that socially preferred adolescents bullied lessthan adolescents who were disliked by their classmates.It should be noted that the effects of number of recipro-cal friends and social preference were small. In addition,antibullying attitudes were associated with less bullyingbehavior. The variability between adolescents withinclasses for bullying was reduced from .20 to .15, indicat-ing that the individual-level variables explained 25% ofthe initial variability in bullying between adolescentswithin classrooms (i.e., 5 of 20, thus 5=20� 100%¼25%). In addition, a chi-square test revealed a significantdrop in deviance of the model after the individual-levelvariables were entered in the prediction of bullying(see Table 3), indicating that the individual-level modelsignificantly fitted the data.

Full model. The full model reveals whether differ-ences in classroom context are related to individualbullying. The classroom context was defined by class-room levels of antibullying attitudes and classroomlevels of bullying, both consisting of the averaged scoreson these variables of all adolescents in a classroom withthe target adolescent score left out. These variables wereentered in the prediction in a grand-mean centered form,so that a score of zero referred to the overall averagelevel of these variables.

When classroom antibullying attitudes were enteredin the prediction, the model for bullying improvedsignificantly, as indicated by the significant regression

TABLE 3

Multilevel Model for Bullying

Bullying

Individual Level Final Model

Predictor b SE b SE

Intercept 2.88�� .06 2.84�� .06

Gender (Boys) .10�� .02 .11�� .02

N of Reciprocal Friends .02�� .01 .02� .01

Social Preference �.03�� .01 �.03�� .01

Individual Antibullying Attitudes �.37�� .01 �.36�� .01

Deviance Dropa (df¼ 4) 776.99��

Classroom Antibullying Attitude �.12� .06

Classroom Bullying Behavior .30�� .08

Deviance Dropb (df¼ 3) 31.66��

aDifference in deviance between empty model and individual-level

model has a Chi-squared distribution and serves as an indication for

model fit.bDifference in deviance between individual-level model and full

model has a Chi-squared distribution and serves as an indication for

model fit.�p< .05. ��p< .01.

794 SCHOLTE, SENTSE, GRANIC

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

coefficients for classroom attitudes (b¼�.31). Becausewe controlled for individual attitudes, the regressioncoefficient for classroom antibullying attitudes refersto the effect of classroom attitude over and above theeffect of individual attitudes. The model shows thatadolescents in classrooms in which a negative attitudetoward bullying was dominant were less likely to bullythan adolescents in classrooms with a more positive atti-tude toward bullying. Of importance, after classroombehavior (i.e., bullying) was entered in the predictionof adolescents’ bullying scores, the association betweenclassroom antibullying attitudes and individual bullyingreduced substantively, from b¼�.31 to b¼�.12 (seeright column of Table 3 for this final model). This sug-gested that classroom bullying behavior partiallymediated the association between classroom antibully-ing attitudes and individual bullying behavior, reducingthe initial regression coefficient of classroom attitudesby 61%. Classroom bullying behavior was positivelyassociated to individual bullying (see Table 3). To testwhether classroom bullying behavior significantlymediated the effect of classroom antibullying attitudes,we performed the Sobel test using a bootstrap approach(cf. Preacher & Hayes, 2004). This test confirmed ourinitial interpretation of the multilevel analysis, with thesuggested mediation being highly significant (z¼�4.58,p< .001).

In this final model, the variability between classroomswas not significant anymore. We explained all the initialbetween-classroom variance, which was 7% of the initialtotal variance in bullying. The deviance of the full modelcompared to the individual-level model showed a signifi-cant drop as indicated by the chi-square test (seeTable 3), indicating that this model fit the data signifi-cantly better than the individual-level model.

DISCUSSION

The main aim of the present study was to examine towhat extent classroom attitudes and behavior were asso-ciated with individual bullying behavior, after control-ling for individual difference characteristics. Findingsignificant associations could have important implica-tions for theorizing about bullying, as it would meanthat conceptual models that only include individual dif-ference characteristics would be restricted because theyneglect the effects of contextual influences. Our studyis among the first to examine behavior and attitudes atthe classroom level simultaneously in relation to individ-ual bullying.

In the present study, we demonstrated that bothindividual-level factors and the classroom context aresignificantly related to adolescents’ bullying. Asexpected, individual antibullying attitudes were related

to bullying, which is in line with previous studies(Andreou et al., 2005; Boulton et al., 1999; Eslea &Smith, 2000). Furthermore, boys were more likely tobe bullies than girls (e.g., Bauer et al., 2007; Nanselet al., 2001; Scholte et al., 2007). In addition, the numberof friends and social preference were both weakly yetsignificantly related to bullying. The fact that adoles-cents who had more reciprocal friends were more likelyto bully suggest that these friends may somehowreinforce bullies in their behavior, for example, byattributing a higher social status to these bullies (cf.Pellegrini & Long, 2002; Salmivalli et al., 1997).Nevertheless, given the small effects, replication of thesefindings in other studies is warranted.

In support of our first hypothesis, we found that theclassroom context in terms of classroom antibullyingattitudes was indeed significantly related to bullying atthe individual level, even after controlling for individualantibullying attitudes and other important individualdifference characteristics such as gender, social prefer-ence, and number of reciprocal friends. This findingcorroborates earlier findings on the importance ofclassroom attitudes in relation to bullying (Salmivalli& Voeten, 2004) or aggressive behavior in general(Henry et al., 2000). We found that adolescents in class-rooms that held permissive attitudes toward bullyingwere more likely to bully themselves. It seems that theway classmates in the classroom think about bullyingin general affects how adolescents themselves areinvolved. This may provide empirical support for thepossible effectiveness of targeting classroom attitudesin antibullying intervention programs to decrease bully-ing in schools (e.g., see Smith & Ananiadou, 2003, for anoverview of programs).

The second major finding was that in addition toclassroom attitudes, the general level of bullying in theclassroom was significantly related to individual bully-ing. Two points are essential here as they highlight theimportance of classroom behavior relative to classroomattitudes. First, the results provide a first indication thatthe general level of behavior in the classroom might beas important, if not more important, than classroomattitudes to drive individual bullying involvement.1 Ingeneral, studies on explicit attitudes toward bullyingreport that school classes generally hold a negative atti-tude toward bullying (e.g., Eslea & Smith, 2000). Thereare, however, bullies in almost every school class despitethe presence of a dominating negative classroom atti-tude. When bullies get away with their behavior without

1Although the coefficients were unstandardized, the classroom

attitudes and behavior were assessed at the same scale level and thus

the unstandardized coefficients of bullying and victimization did offer

an indication of the importance of classroom behavior relative to

classroom attitudes.

BULLYING ATTITUDES AND BEHAVIOR 795

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

being punished or are even rewarded by an increase insocial status or dominance (cf. Pellegrini & Long,2002; Sentse et al., 2007), this might be a strong signaltoward classmates counteracting the possible effects ofgenerally held negative attitudes.

Second, when classroom behavior was entered intothe prediction of individual bullying, there was a sub-stantial drop in the association between classroom atti-tudes and individual-level bullying. In other words, theeffects of classroom bullying attitudes might be partlymediated by classroom behaviors. Indirect proof comesfrom the 61% drop in the effect of classroom attitudesafter the inclusion of classroom behavior. An additionalformal test of mediation confirmed this finding. Eventhough attitudes seem to be important predictors ofwhat adolescents do or experience at the individual level,once it is taken into account what actually happens inthe class in terms of behavior, attitudes seem to be lessrelevant. In this respect, it seems as if actions (i.e., whatclassmates do) speak louder than words (i.e., what theythink about bullying). However, because our study wascross-sectional, we cannot make conclusions about caus-ality due to the correlational design of the study.Although our findings seem to suggest that classroombullying behavior levels mediate classroom attitudes,longitudinal studies on bullying are needed to test actualmediation and to describe any causal links.

The positive association between classroom bullyingbehavior and adolescents’ bullying is consistent withthe results from studies on peer group behavior andinvolvement in bullying (Espelage et al., 2003). The find-ing that higher levels of bullying on a classroom level arerelated to higher levels of individual bullying suggeststhat adolescents reinforce bullying behaviors in theirclassmates. This would be in line with research showingthat peers are important socializing agents on a numberof behaviors. For example, an increasing number ofstudies on aggression have revealed that peer contagion(i.e., peer influence affecting deviant behavior of anindividual) exists in early childhood (e.g., Goldstein,Arnold, Rosenberg, Stowe, & Ortiz, 2001; Hanish,Martin, Fabes, Leonard, & Herzog, 2005), middle child-hood (Boxer, Guerra, Huesmann, & Morales, 2005;Warren, Schoppelrey, Moberg, & McDonald, 2005),and adolescence (e.g., Cohen & Prinstein, 2006;Espelage et al., 2003).

There are several mechanisms whereby classroomnorms could influence an individual’s behaviors.Through observation, individuals may learn thataggression or bullying others is rewarding as it may leadto increased social status or dominance within the class(Cohen & Prinstein, 2006; Pellegrini & Long, 2002).Furthermore, adolescents may desire to befriend avalued peer and could take over beliefs and behaviorsof that peer to increase the likelihood of actually

becoming friends (Hanish et al., 2005; Mrug, Hoza, &Bukowski, 2004). In addition, according to the socialidentity theory (Tajfel & Turner, 1979) adolescentscould imitate the behavior and beliefs from peers thatexemplify the identity they want to hold (Duffy &Nesdale, 2009; Olaja & Nesdale, 2004). It has also beensuggested that behavioral norms foster compliance andconformity from group members (Juvonen & Galvan,2009). As a result, adolescents may bully to avoidexclusion from a peer group. Moreover, the mere factthat individuals are victimized, or even witness victimi-zation, may predict their levels of aggression (Boxer,Edwards-Leeper, Goldstein, Musher-Eizenman, &Dubow, 2003). Finally, Boxer et al. (2005) describedthe principle of discrepancy-proportional peer influence,showing that in intervention groups, the level ofaggression of individual children changed in the peergroup’s average.

Study Limitations

A limitation of our study concerns the definition of thesocial context. We defined the context as the classroom,since bullying takes place mainly in school classes(Smith & Brain, 2000). For statistical reasons (i.e., manyschools in our study consisted of only one class, andthus the class and school level were overlapping) wecould not include the larger school context as a possiblecorrelate of bullying. Furthermore, other levels of thesocial context than just individual and classroom factorsare involved in bullying (cf. social-ecological model;Swearer & Espelage, 2004). For example, it has beensuggested that the school climate is substantially relatedto bullying and victimization (see Orpinas & Horne,2006). Frequently implemented intervention programssuch as Olweus’s antibullying program (Olweus, 1994;see Smith & Ananiadou, 2003) and the program conduc-ted by Salmivalli et al. (2005) apply a multilevelapproach, focusing on the individual, classroom andschool level. Our study focused on the norms and atti-tudes on the classroom level. Although classroomsnorms may be important, many of the mechanisms thatexplain the link between these norms and individualbehavior pertain to smaller peer groups rather thanthe larger class. As recent research indicates, normsand attitudes of smaller peer groups inside a class, inaddition to classroom level norms, affect individualbehavior (Duffy & Nesdale, 2009).

Implications for Research, Policy, and Practice

Our study had a number of strengths such as a largesample, multi-informant data (i.e., self-report and peerreport data) and a multilevel approach to account forour nested data. It provided new insights into factors

796 SCHOLTE, SENTSE, GRANIC

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

that are related to bullying. It also has practical implica-tions for further research. Themajority of studies on bully-ing have described individual difference characteristics ofchildren and adolescents as potential predictors of bully-ing involvement. Our study reveals that these characteris-tics may be important and remain significant but that thelarger social context in which bullying takes place shouldbe included as well in order to obtain a more completeunderstanding of this social problem. Previous studies thathave taken the class context into account have usuallyfocused on classroom attitudes (Henry et al., 2000;Salmivalli & Voeten, 2004). As our study showed thatthe magnitude of the association of attitudes with individ-ual bullying decreased substantially when controlling forclassroom behavior, it is likely that these studies may haveoverestimated the effects of classroom attitudes.

Studies on the association between bullying attitudesand bullying have exclusively focused on explicit atti-tudes, assessed in survey studies. Given that manyschools advocate an antibullying climate and employstrategies to establish nonpermissive attitudes amongadolescents toward bullying in school, assessing explicitattitudes through survey studies might evoke sociallydesirable answers. Recent research suggests that indivi-duals’ implicit attitudes may be important predictorsof behavior as well (e.g., Greenwald & Banaji, 1995).This has been shown to be the case with aggression(Gollwitzer, Banse, Eisenbach, & Naumann, 2007) anddiscrimination (Fazio & Olson, 2003). An importantnext step in bullying research would be to examine towhat extent implicit attitudes, in addition to explicit atti-tudes, are related to individual bullying. It may well bethat implicit attitudes interact with explicit attitudes,which would provide a further explanation for whysome adolescents who report to hold negative attitudestoward bullying nevertheless encourage bullying beha-viors of others or even bully themselves (Goethem,Scholte, & Wiers, in press).

To more fully understand bullying and the role ofbehavioral norms and attitudes, future research mightalso include the role of teachers and parents, as well assmaller peer groups. Empirical studies that simul-taneously examine all these levels are lacking. It wouldbe highly relevant to examine to what extent the schoolclimate and family factors predict individual bullying orvictimization, in addition to significant individual andclassroom factors that have been identified in thepresent and previous studies. Finally, future researchshould compare the effects of attitudes and behavioracross different age groups to be conclusive on theimportance of the classroom context for children before,during, and after the transition from elementary tosecondary school.

Our study also has important implications for bully-ing interventions. Many bullying interventions focus on

changing students’ attitudes toward bullying. Althoughthis focus seems justified by the fact that attitudes arerelated to bullying (Salmivalli & Voeten, 2004), our find-ings imply that this focus may be too narrow. Given thatour study suggests that the effects of attitudes aremediated by what is actually happening in class in termsof bullying (i.e., classroom bullying behavior), inter-vention programs should also direct their focus ondiminishing the actual bullying behavior. Moreover,because classroom attitudes and bullying behavior weresignificantly related to bullying after controlling forindividual-level attitudes, our study also implies thatfocusing on individual students (e.g., bullies) may notbe appropriate. Rather, antibullying interventionsshould target classrooms as a whole, with all class mem-bers included. This may be especially important justafter the transition from primary to secondary schools,when entering a new social group, adolescents will tryto adjust to their new group and might be inclined todisplay behaviors that are socially rewarded in termsof increased dominance (cf. Pellgrini & Long, 2002).

REFERENCES

Andreou, E. (2004). Bully=victim problems and their association with

psychological constructs in 8- to 12-year-old Greek school children.

Aggressive Behavior, 26, 49–56.

Andreou, E., Vlachou, A., & Didaskalou, E. (2005). The roles of

self-efficacy, peer interactions and attitudes in bully-victim incidents:

Implications for intervention policy-practices. School Psychology

International, 26, 545–562.

Bauer, N. S., Lozano, P., & Rivara, F. P. (2007). The effectiveness of

the Olweus Bullying Prevention program in public middle schools: A

controlled trial. Journal of Adolescent Health, 40, 266–274.

Bellmore, A. D., Witkow, M. R., Graham, S., & Juvonen, J. (2004).

Beyond the individual: The impact of ethnic context and classroom

behavioural norms on victims’ adjustment. Developmental

Psychology, 40, 1159–1172.

Boulton, M., Bucci, E., & Hawker, D. (1999). Swedish and English

secondary school pupils’ attitudes towards, and conceptions of,

bullying: Concurrent links with bully=victim involvement. Scandina-

vian Journal of Psychology, 40, 277–284.

Boulton, M. J., & Smith, P. K. (1994). Bully=victim problems in

middle school children: Stability, self-perceived competence, peer

perceptions and peer acceptance. British Journal of Developmental

Psychology, 12, 315–329.

Boulton, M. J., Trueman, M., & Flemington, I. (2002). Associations

between secondary school pupils’ definitions of bullying, attitudes

towards bullying, and tendencies to engage in bullying: Age and

sex differences. Educational Studies, 28, 353–370.

Boxer, P., Edwards-Leeper, L., Goldstein, S. E., Musher-Eizenman,

D., & Dubow, E. F. (2003). Exposure to ‘‘low level’’ aggression in

school: Associations with aggression, future expectations, and per-

ceived safety. Violence and Victims, 18, 691–705.

Boxer, P., Guerra, N. G., Huesmann, L. R., & Morales, J. (2005).

Proximal peer-level effects of small-group selected prevention on

aggression in elementary school children: An investigation of the

peer contagion hypothesis. Journal of Abnormal Child Psychology,

33, 325–338.

BULLYING ATTITUDES AND BEHAVIOR 797

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

Caravita, S. C., Di Blasio, P., & Salmivalli, C. (2009). Unique and

interactive effects of empathy and social status on involvement in

bullying. Social Development, 18, 140–163.

Cialdini, R. B., Kallgren, C. A., & Reno, R. R. (1991). A focus theory

of normative conduct. Advances in Experimental Social Psychology,

24, 201–234.

Cohen, G. J., & Prinstein, M. J. (2006). Peer contagion of aggression

and health risk behavior among adolescent males: An experimental

investigation of effects on public conduct and private attitudes.

Child Development, 77, 967–983.

Coie, J. D., & Dodge, K. A. (1983). Continuities and changes in chil-

dren’s social status: A five-year longitudinal study. Merrill-Palmer

Quarterly, 29, 261–282.

Coyne, I., Craig, J., & Chong, P. S. (2004). Workplace bullying in a

group context. British Journal of Guidance and Counseling, 32,

301–317.

Due, P., Holstein, B. E., Lynch, J., Diderichsen, F., Nic Gabhein, S.,

Scheidt, P., et al. (2005). Bullying and symptoms among school-aged

children: International comparative cross sectional study in 28 coun-

tries. European Journal of Public Health, 15, 128–132.

Duffy, A. L., & Nesdale, D. (2009). Peer groups, social identity, and

children’s bullying behavior. Social Development, 18, 121–139.

Eslea, M., Menesini, E., Morita, Y., O’Moore, M., Mora-Merchan, J.

A., Pereira, B., et al. (2003). Friendship and loneliness among bullies

and victims: Data from seven countries. Aggressive Behavior, 30,

71–83.

Eslea, M., & Smith, P. K. (2000). Pupil and parent attitudes towards

bullying in primary schools. European Journal of Psychology of

Education, 15, 207–219.

Espelage, D. L., Holt, M. K., & Henkel, R. R. (2003). Examination of

peer-group contextual effects on aggression during early ado-

lescence. Child Development, 74, 205–220.

Fazio, R. H., & Olson, M. A. (2003). Implicit measures in social

cognition research: Their meaning and uses. Annual Review of

Psychology, 54, 297–327.

Goethem, A. van, Scholte, R. H. J., & Wiers, R. (in press). Explicit-

and implicit bullying attitudes in relation to bullying behavior.

Journal of Abnormal Child Psychology.

Goldstein, N. E., Arnold, D. H., Rosenberg, J. L., Stowe, R. M., &

Ortiz, C. (2001). Contagion of aggression in day care classrooms

as a function of peer and teacher responses. Journal of Educational

Psychology, 93, 708–719.

Gollwitzer, M., Banse, R., Eisenbach, K., & Naumann, A. (2007).

Effectiveness of the Vienna Social Competence Training on explicit

and implicit aggression: Evidence from an Aggressiveness–IAT.

European Journal of Psychological Assessment, 23, 150–156.

Goossens, F. A., Olthof, T., & Dekker, P. H. (2006). New participant

role scales: Comparison between various criteria for assigning

roles and indications for their validity. Aggressive Behavior, 32,

343–357.

Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition:

Attitudes, self-esteem, and stereotypes. Psychological Review, 102,

4–27.

Hanewinkel, R., & Eichler, D. (1999). Ergebnissse einer Intervention-

studie zur Pravention schulischer Gewalt. In M. Schafer & D. Frey

(Eds.), Aggression und Gewalt unter Kindern und Jugentlichen

(pp. 245–264). Gottingen, Germany: Hogrefe Verlag.

Hanish, L. A., Martin, C. L., Fabes, R. A., Leonard, S., & Herzog, M.

(2005). Exposure to externalizing peers in early childhood: Homo-

phily and peer contagion processes. Journal of Abnormal Child

Psychology, 33, 267–281.

Hartup, W. W. (1996). The company they keep: Friendships and their

developmental significance. Child Development, 67, 1–13.

Henry, D. (2001). Classroom context and the development of

aggression: The role of normative processes. In F. Columbus

(Ed.), Advances in psychology research (Vol. 6, pp. 193–227).

Hauppage, NY: Nova Science.

Henry, D., Guerra, N., Huesmann, R., Tolan, P., Van Acker, R., &

Eron, L. (2000). Normative influences on aggression in urban

elementary school classrooms. American Journal of Community

Psychology, 28, 59–81.

Hodges, E. V. E, & Perry, D. G. (1999). Personal and interpersonal

antecedents and consequences of victimization by peers. Journal of

Personality and Social Psychology, 76, 677–685.

Juvonen, J., & Galvan, A. (2009). Bullying as a means to foster

compliance. In M. J. Harris (Ed.), Bullying, rejection, and peer victi-

mization: A social cognitive neuroscience perspective (pp. 299–318).

New York: Springer.

Kuppens, S., Grietens, H., Onghena, P., Michiels, D., & Subramanian,

S. V. (2008). Individual and classroom variables associated with

relational aggression in elementary-school aged children: A multile-

vel analysis. Journal of School Psychology, 46, 639–660.

Ladd, G. W. (1983). Social networks of popular, average, and rejected

children in school settings. Merrill–Palmer Quarterly, 29, 283–307.

Lee, V. E. (2000). Using hierarchical linear modeling to study social

contexts: The case of school effects. Educational Psychologist, 35,

125–141.

Menesini, E., Eslea, M., Smith, P. K., Genta, M. L., Giannetti, E.,

Fonzi, A., et al. (1997). Cross-national comparison of children’s atti-

tudes towards bully=victim problems in school. Aggressive Behavior,

23, 245–257.

Mrug, S., Hoza, B., & Bukowksi, W. M. (2004). Choosing or being

chosen by aggressive-disruptive peers: Do they contribute to

children’s externalizing and internalizing problems? Journal of

Abnormal Child Psychology, 32, 53–65.

Nansel, T. R., Craig, W., Overpeck, M. D., Pilla, R. S., Ruan, W. J.,

Simons-Morton, B., et al. (2001). Bullying behaviours among U.S.

youth: Prevalence and association with psychosocial adjustment.

Journal of the American Medical Association, 16, 2094–2100.

Nansel, T. R., Craig, W., Overpeck, M. D., Saluja, G., & Ruan, J.

(2004). Cross-national consistency in the relationship between bully-

ing behaviors and psychosocial adjustment. Archives of Pediatrics

and Adolescent Medicine, 158, 730–736.

Nesdale, D., Durkin, K., Maass, A., Kiesner, J., & Griffiths, J. A.

(2008). Effects of group norms on children’s intentions to bully.

Social Development, 17, 889–907.

Nesdale, D., Milliner, E., Duffy, A., & Griffiths, J. A. (2009). Group

membership, group norms, empathy, and young children’s inten-

tions to aggress. Aggressive Behavior, 35, 244–258.

Newcomb, A. F., & Bukowski, W. M. (1983). Social impact and social

preference as determinants of children’s peer group status. Develop-

mental Psychology, 19, 856–867.

Olaja, K., & Nesdale, D. (2004). Bullying and social identity: The

effects of group norms and distinctiveness threat on attitudes

towards bullying. British Journal of Developmental Psychology, 22,

19–35.

Olweus, D. (1989). The Olweus Bully=victim Questionnaire. Bergen,

Norway: Mimeo.

Olweus, D. (1993). Bullying at school. Cambridge, MA: Blackwell.

Olweus, D. (1994). Annotation: Bullying at school: Basic facts and

effects of a school based intervention program. Journal of Child

Psychology and Psychiatry, 35, 1171–1190.

Olweus, D., & Endresen, I. M. (1998). The importance of sex-of-stimu-

lus object: Age trends and sex differences in empathetic responsive-

ness. Social Development, 3, 370–388.

Olweus, D., & Limber, S. (1999). Bullying Prevention Programme. In

Blueprints for violence prevention, Vol. 9. Golden, CO: Venture.

Orpinas, P., & Horne, A. M. (2006). Bullying prevention: Creating a

positive school climate and developing social competence. Washing-

ton, DC: American Psychological Association.

798 SCHOLTE, SENTSE, GRANIC

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014

Pellegrini, A. D. (1998). Bullies and victims in school: A review and

call for research. Journal of Applied Developmental Psychology, 19,

165–176.

Pellegrini, A. D., Bartini, M., & Brooks, F. (1999). School bullies,

victims, and aggressive victims: Factors relating to group affiliation

and victimization in early adolescence. Journal of Educational

Psychology, 91, 26–224.

Pellegrini, A. D., & Long, J. D. (2002). A longitudinal study of

bullying, dominance, and victimization during the transition from

primary school through secondary school. British Journal of

Developmental Psychology, 20, 259–280.

Pepler, D., Craig, W., Ziegler, S., & Charach, A. (1994). An evaluation

of an anti-bullying intervention in Toronto schools. Canadian

Journal of Community Mental Health, 13, 95–110.

Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for

estimating indirect effects in simple mediation models. Behavior

Research Methods, Instruments, and Computers, 36, 717–731.

Rasbash, J., Browne, W., Goldstein, H., Yang, M., Plewis, I., Healy,

M., et al. (2000). A user’s guide to MlwiN. London: Institute of

Education.

Rigby, K. (2004). Addressing bullying in schools: Theoretical perspec-

tives and their implications. School Psychology International, 25,

287–300.

Rigby, K., & Slee, P. (1991). Bullying among Australian school chil-

dren: Reported behavior and attitudes towards victims. Journal of

Social Psychology, 131, 615–627.

Salmivalli, C., Huttunen, A., & Lagerspetz, K. M. J. (1997). Peer net-

works and bullying in schools. Scandinavian Journal of Psychology,

38, 305–312.

Salmivalli, C., Kaukiainen, A., & Voeten, M. (2005). Anti-bullying

intervention: Implementation and outcome. British Journal of

Educational Psychology, 75, 465–487.

Salmivalli, C., Lagerspetz, K., Bjorkqvist, K., Osterman, K., &

Kaukiainen, A. (1996). Bullying as a group process: Participant roles

and their relations to social status within the group. Aggressive

Behavior, 22, 1–15.

Salmivalli, C., & Voeten, M. (2004). Connections between attitudes,

group norms, and behavior in bullying situations. International

Journal of Behavioral Development, 28, 246–258.

Samara, M., & Smith, P. K. (2008). How schools tackle bullying, and

the use of whole school policies: Changes over the last decade.

Educational Psychology, 28, 663–676.

Schafer, M., Korn, S., Brodbeck, F. C., Wolke, D., & Schultz, H.

(2005). Bullying roles in changing contexts: The stability of victim

and bully roles from primary to secondary school. International

Journal of Behavioral Development, 29, 323–335.

Scholte, R. H. J., Engels, R. C. M. E., Overbeek, G., de Kemp, R. A.

T., & Haselager, G. J. M. (2007). Stability in bullying and victimiza-

tion and its association with social adjustment in childhood and ado-

lescence. Journal of Abnormal Child Psychology, 35, 217–228.

Scholte, R. H. J., Overbeek, G., ten Brink, G., Rommes, E., de Kemp,

R. A. T., Goossens, L., et al. (2009). The significance of reciprocal

and unilateral friendships for peer victimization in adolescence.

Journal of Youth and Adolescence, 38, 89–100.

Sentse, M., Scholte, R. H. J., Salmivalli, C., & Voeten, M. (2007).

Person–group Dissimilarity in involvement in bullying and its

relation with social status. Journal of Abnormal Child Psychology,

35, 1009–1019.

Smith, P. K., & Ananiadou, K. (2003). The nature of school bullying

and the effectiveness of school-based interventions. Journal of

Applied Psychoanalytic Studies, 5, 189–209.

Smith, P. K., & Brain, P. (2000). Bullying in schools: Lessons from two

decades of research. Aggressive Behavior, 26, 1–9.

Smith, P. K., Pepler, D., & Rigby, K. (Eds.). Bullying in schools: How

successful can interventions be? New York: Cambridge University

Press.

Smith, P. K., & Sharp, S. (Eds.). (1994). School bullying: Insights and

perspectives. London: Routledge.

Smith, P. K., Singer, M., Hoel, H., & Cooper, C. L. (2003). Victimiza-

tion in the school and the workplace: Are there any links? British

Journal of Psychology, 94, 175–188.

Stevens, V., De Bourdeaudhuij, I., & Oost, P. van. (2000). Bullying in

Flemish schools: An evaluation of anti-bullying intervention in

primary and secondary schools. British Journal of Educational

Psychology, 70, 195–210.

Stevens, V., Oost, P. van, & De Bourdeaudhuij, I. (2000). The effects of

an anti-bullying intervention programme on peers’ attitudes and

behaviour. Journal of Adolescence, 23, 21–34.

Stormshak, E. A., Bierman, K. L., Vruschi, C., Dodge, K. E., & Coie,

J. D. (1999). The relation between behavior problems and peer

preference in different classroom contexts. Child Development, 70,

169–182.

Sutton, J., & Smith, P. K. (1999). Bullying as a group process: An

adaptation of the participant role approach. Aggressive Behavior,

25, 97–111

Sutton, J., Smith, P. K., & Swettenham, J. (1999). Social cognition and

bullying: Social inadequacy or skilled manipulation? British Journal

of Developmental Psychology, 17, 435–450.

Swearer, S. M., & Espelage, D. L. (2004). Introduction: A

social-ecological framework of bullying among youth. In D. L.

Espelage & S. M. Swearer (Eds.), Bullying in American schools: A

social-ecological perspective on prevention and intervention

(pp. 1–12). Mahwah, NJ: Erlbaum.

Swenson, L. P. & Rose, A. J. (2009). Friends’ knowledge of youth

internalizing and externalizing adjustment: Accuracy, bias, and the

influences of gender, grade, positive friendship quality, and self

disclosure. Journal of Abnormal Child Psychology, 37, 887–901.

Tajfel, H., & Turner, J. (1979). An integrative theory of intergroup

conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology

of intergroup relations (pp. 33–47). Monterey, CA: Brooks=Cole.

Warren, K., Schoppelrey, S., Moberg, D. P., & McDonald, M. (2005).

A model of contagion through competition in the aggressive

behaviors of elementary school students. Journal of Abnormal Child

Psychology, 33, 283–292.

BULLYING ATTITUDES AND BEHAVIOR 799

Dow

nloa

ded

by [

Uni

vers

ity o

f C

alif

orni

a Sa

nta

Cru

z] a

t 05:

19 1

9 N

ovem

ber

2014