Bullying Among Adolescent Football Players:Role of Masculinity and Moral Atmosphere

Jesse A. Steinfeldt, Ellen L. Vaughan, andJulie R. LaFollette

Indiana University-Bloomington

Matthew C. SteinfeldtFort Lewis College

Identifying practices of masculinity socialization that contribute to the establishment ofgender privilege can help address violence and bullying in schools (Connell, 1996).Because the sport of football is considered an important contributor to masculinityconstruction, establishing peer networks, and creating hierarchies of student status, thisstudy examined the influence of social norms (i.e., moral atmosphere, meanings ofadolescent masculinity) on bullying beliefs and behaviors of 206 high school footballplayers. Results demonstrated that moral atmosphere (Peer Influence, Influential MaleFigure) and adherence to male role norms significantly predicted bullying, but thestrongest predictor was the perception of whether the most influential male in a player’slife would approve of the bullying behavior. In addition to prevention interventionshighlighting the role of influential men and masculinity norms in this process, impli-cations for practice suggest that football players can use their peer influence and statusas center sport participants to create a school culture that does not tolerate bullying.

Keywords: meanings of adolescent masculinity, gender role norms, high school student-athletes,moral functioning, aggression, American football

School bullying is gaining recognition as aserious threat to the mental and physical healthof adolescents around the world, occurringacross a diverse demographic spectrum (Al-uede, Adeleke, Omoike, & Afen-Akpaida,2008; Coggan, Bennett, Hooper, & Dickinson,2003; Wild, Flisher, Bhana, & Lombard, 2004).Little consensus exists in the literature concerningthe profile of a bully (Aluede et al., 2008; Kreager,2009; Vaillancourt, Hymel, & McDougall, 2003).However, one popular societal stereotype of abully is that of the socially powerful jock, mosttypically a football player. This stereotype haslong been present in popular culture (e.g., Re-venge of the Nerds), but a relevant contempo-rary example can be found in the TV showGlee. In Glee, the football players and cheer-

leaders enjoy social dominion over the highschool, while the lowly “gleeks” (i.e., kids inthe glee club) are continually harassed, humili-ated, and rejected by their popular peers. Thefootball players are largely portrayed as insen-sitive, unintelligent, bullies who enjoy torment-ing students lower in the social hierarchy.

This example illustrates how athletes, partic-ularly football players, are portrayed as perpe-trators of bullying and violence. One possibleexplanation for this portrayal is that the sport offootball not only values, encourages, andteaches instrumental aggression on the field(Messner, 1990, 2002), but it also serves as asite where boys learn values and behaviors (e.g.,competition, toughness, winning-at-all costs)that are considered to be culturally valued as-pects of masculinity (Messner, 1996). However,despite the aggressive and even violent natureof football on the field—combined with societalstereotypes of football players as aggressivebullies off the field—there is a dearth of empir-ical research that directly investigates bullyingin this domain. Thus, football represents aninteresting site to examine how football playersbalance the use of instrumental aggression onthe field with standards of manhood when they

This article was published Online First January 23, 2012.Jesse A. Steinfeldt, Ellen L. Vaughan, and Julie R. La-

Follette, Department of Counseling and Educational Psy-chology, Indiana University-Bloomington; Matthew C.Steinfeldt, Department of Athletics, Fort Lewis College.

Correspondence concerning this article should be ad-dressed to Jesse A. Steinfeldt, Department of Counselingand Educational Psychology, Indiana University, 201 NorthRose Avenue, Bloomington, IN 47403. E-mail: jesstein@indiana.edu

Psychology of Men & Masculinity © 2012 American Psychological Association2012, Vol. 13, No. 4, 340–353 1524-9220/12/$12.00 DOI: 10.1037/a0026645

340

are off the field. By using a model moral action(Rest, 1983, 1984) to understand social normsoperating within this unique context, this studyintended to determine if the norms of moralatmosphere and gender role norms associatedwith meanings of adolescent masculinity pre-dicted moral functioning and bullying behaviorsamong high school football players.

Bullying and School Violence

Bullying is an all too common phenomenon.National research suggests that greater than50% of students in Grades 6 to 10 are impactedby bullying (Wang, Iannotti, & Nansel, 2009).There is mounting evidence that bullyinggreatly influences the psychosocial functioningof both perpetrators and victims (Coggan et al.,2003; Dukes, Stein, & Zane, 2009; Espelage &Holt, 2001; Kim & Leventhal, 2008; McCabe,Antony, Summerfeldt, Liss, & Swinson, 2003;Storch, Bagner, Geffken, & Baumeister, 2004).Broadly defined, bullying is characterized asbehavior that is intended to inflict harm orstress, occurs repeatedly over a period of time,and involves an inequity of strength or power(Dukes et al., 2009). Bullying behavior is foundto increase when it is socially endorsed bygroup norms, and bullying is generally consid-ered more acceptable when it is directed towarda nongroup member (Duffy & Nesdale, 2009;Gini, 2008; Ojala & Nesdale, 2004). Bulliestend to identify other bullies as friends, suggest-ing that bullying behavior is socially reinforced(Espelage & Holt, 2001).

Bullying occurs in several forms, includingphysical (e.g., hitting, pushing, and kicking);verbal (e.g., making threatening remarks, name-calling, teasing); and social (e.g., social exclu-sion, rumor spreading). Another contemporaryform of bullying is cyberbullying, wherein thisform of social bullying is carried out usingtechnology such as cell phones and online socialnetworking sites (Dukes et al., 2009; Poteat &Rivers, 2010; Wang et al., 2009). Wang et al.(2009) reported a high prevalence rate for in-volvement in bullying, whether victim or per-petrator, among participants from a nationallyrepresentative sample of 7,182 adolescent stu-dents. Verbal bullying was the most common(53.6%) closely followed by social bullying(51.4%). Physical bullying was less frequent(20.8%) and 13.6% of students reported in-

volvement in cyberbullying, although given thatthis survey was completed in 2005, cyberbully-ing has likely increased as technology has ad-vanced. In 2007, Kowalski and Limber reportedthat 22% of a sample of 3,767 middle schoolstudents had experience with cyberbullying.

While the detrimental effects of bullying onvictims has long been recognized and studied,the impact that bullying has on perpetrators isnot as well understood (Dukes et al., 2009).Both bullies and their victims are at greater riskfor social anxiety, lower self-esteem, depres-sion, hopelessness, and suicidal ideation (Cog-gan et al., 2003; Dukes et al., 2009; Espelage &Holt, 2001; Jankauskiene, Kardelis, Sukys, &Kardeliene, 2008; Kim & Leventhal, 2008;Storch et al., 2004). Bulling behavior has alsobeen associated with increased alcohol and druguse among college students (Storch et al.,2004). Some risk factors associated with agreater likelihood of the development of bully-ing behavior include: having unmarried parents,having maternal authoritarian or permissiveparenting, having a tendency toward moral dis-engagement versus moral sensibility, and beingof the male gender (Gini, 2006; South & Wood,2006; Underwood, Bernon, & Rosen, 2009).

In regard to gender influence on bullying,research suggests that bullies tend to be boys(Beaty & Alexeyev, 2008) and that using ho-mophobic language is particularly commonamong male bullies (Poteat & Rivers, 2010).“Bullying is a cycle that begins and ends withthe Boy Code, the strict rules of masculinity thatpunish boys who seem feminine, weak, or gayto their peers. An escape from teasing and bul-lying is to act tough, succeed at sports, hidetheir emotions, and avoid things traditionallythought of as “girly” or feminine” (Pollack,2000, p. 108). Thus, boys who worry that theyare not living up to the Boy Code may feelpressure to tease and even bully other boys inorder to avoid being victimized by those samebehaviors.

Masculinity Norms and Sport Participation

According to Connell (1996), identifying therituals and practices of masculine socializationthat contribute to the establishment of genderprivilege is an important step in addressing vi-olence and bullying in schools. In many highschools in the United States, sports are consid-

341BULLYING AMONG ADOLESCENT FOOTBALL PLAYERS

ered to be an important contributor to the estab-lishment of peer networks and to the creation ofhierarchies of student status (Kreager, 2007).Gilbert and Gilbert (1998) suggested that sportparticipation increases the likelihood that anadolescent male will be welcomed into peergroups, whereas young men who do not partic-ipate in sport often find difficulty gaining ac-ceptance from their peers and gaining access tosocial networks. While sport is considered to bea major contributor to a school’s gender regimeand social hierarchy, some sports wield greaterinfluence than others in this process.

In an attempt to distinguish between sportsthat have differential levels of power and pres-tige in promoting preferred modes of masculin-ity and social status, Messner (2002) classifiedsports as falling into the categories of centersports or margin sports. Valued more than mar-gin sports, center sports (e.g., football, basket-ball) are sports that are close to the institutionalcenter, based on their long traditions withinschool and community structures. Because ofcenter sports’ historical position in relation toinstitutional ties, sources of funding, and socialtradition, these sports create spaces wherenorms of masculinity are readily conveyed andbecome entrenched. Furthermore, these sportsenjoy a degree of insulation from pressures ofexternal social control that might otherwise dis-courage antisocial behavior by their athletes(Messner, 2002). For instance, by rewardingphysical aggression with on-the-field successand increased prestige in the sport, football isoften perceived as not only socially elevating itsparticipants above their peers, but also, “in-creasing [participants’] off-the-field violencetoward perceived outsiders and “weaker” stu-dents” (Kreager, 2007, p. 706).

In his ethnographic study of high school foot-ball in Texas, Foley (2001) described the waysthat football contributes to the social hierarchyand how it promotes dominant gender normsamong its participants, in the school, andthroughout the community. The manner inwhich football prioritizes certain forms of mas-culinity not only elevates football players tohigher levels of social status, but it also indi-rectly marginalizes nonparticipants (Foley,2001). Participation in football may affect boththe creation of masculinity norms enforced byteam culture, and the amount of pressure placedon athletes to conform to this norm. This pres-

sure to conform to team norms of masculinitymay contribute to aggressive behavior off thefield (Curry, 2000). “Reinforcement of mascu-line gender norms through strict football ritualsin team practice creates a behavioral repertoirethat may extend to other contexts, with gendernorms also informing the interpersonal relation-ships that athletes form off the field” (Gage,2008, p. 1016).

As mentioned earlier, a football player who isintimidating, dominating, and aggressive on thefield earns respect from his teammates andcoaches. According to Elman and McKelvie(2003), this recognition may contribute to aplayer believing he deserves this respect off thefield, which could result in bullying, fighting,and other instances of off-field violence. Someresearch has suggested that there exists a rela-tionship between participating in high contactsports (e.g., football, basketball) and relationalaggression, including fighting and violence to-ward women (Forbes, Adams-Curtis, Pakalka,& White, 2006; Kreager, 2009). In an attempt todetermine the extent to which participating inhigh school sports contributes to male violence,Kreager (2007) analyzed almost 6,400 highschool adolescent males in a secondary data set(i.e., National Longitudinal Study of AdolescentHealth). Kreager concluded that there exists “acontinuum of physical contact and masculinitywhereby highly masculinized contact sports in-crease the risks of violence” (p. 717). Resultssuggested that when compared to nonathletes,high school football players were 40% morelikely to get in a serious fight. Additionally,football players with a high proportion offriends who are their teammates were signifi-cantly more likely to engage in violent behav-iors than those without a similarly high percent-age of football friends. Thus, playing highschool football and being embedded within afootball network significantly increased the riskof serious violence for participants in this sam-ple (Kreager, 2007).

This research assessed participation in sportas a predictive factor, but it did not assessbeliefs about bullying by directly asking ado-lescents about these issues. While there hasbeen relatively limited research that directlyassesses the off-field dynamic of bullyingamong athletes, an abundance of research hasexamined beliefs about violence and aggressionon the fields of play (Shields & Bredemeier,

342 STEINFELDT, VAUGHAN, LAFOLLETTE, AND STEINFELDT

2007). Much of this research has focused onmoral functioning in order to explain antisocialbehavior that occurs during athletic competi-tions (Coulumb-Cabagno & Rascle, 2006;Kavussanu, Roberts, & Ntoumanis, 2002; Ka-vussanu, Stamp, Slade, & Ring, 2009; Sage,Kavussanu, & Duda, 2006; Shields, Brede-meier, Gardner, & Bostron, 1995; Shields,LaVoi, Bredemeier, & Power, 2007; Stuart &Ebbeck, 1995; Stephens & Kavanagh, 2003;Visek & Watson, 2005). According to Shieldsand Bredemeier (2007), the model most consis-tently used to empirically examine moral func-tioning in sport is Rest’s model of moral action(1983, 1984). This model operates under thepremise that within each action, there are fourprocesses: (a) interpreting the situation; (b)forming a moral judgment about the appropriatething to do; (c) deciding what one intends to doby selecting among competing values; and (d)engaging in the behavior. These processes areinteractive, and can be influenced by a varietyof factors (Rest, 1983, 1984). Much of the re-search on moral functioning in sport has exam-ined the last three processes (i.e., judgment,intention, behavior) of Rest’s model (Shields &Bredemeier, 2007). Because it assesses influ-ences of group norms in sport, this model mayrepresent an effective framework to directly ex-amine beliefs about off-field aggression (i.e.,bullying) among football players.

Current Study

While some research—along with pervasivesocietal stereotypes—suggests that playingfootball may contribute to higher rates of vio-lence off the field, there are limited empiricalstudies that directly examine bullying amongfootball players. Accordingly, the purpose ofthis study was to investigate bullying beliefsand behaviors among high school football play-ers. The first goal of the study was to examine ifmeanings of adolescent masculinity and thenorms of moral atmosphere influenced themoral functioning of high school football play-ers as it relates to bullying. We used Rest’s(1983, 1984) model of moral action in an at-tempt to replicate the design of Kavussanu et al.(2002) by adapting the on-field moral function-ing scenarios to be relevant to bullying. In orderto utilize their study’s framework to examineoff-field outcomes among student-athletes, we

substituted Meanings of Adolescent Masculin-ity (Oransky & Fisher, 2009) for the Motiva-tional Climate variable, and instead of Kavus-sanu et al.’s influence of the coach on moralatmosphere, we expanded this moral atmo-sphere norm by assessing the impact of the mostinfluential male in the life of the adolescent.Consistent with Kavussanu et al., the study’sfirst hypothesis was that lower levels of moralatmosphere norms—combined with higher en-dorsement of male role norms—would predictmoral functioning that endorsed a greater accep-tance of bullying, as indicated by higher scoresacross all three levels (i.e., judgment, intention,behavior) of Rest’s model.

The second goal of the study was to deter-mine whether moral atmosphere and meaningsof adolescent masculinity predicted the likeli-hood that these players would engage in any ofthe four bullying behaviors: (a) physical bully-ing; (b) relational cyberbullying; (c) verbal bul-lying—physical threat; and (d) social bully-ing— homophobic. In order to inform thedevelopment of bullying prevention, this studyaimed to test male role norms and the directeffects of the norms of moral atmosphere (i.e.,most influential male, peer influence) on bully-ing behaviors. Thus, the study’s second hypoth-esis was that higher adherence to adolescentmasculine norms, along with perceptions of ac-ceptability of bullying among peers and by themost influential male figure in the adolescent’slife, would predict higher levels of bullyingbehaviors among high school football players.

Method

Participants

The participants in this study were 206 highschool football players between 14 and 18 yearsof age who attended one of three high schools inthe Midwestern United States. The average ageof the participants was 15.77 (SD � 1.51), andthe sample consisted of 48 freshmen, 60 soph-omores, 41 juniors, and 57 seniors. The partic-ipants self-identified their race as EuropeanAmerican (83%), African American (4%), Mul-tiracial (2%), Asian American (2%), HispanicAmerican (3%), and American Indian (3%).Three percent of the participants did not reporta racial identification, and the racial composi-tion of the sample was consistent with the racial

343BULLYING AMONG ADOLESCENT FOOTBALL PLAYERS

demographics of the community. Forty-one ofthe participants played football at the freshmanlevel, 58 played at the junior varsity level, and107 played at the varsity level. When asked whothe most influential male figure in their life was,66% of the players responded it was their father,14% of the players indicated a brother, 8% ofthe players identified their coach, and 12% orthe players responded with “other” and filled insuch people as a teacher, grandfather, uncle,mother, or famous person (e.g., Brett Favre,Jesus).

Measures

Meanings of Adolescent Masculinity.The MAMS (Oransky & Fisher, 2009) is a27-item self-report instrument that uses a4-point Likert-type scale with possible re-sponses ranging from 1 (strongly disagree) to 4(strongly agree). By drawing items from thenarratives of boys, the MAMS is designed toassess endorsement of traditional male rolenorms among adolescents. Higher scores repre-sent higher levels of endorsement of traditionalmale role norms. The MAMS has four sub-scales: (a) Constant Effort (e.g., “Acting manlyshould be the most important goal for guys”);(b) Emotional Restriction (e.g., “If a guy isupset about something, he should hold it in”);(c) Heterosexism (e.g., “Being thought of as gaymakes a guy seem like less of a man”); and (d)Social Teasing (e.g., “In order to fit in, guysmust be able to tease other guys”). These fourtheoretically driven subscales were derivedfrom factor analytic procedures, and presentfour distinct patterns of norms associated withadolescent masculinity. Evidence for the valid-ity of the MAMS has been establishedthrough relationship with MAMS subscalesand measures of psychological adjustment(e.g., anxiety, self-esteem). Oransky and Fi-scher reported additional convergent validitybased on the MAMS’ statistically significantrelationship to other psychometrically vali-dated scales of masculinity, including the Ad-olescent Masculinity Ideology in Relation-ships Scale (MIRS; Chu, Porche, & Tolman,2005) and the Male Role Norm Scale (MRNS;Thompson & Pleck, 1986). In support of thescale’s reliability, Oransky and Fischer re-ported the following internal consistency co-efficients: Constant Effort � .79; Emotional

Restriction � .80; Heterosexism � .80; andSocial Teasing � .61. Consistent with thosefindings, the internal consistency coefficientsfor the current study were as follows: Con-stant Effort � .69; Emotional Restriction �.75; Heterosexism � .78; and Social Teas-ing � .56.

Moral functioning. In this current study,moral functioning was assessed by adapting ameasure used to assess on-field moral function-ing of student-athletes by Kavussanu et al.(2002). This framework has been previouslyreplicated to examine on-field moral function-ing with high school football players (Stein-feldt, Rutkowski, Vaughan, & Steinfeldt, 2011),and it was further adapted to assess bullyingbehaviors and attitudes among high school foot-ball players in this current study. For this modelassessing the moral functioning and moral at-mosphere variables, Kavussanu et al. (2002)utilized a Multitrait-Multimethod (MMTM)analysis to validate the two indicators (i.e.,coach, teammate) of moral atmosphere and thethree indicators (i.e., judgment, intention, be-havior) of moral functioning across the fourscenarios. Utilizing confirmatory factor analysisand structural equation modeling, the results oftheir MMTM analysis demonstrated support forthis model. Thus, results suggest that this modelis appropriate to use as a valid measure forassessing moral atmosphere and moral func-tioning in sport, and the results of our analysesindicate that the adaptations we made resultedin a model that statistically significantly fit thedata.

Through consultation with football coaches,school administrators, and psychologists, thefour scenarios in this measure were revised toreflect potential bullying scenarios that a highschool student might encounter. The adaptedscenarios used in this study are:

(a) Physical Bullying: A classmate is walking down thehall. You bump your shoulder into him as he passes sothat his books fall out of his hands onto the floor.

(b) Relational Cyberbullying: You take a picture of aclassmate while he is crying and forward it to otherstudents.

(c) Verbal Bullying—Physical Threat: You threaten tophysically harm a classmate if he doesn’t allow you tosit where you want in the lunchroom.

(d) Social Bullying—Homophobic: You call a class-mate a degrading name that questions his sexual ori-entation in front of other students.

344 STEINFELDT, VAUGHAN, LAFOLLETTE, AND STEINFELDT

These scenarios were assessed across the lastthree levels (i.e., judgment, intention, behavior)of Rest’s (1983, 1984) model of moral action.For the first level, Judgment, participants wereasked after each scenario to judge how appro-priate they deemed the behavior to be. Theitems used a 5-point Likert-type scale with pos-sible responses ranging from 1 (never appropri-ate) to 5 (always appropriate). Scores weresummed across all four scenarios to provide atotal score for Judgment. The internal consis-tency coefficient for the Judgment scale was.87, higher than the Cronbach’s alpha of .67reported by Kavussanu et al. (2002) for thisscale. A similar procedure was followed for theintention and behavior levels. Participants wereasked if they would engage in the behavior (i.e.,intention), and were then asked how often theyhad actually engaged in the behavior in the pasttwo months (i.e., behavior). Both Intention andBehavior were assessed using a 5-point Lik-ert-type scale with possible responses rangingfrom 1 (never) to 5 (very often). Scores forthese scales are also summed across all fourscenarios. The internal consistency coeffi-cients for the Intention scale (� � .81) andBehavior scale (� � .75) were also consistentwith the Cronbach’s alphas of .72 and .70reported by Kavussanu et al. (2002) on therespective scales.

Moral atmosphere. Moral atmospherecan be considered the collective norms that con-tribute to the unique construction of a group’sethos (i.e., moral climate; Kavussanu et al.,2002; Shields et al., 2007). In this study, moralatmosphere was comprised of two dimensionsof moral influence: Peer Influence norms, andthe norms established by the participant’s MostInfluential Male. To assess Peer Influencenorms across the four scenarios of moral func-tioning, participants were asked how many oftheir friends they thought would engage in thedescribed behavior. A 5-point Likert-type scalewas used, with possible responses including 1(none), 2 (a few), 3 (about half), 4 (most), and 5(all). To assess the norms of the Most Influen-tial Male variable across the four scenarios,participants were asked how they thought themost influential male in their life would viewthe behavior in the scenario. A 5-point Likert-type scale was used, with possible responsesranging from 1 (strongly discourage) to 5(strongly encourage). Scores were summed

across all four scenarios to provide a total scorefor Peer Influence and a total score for MostInfluential Male. The internal consistency coef-ficient was � � .74 for Peer Influence and � �.86 for Most Influential Male, consistent withthe Cronbach’s alphas of .82 and .70 reportedby Kavussanu et al. (2002) on the comparablescales of Teammate Influence and Coach Influ-ence, respectively.

Social desirability. In order to determineif participants responded to questions in a so-cially desirable manner, the short version of theMarlow-Crowne Social Desirability Scale(Strahan & Gerbasi, 1972) was administered.The scale contains 10 items that describe so-cially desirable attributes (e.g., “I never resentbeing asked to return a favor”), and respondentsare asked to indicate if the statement is true orfalse, as it relates to them personally. A partic-ipant received a score of 1 on that item if heresponded in a socially desirable manner, and ascore of 0 was awarded if the response was notsocially desirable. The total score was calcu-lated by summing the numbers assigned to eachstatement.

Procedures

Research was conducted in compliance withInstitutional Review Board approval from thefirst author’s institution, and in compliance withthe review process of the school district. Afterreceiving this approval, school system adminis-trators, school principals, athletic directors, andcoaches were contacted. The first author wasinvited to attend team meeting where he distrib-uted parental consent forms to participants andexplained the rationale for the study. At a laterteam meeting, participants were provided theopportunity for voluntary participation if theyreturned with signed parental consent forms.Participants took approximately 15 to 20 min-utes to complete the survey packet. In additionto assurances of anonymity, participants wereinformed that all their data would be kept con-fidential and in a safe locked location. In orderto ensure voluntary participation, participantswere informed that they could write in theirplaybooks and turn in a blank survey packet atthe end if they did not want to participate in thestudy.

345BULLYING AMONG ADOLESCENT FOOTBALL PLAYERS

Data Analytic Plan

Structural Equation Modeling using AMOS(Arbuckle, 2008) was used to test the two hy-pothesized models of relationships betweenmoral atmosphere, male role norms and twooutcomes: moral functioning and bullying be-haviors. In each structural equation, model errorterms were correlated for each scenario used tomeasure moral atmosphere for influential maleand peers. The Comparative Fit Index (CFI) andthe Root Mean Square Error of Approximation(RMSEA) were used to assess model fit, as ChiSquare statistics are sensitive to both samplesize and high correlations and are thus lessindicative of good model fit (Kline, 2005). Val-ues greater than .95 on the CFI and less than .06on RMSEA are indicative of good model fit(Schreiber, Stage, King, Nora, & Barlow,2006). According to Kline (2005), the currentsample size of 206 is adequate for SEM, asvalidated by the results of the post hoc poweranalyses for RMSEA that we conducted. Usingpower analysis procedures suggested byPreacher and Coffman (2006), power estimatesfor the current study were greater than .80 forboth models. Furthermore, all variables wereassessed for normality. Four variables were logtransformed due to high levels of skew (val-ues � 2). These included two variables repre-

senting “relational cyberbullying” and “physi-cal threat” for the influential male, bullyingbehavior, and social desirability.

Results

Structural Equation Model 1: PredictingMoral Functioning

The first hypothesized model (see Figure 1)estimated the relations between moral atmo-sphere, male role norms, and moral functioningwhile controlling for age and social desirability.The first model achieved good model fit (CFI �.98; RMSEA � .04 [90% C.I. � .02 – .06]).Figure 1 presents standardized regression weightsand significance for each path. There were signif-icant paths between moral atmosphere and moralfunctioning (� � .92; p � .001) and moral atmo-sphere and meanings of adolescent masculinity(� � .27; p � .01). Those adolescents who per-ceived that their friends and the most influentialmale in their life had engaged in or would encour-age bullying were more likely to have judgedbullying as appropriate, disclosed their intention tobully, and engaged in bullying. Likewise, percep-tions of permissive attitudes about bullying fromfriends and influential males were also positivelyrelated to identification with male role norms.Finally, identification with traditional male role

Figure 1. Model of moral atmosphere, masculinity, and moral functioning among highschool football players. Note: � p � .05. �� p � .001.

346 STEINFELDT, VAUGHAN, LAFOLLETTE, AND STEINFELDT

norms was also significantly related to moral func-tioning (� � .15; p � .05). Thus, a higher en-dorsement of male role norms was also related tomore permissive attitudes about bullying. Thepath from age to moral functioning was not sig-nificant. Social desirability significantly predictedmoral functioning (� � �.13; p � .01) withgreater levels of social desirability being related toless permissive attitudes about bullying. Overall,the social norms of a moral atmosphere that con-dones bullying and a strong identification withmale role norms were positively related to moralfunctioning that reflects judgments, intentions,and behaviors that support bullying.

Structural Equation Model 2: PredictingBullying Behaviors

The second hypothesized model (see Figure 2)aimed to test the relations between the importanceof influential male and peer dimensions of moralatmosphere, masculinity, and bullying behaviorswhile controlling for age and social desirability.The aim of this second model was to understandthe roles that adults, friends, and male role normsplay in bullying behaviors among football playersin order to inform prevention efforts. This second

model also achieved good model fit (CFI � .97;RMSEA � .05 [90% C.I. � .03 – .07]). Standard-ized regression weights and significance for eachpath are presented in Figure 2. Of primary interestwere the paths from influential male and peers tobullying behaviors. The moral atmosphere normcultivated by participants’ perceptions about theattitudes of their most influential male figure werepositively and significantly related to bullying be-haviors (� � .51; p � .001). The more a playerperceived than his most influential male figurewas supportive of bullying, the more the playerreported having recently engaged in bullying be-haviors. Likewise, the moral atmosphere norm ofpeer influence was also significantly related tobullying behaviors, (� � .32; p � .05). The rela-tionship between the moral atmosphere of influ-ential males positively related to that of peers(� � .76; p � .001). In terms of meanings ofadolescent masculinity, the perceived moral atmo-sphere created by the most influential male figurewas also related to higher identification with moretraditional male role norms (� � .25; p � .01).Higher identification with male role norms was inturn related to more bullying behaviors (� � .15;p � .05). Taken together, results demonstrate that

Figure 2. Norms of moral atmosphere, masculinity, and bullying behaviors among highschool football players. Note: � p � .05. �� p � .001.

347BULLYING AMONG ADOLESCENT FOOTBALL PLAYERS

the perceived moral atmosphere created by influ-ential males had the most impact on bullyingbehaviors among adolescent football players.Again, age and social desirability were not signif-icantly related to bullying behaviors.

Discussion

Impact of Social Norms on MoralFunctioning Process of Bullying

The results of the study indicated that collec-tive norms that comprise the team’s moral at-mosphere—in addition to adherence to malerole norms—have an impact on how highschool football players perceive bullying. Theresults suggested that a football player wasmore likely to endorse the acceptability of bul-lying if he perceived a lower moral atmosphere(i.e., he reported that his friends were doing it,he perceived that bullying was acceptable by themost influential male in his life) and if he re-ported higher endorsement of male role norms.Past research has linked conformity to mascu-line role norms to a host of negative outcomes,including low self-esteem, depression, alcoholabuse, and sexually aggressive behavior (Locke& Mahalik, 2005; Mahalik et al., 2003; Oransky& Fisher, 2009). However, compared to thevolume of studies examining the experience ofadult males (see O’Neil, 2008), little researchhas been devoted to adolescent males’ experi-ences with gender role socialization. This lineof research could provide a better developmen-tal understanding of adolescents’ emerging ex-periences with masculinity. Also, no research todate has directly examined if endorsement ofmasculine norms within the unique context ofspecific sport cultures might be related to bul-lying or other issues of school violence. Thus,

this study addressed a variety of gaps in theliterature by reporting that meanings of adoles-cent masculinity, in combination with the normsof moral atmosphere, predicted the acceptabilityof bullying beliefs and behaviors among highschool football players.

Impact of Social Norms on BullyingBehaviors

The results of our study indicated that highschool football players did not report a highdegree of involvement with any of the fourspecific bullying behaviors (see Table 1). Themajority of players responded that they have neveror that they have rarely engaged in any of thebullying behaviors. This finding does not providesupport for societal perceptions of football playersas stereotypic bullies. Although we didn’t have acomparison group (e.g., nonathletes), this is aninteresting finding given that there is little con-sensus in the literature on what constitutes avalid profile of a typical bully (Aluede et al.,2008; Beaty & Alexeyev, 2008; Kreager, 2009;Vaillancourt et al., 2003). Some researchers docharacterize bullies as being members of thepopular group who gain strength by bullyingmembers of other groups (Duncan, 1999). Onthe other hand, “though some people may ste-reotype them [bullies] as being cocky, mali-cious, hyperaggressive individuals, most boybullies, in reality, are insecure, isolated, andexceedingly unhappy” (Pollack, 2000, p. 109).These conflicting theoretical positions, whencombined with the results of this study, suggestthat more research is needed in order to estab-lish a valid profile of a typical bully that canelucidate societal stereotypes of bullies.

In addition to reporting the frequencies ofbullying behaviors reported by high school

Table 1Responses and Frequencies of Bullying Behaviors of High School Football Players (N � 206)

Behavior Never Rarely Sometimes Often Always

Physical Bullying 143 34 20 3 669% 17% 10% 1% 3%

Relational Cyberbullying 179 13 6 5 387% 6% 3% 2% 2%

Verbal Bullying (Physical Threat) 153 31 14 5 374% 15% 7% 2% 2%

Social Bullying (Homophobic) 93 55 35 12 1145% 27% 17% 6% 5%

348 STEINFELDT, VAUGHAN, LAFOLLETTE, AND STEINFELDT

football players in this sample, the study’s re-sults provided support for our second hypothe-sis. Higher adherence to adolescent male rolenorms, along with perceptions of acceptabilityof bullying by peers and by the most influentialmale figure in the adolescent’s life, predictedhigher levels of bullying behaviors. Given thefocus on peer influence and peer pressure in thedevelopmental literature (Espelage & Holt,2001; Heilbron & Prinstein, 2008), one mightexpect friends to be the strongest predictor ofbullying. Although peer influence was signifi-cantly related to bullying behavior in our model,the perceived beliefs of the most influentialmale figure were the strongest predictor of bul-lying behaviors. This is an interesting findingwhich may speak to the structure and culture ofsport because authority figures wield great in-fluence and power over athletes within thisunique context (Sabo & Panepinto, 1990). Tothis point, fathers, brothers, and coaches werereported as the most influential male figures inthe lives of the high school football players inthis sample. These results suggest that fatherscan play a significant role in shaping their sons’interpersonal skills as it relates to decisionswhether or not to be a bully. Taken togetherwith research on the influence of coaches insport (Becker, 2009; Gould, Collins, Lauer, &Chung, 2006; Steinfeldt, Foltz, et al., 2011), theresults of this study suggest that coaches, fa-thers, brothers, and other influential male fig-ures can influence not only the development ofmale role norms, but these men can also posi-tively impact the bullying beliefs and behaviorsof adolescent football players.

Implications for Prevention Programming

The implications of the current study havedirect relevance to research and practice ele-ments of prevention, based on the best practiceguidelines for psychologists in the area of pre-vention practice, research, training, and socialadvocacy (Hage et al., 2007). Initial steps pre-vention research aims to identify those risk andprotective factors that are salient to a particularproblem (i.e., bullying). In the current study, wefound that adherence to masculine norms, peerinfluence, and the encouragement of bullyingbehaviors by influential men were directlylinked to engagement in bullying behaviors. Askey elements of the preintervention stage ofprevention research (Hage et al., 2007), thesefindings can inform the development of bully-ing prevention interventions for high schoolcommunities. Parental and adult monitoringhave been shown to be protective factors inreducing relational aggression in schools (Lead-beater, Banister, Ellis, & Yeung, 2008; Toturaet al., 2009). Efforts toward decreasing bullyingbehavior have been shown to be most effectivewhen directed toward family and school do-mains (Wild et al., 2004). The results of thisstudy suggest that influential males may need toplay a central role in the intervention, and thatdiscussions about adherence to traditional mas-culine norms should be included in these inter-ventions. Thus, it might be effective to trainfathers and coaches to cofacilitate an interven-tion component on masculine norms and howthese norms influence the behavior of their sonsand players.

In accordance with the prevention guidelines,psychologists in this setting would aim to “pre-vent human suffering through the developmentof proactive interventions” (Hage et al., 2007, p.501) that work toward eliminating bullying andschool violence. Thus, not only do the currentresults have implications for the development ofprevention interventions, they can also informthe current practice of psychologists who workwith adolescent student-athletes. One way thatthe current research informs this practice is byfocusing on bullying within a specific popula-tion in a specific context. In fact, preventionefforts that address bullying and school vio-lence have aimed to understand the contexts inwhich youth are developing with particular at-tention to interactions between parents, peers,

Table 2Means and Standard Deviations for Main StudyVariables (N � 206)

Variable Mean SD

Heterosexism 23.79 4.06Emotional Restriction 16.15 3.21Social Teasing 13.97 2.16Constant Effort 17.95 2.98Most Influential Male 5.90 2.88Teammate Influence 8.78 2.76Judgment 6.70 3.49Intention 7.03 3.30Behavior 6.18 2.83Social Desirability 4.74 2.78

349BULLYING AMONG ADOLESCENT FOOTBALL PLAYERS

and schools and deliver prevention within thesecontexts (Espelage & Horne, 2008). In the con-text of sport, psychologists are called upon toplay a big role in the psychosocial developmentof student-athletes (Steinfeldt, Foltz, et al.,2011). Thus, psychologists working with ado-lescent football players may want to considerbullying within the broader context, particularlythe ways that traditional masculine norms areconveyed by peers and influential males withinthe unique context of football.

Not dissimilar to the development of empir-ically based prevention, psychologists in currentpractice might not only enlist the help ofcoaches, fathers, and brothers in choosing anddelivering their interventions, psychologistsmay also choose to involve football playersthemselves in this process. Similar to the sig-nificant relationship between peer influence andbullying in our study, a robust body of literaturehas suggested that peers play a role in thisprocess (Duffy & Nesdale, 2009; Espelage &Holt, 2001; Heilbron & Prinstein, 2008; Ojala& Nesdale, 2004). Thus, football players can beencouraged to use their peer influence—enhanced by their potentially elevated socialstatus as center sport participants—to set andenforce the norm that bullying is unacceptablein their schools and communities. As such, inaddition to enlisting influential males in imple-menting bullying prevention programming,identifying the most effective peer leaders (e.g.,high school football players) may also be im-portant in prevention efforts.

Limitations

This study has a number of limitations tonote. First, the study design was cross-sectionaland self-report in nature. From a social learningand developmental perspective, it makes sensethat influential males and adherence to mascu-line norms would influence moral functioningin regards to bullying. However, the temporalordering of these constructs cannot be defini-tively determined because of the cross-sectionalnature of the study. Additionally, the experienceof this sample of football players may not berepresentative of the experience of other ado-lescent males from different regions of thecountry. The particular schools in this districtmay have unique policies, personnel, and ex-periences— both successful and challeng-

ing—that may not represent the experiencesof other school districts, particularly as itrelates to bullying and school violence. Thus,the results of this study may not be general-izable to other schools, to other groups ofadolescents, or to other groups of high schoolfootball players. Further research should con-tinue to explore the dynamics of bullyingwithin the unique context of sport in order totriangulate the results of this study with theresults from other sites.

An additional limitation of this study is thelow reliability reported in some subscales of theMAMS (i.e., Constant Effort, � � .69; SocialTeasing, � � .56). Although the reliability ofthe Social Teasing subscale was low, this inter-nal consistency coefficient is consistent with thelevel (� � .61) reported by Oransky and Fisher(2009) in their original psychometric validationof the scale. Social teasing and bullying may beintuitively connected, if bullying is conceptual-ized as social teasing that escalates when theteasing has been taken too far. However, be-cause of the low levels of reliability reported bythis particular subscale, the results of this studymay not sufficiently represent the relationshipbetween these constructs. Future researchshould continue to examine the psychometricproperties of the MAMS, in addition to exam-ining the relationship between social teasingand bullying. Results should be interpreted inlight of these limitations.

Conclusion

Football is a sport that teaches its playershow to be violent and aggressive on the field(Gage, 2008; Messner, 1990), but little re-search has addressed if players learn to com-partmentalize this instrumental aggressionand deem it to be appropriate only on thefield. Societal portrayals of football playersdepict them as perpetrators of bullying on awide scale, despite a lack of empirical re-search to justify these stereotypes. The resultsof our study suggest that bullying was pre-dicted by social norms— both the norms ofmoral atmosphere and endorsement of malerole norms. The strongest predictor of bully-ing was the perception of whether a player’smost influential male in his life would ap-prove of the behavior. In addition to preven-tion interventions that highlight the role of

350 STEINFELDT, VAUGHAN, LAFOLLETTE, AND STEINFELDT

addressing masculine norms and including in-fluential males in this process, implicationsfor intervention programming also suggestthat football players can be instrumental increating a school culture that does not toleratebullying. Because the center sport of footballoften provides high school football playerswith access to higher levels of the school’ssocial hierarchy (Gilbert & Gilbert, 1998;Kreager, 2007; Messner, 2002), interventionsthat utilize peer leadership models can accessfootball players’ relational capital and socialstatus to combat school violence and posi-tively impact the overall school environment.

References

Aluede, O., Adeleke, F., Omoike, D., & Afen-Akpaida, J. (2008). A review of the extent, nature,characteristics and effects of bullying behaviour inschools. Journal of Instructional Psychology, 35,151–158.

Arbuckle, J. L. (2008). AMOS 17.0 (Computer soft-ware). Crawfordville, FL: AMOS DevelopmentCorporation.

Beaty, L. A., & Alexeyev, E. B. (2008). The problemof school bullies: What the research tells us. Ado-lescence, 43, 1–11.

Becker, A. (2009). It’s not what they do, It’s howthey do it: Athlete experiences of great coaching.International Journal of Sports Science & Coach-ing, 4, 93–119. doi:10.1260/1747-9541.4.1.93

Chu, J., Porche, M., & Tolman, D. (2005). The ado-lescent masculinity ideology in relationship scale:Development and validation of a new measure forboys. Men and Masculinities, 8, 93–115. doi:10.1177/1097184X03257453

Coggan, C., Bennett, S., Hooper, R., & Dickinson, P.(2003). Association between bullying and mentalhealth status in New Zealand adolescents. TheInternational Journal of Mental Health Promo-tion, 5, 16–22.

Connell, R. W. (1996). Teaching the boys: Newresearch on masculinity, and gender strategies forschools. Teachers College Record, 98, 206–235.

Coulumb-Cabagno, G., & Rascle, O. (2006). Teamsports players’ observed aggression as a functionof gender, competitive level, and sport type. Jour-nal of Applied Sport Psychology, 36, 1980–2000.doi:10.1111/j.0021-9029.2006.00090.x

Curry, T. J. (2000). Booze and bar fights: A journeyto the dark side of college athletics. In J. McKay,M. A. Messner, & D. Sabo (Eds.), Masculinities,gender relations and sport (pp. 162–175). Thou-sand Oaks, CA: Sage.

Duffy, A. L., & Nesdale, D. (2009). Peer groups,social identity, and children’s bullying behavior.Social Development, 18, 121–139. doi:10.1111/j.1467-9507.2008.00484.x

Dukes, R. L., Stein, J. A., & Zane, J. I. (2009). Effectof relational bullying on attitudes, behavior andinjury among adolescent bullies, victims and bul-ly-victims. The Social Science Journal, 46, 671–688. doi:10.1016/j.soscij.2009.05.006

Duncan, R. (1999). Maltreatment by parents andpeers: The relationship between child abuse, bullyvictimization, and psychological distress. ChildMaltreatment, 19, 45–56. doi:10.1177/1077559599004001005

Elman, W. F., & McKelvie, S. J. (2003). Narcissismin football players: Stereotype or reality? AthleticInsight: The Online Journal of Sport Psychol-ogy, 5, 1–9.

Espelage, D. L., & Holt, M. K. (2001). Bullying andvictimization during early adolescence: Peer influ-ences and psychosocial correlates. Journal ofEmotional Abuse, 2, 123–142. doi:10.1300/J135v02n02_08

Espelage, D. L., & Horne, A. M. (2007). Schoolviolence and bullying prevention: From research-based explanations to empirically based solutions.In S. D. Brown & R. W. Lent (Eds.), Handbook ofcounseling psychology (4th ed., pp. 588–598).Hoboken, NJ: Wiley.

Foley, D. E. (2001). The great American footballritual: Reproducing race, class, and gender in-equality. In A. Yiannakis & M. J. Melnick (Eds.),Contemporary issues in sociology of sport. Cham-paign, IL: Human Kinetics.

Forbes, G., Adams-Curtis, L., Pakalka, A., & White,K. (2006). Dating aggression, sexual coercion, andaggression-supporting attitudes among collegemen as a function of participation in aggressivehigh school sports. Violence Against Women, 12,441–455. doi:10.1177/1077801206288126

Gage, E. A. (2008). Gender attitudes and sexualbehavior. Violence Against Women, 14, 1014–1032. doi:10.1177/1077801208321987

Gilbert, R., & Gilbert, P. (1998). Masculinity goes toschool. New York, NY: Routledge.

Gini, G. (2006). Social cognition and moral cognitionin bullying: What’s wrong? Aggressive Behav-ior, 32, 528–539. doi:10.1002/ab.20153

Gini, G. (2008). Italian elementary and middle schoolstudents’ blaming the victim of bullying and per-ception of school moral atmosphere. The Elemen-tary School Journal, 108, 335–354. doi:10.1086/528975

Gould, D., Collins, K., Lauer, L., & Chung, Y.(2007). Coaching life skills through football: Astudy of award winning high school footballcoaches. Journal of Applied Sport Psychology, 19,16–37. doi:10.1080/10413200601113786

351BULLYING AMONG ADOLESCENT FOOTBALL PLAYERS

Hage, S. M., Romano, J. L., Conye, R. K., Kenny, M.,Matthews, C., Schwartz, J. P., & Waldo, M. (2007).Best practice guidelines on prevention practice, re-search, training, and social advocacy for psycholo-gists. The Counseling Psychologist, 35, 493–566.doi:10.1177/0011000006291411

Heilbron, N., & Prinstein, M. J. (2008). A review andreconceptualization of social aggression: Adaptiveand maladaptive correlates. Clinical Child andFamily Psychology Review, 11, 176–217. doi:10.1007/s10567-008-0037-9

Jankauskiene, R., Kardelis, K., Sukys, S., & Karde-liene, L. (2008). Associations between school bul-lying and psychosocial factors. Social Behaviorand Personality, 36, 145–162. doi:10.2224/sbp.2008.36.2.145

Kavussanu, M., Roberts, G. C., & Ntoumanis, N.(2002). Contextual influences on moral function-ing of college basketball players. The Sport Psy-chologist, 16, 347–367.

Kavussanu, M., Stamp, R., Slade, G., & Ring, C.(2009). Observed prosocial and antisocial behav-iors in male and female soccer players. Journal ofApplied Sport Psychology, 21, S62–S76. doi:10.1080/10413200802624292

Kim, Y. S., & Leventhal, B. (2008). Bullying andsuicide. A review. International Journal of Ado-lescent Medicine and Health, 20, 133–154. doi:10.1515/IJAMH.2008.20.2.133

Kline, R. B. (2005). Principles and practice of struc-tural equation modeling (2nd ed.). New York, NY:Guilford Press.

Kowalski, R. M., & Limber, S. P. (2007). Electronicbullying among middle school students. Journal ofAdolescent Health, 41, S22–S30. doi:10.1016/j.jadohealth.2007.08.017

Kreager, D. A. (2007). Unnecessary roughness:School sports, peer networks, and male adolescentviolence. American Sociological Review, 72, 705–724, doi:10.1177/000312240707200503

Kreager, D. A. (2009). Sports machismo may be acue to male teen violence. Nutrition Health Re-view: The Consumer’s Medical Journal, 100, 6.

Leadbeater, B. J., Banister, E. M., Ellis, W. E., &Yeung, R. (2008). Victimization and relational ag-gression in adolescent romantic relationships: Theinfluence of parental and peer behaviors, and indi-vidual adjustment. Journal of Youth and Adoles-cence, 37, 359 –372. doi:10.1007/s10964-007-9269-0

Locke, B. D., & Mahalik, J. R. (2005). Examiningmasculinity norms, problem drinking, and athleticinvolvement as predictors of sexual aggression incollege men. Psychology of Men & Masculinity, 5,279–283.

Mahalik, J. R., Locke, B., Ludlow, L., Diemer, M.,Scott, R. P. J., Gottfried, M., & Freitas, G. (2003).Development of the Conformity to Masculine

Norms Inventory. Psychology of Men & Masculin-ity, 4, 3–25. doi:10.1037/1524-9220.4.1.3

McCabe, R. E., Antony, M. M., Summerfeldt, L. J.,Liss, A., & Swinson, R. P. (2003). Preliminaryexamination of the relationship between anxietydisorders in adults and self- reported history ofteasing or bullying experiences. Cognitive Behav-iour Therapy, 32, 187–193. doi:10.1080/16506070310005051

Messner, M. A. (1990). When bodies are weapons:Masculinity and violence in sport. InternationalReview for Sociology of Sport, 25, 203–220. doi:10.1177/101269029002500303

Messner, M. A. (1996). Masculinities and athleticcareers. In E. N. Chow, D. Wilkinson, & M. BacaZinn (Eds.), Race, class and gender: Commonbonds, different voices (pp. 70–-86). ThousandOaks, CA: Sage.

Messner, M. A. (2002). Taking the field: Men, womenand sports. Minneapolis, MN: University of Min-nesota Press.

Ojala, K., & Nesdale, D. (2004). Bullying and socialidentity: The effects of group norms and distinc-tiveness threat on attitudes towards bullying. Brit-ish Journal of Developmental Psychology, 22, 19–35. doi:10.1348/026151004772901096

O’Neil, J. M. (2008). Summarizing 25 years of re-search on men’s gender role conflict using theGender Role Conflict Scale: New research para-digms and clinical implications. The CounselingPsychologist, 36, 358 – 445. doi:10.1177/0011000008317057

Oransky, M., & Fisher, C. (2009). The developmentand validation of the meanings of adolescent mas-culinity scale. Psychology of Men & Masculin-ity, 10, 57–72. doi:10.1037/a0013612

Pollack, W. S. (2000). Bullying and teasing: Fittingin and being left out. In W. S. Pollack & T. Shuster(Eds.), Real boys voices (pp. 106–141). NewYork, NY: Penguin Press.

Poteat, V. P., & Rivers, I. (2010). The use of ho-mophobic language across bullying roles duringadolescence. Journal of Applied DevelopmentalPsychology, 31, 166–172. doi:10.1016/j.appdev.2009.11.005

Preacher, K. J., & Coffman, D. L. (2006, May).Computing power and minimum sample size forRMSEA [Computer software]. Available fromhttp://quantpsy.org/

Rest, J. R. (1983). Morality. In J. Flavell & E. Mark-man (Eds.), Manual of child psychology, P. Mus-sen (Gen. Ed.), Vol. 3: Cognitive development (pp.356–629). New York, NY: Wiley.

Rest, J. R. (1984). The major components of moral-ity. In W. Kurtines & J. Gewirtz (Eds.), Morality,moral behavior, and moral development (pp. 356–429). New York, NY: Wiley.

352 STEINFELDT, VAUGHAN, LAFOLLETTE, AND STEINFELDT

Sabo, D. F., & Panepinto, J. (1990). Football ritualand the social reproduction of masculinity. InM. A. Messner & D. F. Sabo (Eds.), Sport, men,and the gender order (pp. 115–126). Champaign,IL: Human Kinetics.

Sage, L., Kavussanu, M., & Duda, J. (2006). Goalorientations and moral identity as predictors of proso-cial and antisocial functioning in male associationfootball players. Journal of Sports Sciences, 24, 455–466. doi:10.1080/02640410500244531

Schreiber, J. B., Stage, F. K., King, J., Nora, A., &Barlow, E. A. (2006). Reporting structural equationmodeling and confirmatory factor analysis results: Areview. Journal of Educational Research, 99, 323–337. doi:10.3200/JOER.99.6.323-338

Shields, D. L., & Bredemeier, B. L. (2007). Ad-vances in sport morality research. In G. Tenen-baum & R. Eklund (Eds.), Handbook of sportpsychology (3rd ed., pp. 662–684). Hoboken, NJ:Wiley

Shields, D. L., Bredemeier, B. L., Gardner, D. E., &Bostrom, A. (1995). Leadership, cohesion, andteam norms regarding cheating and aggression.Sociology of Sport Journal, 12, 324–336.

Shields, D. L., LaVoi, N. M., Bredemeier, B. J., &Power, F. C. (2007). Predictors of poor sportsper-sonship in youth sports: Personal attitudes andsocial influences. Journal of Sport & Exercise Psy-chology, 29, 747–762.

South, C. R., & Wood, J. (2006). Bullying in prisons:The importance of perceived social status, pris-onization, and moral disengagement. AggressiveBehavior, 32, 490–501. doi:10.1002/ab.20149

Steinfeldt, J. A., Foltz, B. D., Mungro, J., Speight,Q. L., Wong, Y. J., & Blumberg, J. (2011). Mas-culinity socialization in sports: Influence of collegefootball coaches. Psychology of Men & Masculin-ity, 12, 247–259. doi:10.1037/a0020170

Steinfeldt, J. A., Rutkowski, L., Vaughan, E. L., &Steinfeldt, M. C. (2011). Masculinity, moral atmo-sphere, and moral functioning of high school foot-ball players. Journal of Sport and Exercise Psy-chology, 33, 215–234.

Stephens, D. E., & Kavanagh, B. (2003). Aggressionin Canadian youth ice hockey: The role of moralatmosphere. International Sports Journal, 7, 109–119.

Storch, E. A., Bagner, D. M., Geffken, G. R., &Baumeister, A. L. (2004). Association betweenovert and relational aggression and psychosocial

adjustment in undergraduate college students. Vi-olence and Victims, 19, 689–700. doi:10.1891/vivi.19.6.689.66342

Strahan, R., & Gerbasi, K. C. (1972). Short, homo-geneous versions of the Marlow-Crowne Social De-sirability Scale. Journal of Clinical Psychology, 28,191–193. doi:10.1002/1097-4679(197204)28:2�191::AID-JCLP2270280220�3.0.CO;2-G

Stuart, M. E., & Ebbeck, V. (1995). The influence ofperceived social approval and moral developmentin youth sport. Pediatric Exercise Journal, 7, 270–280.

Thompson, E. H., & Pleck, J. H. (1986). The structureof male norms. American Behavioral Scientist, 29,531–543. doi:10.1177/000276486029005003

Totura, C. M. W., MacKinnon-Lewis, C., Gesten,E. L., Gadd, R., Divine, K. P., Dunham, S., &Kamboukos, D. (2009). Bullying and victimizationamong boys and girls in middle school: The influ-ence of perceived family and school contexts. TheJournal of Early Adolescence, 29, 571–609. doi:10.1177/0272431608324190

Underwood, M. K., Beron, K. J., & Rosen, L. H.(2009). Continuity and change in social and phys-ical aggression from middle childhood throughearly adolescence. Aggressive Behavior, 35, 357–375. doi:10.1002/ab.20313

Vaillancourt, T., Hymel, S., & McDougall, P. (2003).Bullying is power: Implications for school-basedintervention strategies. Journal of Applied SchoolPsychology, 19, 157–176. doi:10.1300/J008v19n02_10

Visek, A., & Watson, J. (2005). Ice hockey players’legitimacy of aggression and professionalization ofattitudes. The Sport Psychologist, 19, 178–192.

Wang, J., Iannotti, R. J., & Nansel, T. R. (2009).School bullying among adolescents in the UnitedStates: Physical, verbal, relational, and cyber.Journal of Adolescent Health, 45, 368–375. doi:10.1016/j.jadohealth.2009.03.021

Wild, L. G., Flisher, A. J., Bhana, A., & Lombard, C.(2004). Associations among adolescent risk behav-iours and self-esteem in six domains. Journal ofChild Psychology and Psychiatry, 45, 1454–1467.doi:10.1111/j.1469-7610.2004.00330.x

Received June 30, 2011Revision received November 6, 2011

Accepted November 9, 2011 �

353BULLYING AMONG ADOLESCENT FOOTBALL PLAYERS