Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Jul 29.
Published in final edited form as: J Early Adolesc. 2014 May 1;34(4):463–485. doi: 10.1177/0272431613495446

Peers’ Perceptions of Gender Nonconformity: Associations with Overt and Relational Peer Victimization and Aggression in Early Adolescence

Russell B Toomey a,✉, Noel A Card b, Deborah M Casper c
PMCID: PMC4518457  NIHMSID: NIHMS708809  PMID: 26236066

Abstract

The current study used reports from 318 early adolescents to examine the associations of peer-reported gender nonconformity with peer- and self-reported overt and relational victimization and aggression and possible sex differences in these associations. Multiple-group structural equation modeling revealed that higher levels of peer-reported gender nonconformity were associated with higher self- and peer-reports of overt and relational victimization and aggression among males and females. The association between peer-reported gender nonconformity and peer-reported overt aggression was moderated by participant sex, such that the association was stronger for females compared to males. Results suggest that perceived gender nonconformity is associated with problematic peer relations, especially among females, in early adolescence and implications of these associations are discussed.

Keywords: Gender nonconformity, peer victimization, peer aggression

Societal gender norms restrict individuals to gender identities and expressions that conform to the male–female binary (Oswald, Blume, & Marks, 2005). For example, boys are socialized to enjoy masculine activities (e.g., playing sports) and dress and behave in masculine ways. By middle childhood, these gendered norms are apparent in the everyday choices of young people including their interests, choice of activities, and clothing choices (e.g., Collaer & Hines, 1995). These gender norms are negotiated and reinforced in heteronormative school climates that reflect broader societal expectations and pressures about what constitutes “normal” gender and sexuality (Chesir-Teran, 2003; Jackson, 2006). Thus, older children and adolescents who do not conform to these expectations may be rejected by their peers (e.g., Smith & Leaper, 2006) or labeled as “deviant” or “weird” by their classmates.

Developmental research has documented that the level of importance of conformity to gender norms follows a nonlinear pattern that is initially high in kindergarten, followed by a decrease in middle childhood and another increase in early adolescence (e.g., Alfieri, 1996; Hill & Lynch, 1983; Stoddart & Turiel, 1985). At the same time, early adolescence is a time when peer relations become increasingly important (e.g., Brown, 1990; Cairns, Xie, & Leung, 1998). Thus, early adolescence marks a critical time to understand the peer relationships of adolescents who are considered gender nonconforming by their classmates. The focus of this study is on gender nonconformity, which can be defined as the degree to which an individual is perceived to not adhere to societal gender expectations or norms (e.g., boys who are less masculine than other boys or act “too feminine”; e.g., Gay, Lesbian, and Straight Education Network [GLSEN] & Harris Interactive, 2012). Given the heightened attention to gender nonconformity and the importance placed on peer relationships during early adolescence, the purpose of this study is to examine the associations among gender nonconformity, peer victimization, and peer aggression in a school-based sample of early adolescent middle school students.

Gender Nonconformity and Peer Relationships

Peer victimization

Research on the peer relationships of gender nonconforming youth has largely focused on peer victimization experiences of those who do not conform to gendered norms. Of these studies, most have found that higher levels of gender nonconformity are associated with greater peer victimization among adolescents (e.g., Aspenlieder, Buchanan, McDougall, & Sippola, 2009; D’Augelli, Grossman & Starks, 2006). In contrast to these studies, a recent study (Wallien, Veenstra, Kreukels, & Cohen-Kettenis, 2010) of gender dysphonic children (i.e., children with a Gender Identity Disorder [GID] diagnosis or who were at a subthreshold for GID; mean age = 10.47 years) found no difference in levels of peer victimization between the gender referred children in their sample and a control sample of non-referred youth. Nonetheless, it is important to note that the children in this sample were not yet in adolescence, when an increase in adherence to gender roles is expected (e.g., Hill & Lynch, 1983; Stoddart & Turiel, 1985). Further, as suggested by Wallien and colleagues, given that the gender referred children in their study had likely been immersed in the same peer group during their elementary education years, it is possible that any negative peer attitudes related to gender variance may have become less salient over time. Nonetheless, with the exception of this study, research has largely come to the conclusion that early adolescents typically negatively evaluate gender nonconformity.

Studies that have examined the association between gender nonconformity and peer victimization in adolescence have generally used homogeneous samples of either presumed heterosexual or sexual minority youth (e.g., gay, lesbian, bisexual), a methodological decision which may ultimately conflate victimization based on sexual orientation and victimization based on gender nonconformity (e.g., D’Augelli et al., 2006). The generalizability of results from studies with homogeneous samples of sexual minority youth to the broader population of heterosexual youth is not known. Importantly, a study of adolescent appraisals of peers using vignettes that differed in terms of the sexual orientation and gender expression (e.g., more or less gender conforming) of the target did find that gender nonconforming targets (regardless of sexual orientation) were viewed more negatively than targets that were more gender conforming (Horn, 2007). This finding suggests that nonconformity to gender norms is potentially a more important social determinant of peer acceptance than sexual orientation. Regardless of sexual orientation, experiencing peer victimization is associated with a myriad of negative psychosocial and academic adjustment problems (e.g., Card, Isaacs, & Hodges, 2007); thus, an understanding of whether gender nonconformity and peer victimization are associated is important for future prevention and intervention efforts aimed at strategically targeting antecedents of peer victimization processes.

In addition to these cross-sectional studies of the association between peer victimization and gender nonconformity, two recent studies (Ewing Lee & Troop-Gordon, 2011a, 2011b) found that peer victimization was differentially associated with changes in gender-typical behaviors by gender in middle childhood (mean age = 10.8 years). One of these studies by Ewing Lee and Troop-Gordon (2011a) documented that the association between gender nonconformity and peer victimization may be bidirectional. In their study, peer victimization was associated with decreases in gender-nonconformity among males (i.e., decreases in feminine behaviors), but not among females. The other study by Ewing Lee and Troop-Gordon (2011b) examined these same associations, but differentiated between overt, verbal, and social exclusion forms of victimization. Consistent with their other study (Ewing Lee & Troop-Gordon, 2011a), findings indicated that overt victimization was associated with decreases in feminine behaviors among males, yet social exclusion was associated with increases in feminine behaviors among males. Further, among females, social exclusion was associated with decreases in masculine behaviors, while overt forms of victimization were associated were associated with decreases in both masculine and feminine behaviors. Nonetheless, while these two studies by Ewing Lee and Troop-Gordon suggested that there may be a prospective association between victimization and less frequent involvement in gender-atypical behaviors among males, particularly in the case of overt victimization, these studies were limited in that they did not assess the cross-lagged paths from masculine and feminine behaviors to later assessments of peer victimization. Further, other studies have identified a prospective association between engagement in gender atypical behavior (aggressive girls and withdrawn boys) and higher levels of peer difficulties (i.e., peer victimization; Kochel, Miller, Updegraff, Ladd, & Kochenderfer-Ladd, 2012). Thus, additional research is needed to better understand the associations between gender nonconformity and poor peer relations.

Peer aggression

Less is known about the association between gender nonconformity and the enactment of peer aggression. To our knowledge, only one study has documented a positive association between gender nonconformity and externalizing problems (Carver, Yunger, & Perry, 2003). Contrastingly, and similar to the findings with peer victimization, the study on the peer experiences of gender dysphoric children by Wallien and colleagues (2010) did not find a significant mean-level difference in levels of enacted bullying between the gender referred youth compared to the control sample (non-referred youth). Yet, the characteristic of gender nonconformity may be linked to peer aggression for several reasons. First, gender nonconforming early adolescents may be more likely to experience peer victimization (e.g., Aspenlieder et al., 2009) or peer rejection, and thus may respond to these experiences with aggression. For instance, Wyss (2004) found that a common reaction to peer victimization based on gender nonconformity was to retaliate using verbal and physical aggression. Second, involvement in peer aggression, depending on the form used, may be an indicator of gender nonconformity to adolescents (e.g., Crick, 1997). That is, youth who engage in gender non-normative behaviors (e.g., girls who engage in overt aggression or boys who engage in withdrawal behaviors; Liben & Bigler, 2002) may be labeled by their peers as gender nonconforming and are at risk for peer relationship problems (e.g., Kochel et al., 2012). Given that boys are more likely than girls to engage in overt forms of aggression (Card et al., 2008; Hyde, 1984; Maccoby & Jacklin, 1974), girls who engage in this type of behavior may be perceived as gender nonconforming (Crick, 1997). Although relational aggression was originally conceptualized as a female form of aggression, a recent meta-analysis documented that there is no meaningful sex difference in use of relational aggression (Archer, 2004; Card, Stucky, Sawalani, & Little, 2008); thus, boys who engage in this behavior may not necessarily be labeled as gender nonconforming by their peers. Similar to studies on peer victimization, peer aggression has also been found to be associated with poor adjustment (e.g., Card et al., 2008). Thus, given conceptualizations and empirical findings about the gendered nature of aggression (i.e., Card et al., 2008; Crick, 1997), and the contrasting findings in prior research, additional research is needed that examines the association between gender nonconformity and peer aggression. That is, questions still remain about whether gender nonconformity among early adolescents is associated with normative or non-normative forms of aggressive behavior.

Forms and Reports of Victimization and Aggression

Multiple forms

The utility of considering multiple forms (i.e., overt and relational) of peer victimization and aggression is clear for understanding child and adolescent psychosocial adjustment (e.g., Card et al., 2008; Crick & Bigbee, 1998). Yet, with the exception of one study by Ewing Lee and Troop-Gordon (2011b), previous literature that has examined the associations of gender nonconformity with peer victimization and aggression lacks attention to the multiple forms of victimization and aggression, instead primarily focusing on victimization (defined as a single construct) or externalizing problems (broadly defined). Examinations of the specific forms of victimization and aggression may provide a clearer picture of the peer relationships of gender nonconforming adolescents, thus allowing for more accurate and informed intervention and prevention efforts. For instance, Ewing Lee and Troop-Gordon (2011b) illuminated interesting differences in the longitudinal associations between overt versus relational forms of victimization and changes in gender typical behaviors. Further, examinations of the forms of aggression allow one to formally test whether gender non-normative forms of aggression are associated with peer perceptions of gender nonconformity, as suggested by Crick (1997). Relative to gender nonconformity, Horn (2005) suggests that gender nonconforming adolescents may be more likely to receive relational forms of victimization, rather than more overt forms.

Multiple reporters

In addition to the consideration of the multiple forms of peer victimization and aggression, multi-informant approaches undoubtedly provide a more comprehensive picture of peer relationships (Juvonen, Nishina, & Graham, 2001). Research has typically examined the association between gender nonconformity and peer victimization with self-report measures or teacher reports. Yet, peers’ perceptions of nonconformity to gender norms might be more meaningfully related to peer relations; for instance, victimization of gender nonconforming adolescents likely occurs because of the perpetrator’s perceptions of the victim’s gender norm violations (Oswald et al., 2005).

To date, only one study has used multiple reporters to examine the association between early adolescent gender nonconformity and peer victimization. In that study, Aspenlieder and colleagues (2009) found important differences in the associations among gender nonconformity and victimization based on the reporter (self versus peer) and target gender among early adolescents (school grades 5 to 9). Specifically, Aspenlieder and colleagues found that both self- and peer-reports of gender nonconformity were positively associated with self- and peer-reports of victimization. In their study, there was a significant positive association between gender nonconformity and self-reported levels of physical and relational victimization for boys; yet, this association was not present for girls. However, only their self-report measure of victimization separated overt and relational forms; therefore, they were not able to assess whether gender nonconformity was differentially associated with relational and overt victimization as perceived by peers. Further, their definition of gender nonconformity was limited to behavioral characteristics (e.g., willing to take risks as masculine) and therefore did not capture other facets (e.g., appearance, engagement in activities), which may be more visible (and salient) indicators of gender nonconformity to peers. That is, while two male adolescents may have similar scores on measures that assess masculine and feminine traits, one of those adolescents may outwardly engage in cross-gender behaviors that are observable by peers whereas the other may not engage in these types of activities.

Likewise, few studies have examined the association between peer aggression and gender nonconformity using multiple reporters. One study that has examined this association did not find a significant association between self-reported gender typicality and peer-reported externalizing problems (Yunger, Carver, & Perry, 2004). In a separate study, girls who self-reported greater dissatisfaction with their gender were perceived by their peers as exhibiting higher rates of externalizing behavior; however, this association was not present for boys (Carver et al., 2003). These two studies defined and measured externalizing problems more broadly, rather than aggression in particular. An additional study did find that early adolescent girls who engaged in aggression (as reported by their peers) had higher levels of peer-reported victimization; however, this study collapsed verbal, physical, and relational aggression into a singular construct (Kochel et al., 2012). Yet, as mentioned earlier, research has shown the importance of examining the multiple forms of aggression instead of examining aggression as a singular construct.

Current Study Goals

This study expanded on the current literature by using a multi-informant approach to examine the associations of (a) peer-reported gender nonconformity with (b) self- and peer- reports of (c) overt and relational forms of (d) peer victimization (experienced) and aggression (enacted). Additionally, given that previous research had identified important differences in the experiences of gender nonconforming males and females (e.g., Carver et al., 2003; D’Augelli et al., 2006, Toomey, McGuire, & Russell, 2012), we examined moderation of these associations by sex.

Based on previous literature (e.g., Aspenlieder et al., 2009), we expected that adolescents who were nominated by a larger number of peers as gender nonconforming would self-report higher levels of overt and relational victimization, and that there would be positive associations between gender nonconformity and peer-reported overt and relational forms of victimization. Further, given that previous research has documented an association between gender nonconformity and externalizing problems (Carver et al., 2003) and more overt forms of retaliation (e.g., Wyss, 2004), we expected to find a positive association between peer-reported gender nonconformity and self- and peer-reports of overt aggression, but not relational aggression. Finally, given that prior research had found that gender nonconformity has more significant consequences for male adolescents (e.g., Carver et al., 2003), we expected that the associations between gender nonconformity and victimization would be stronger for male participants compared to females.

Method

Sample

Participants included 318 middle school students in sixth (n=140) and seventh grades (n=178). The students in seventh grade were divided into three teams for educational purposes and the majority of peer interactions occurred within these teams: 77 were in Team A, 68 were in Team B, and 33 were in Team C. The mean age of the sample was 12.3 years (SD = .70) and approximately half (56.3%) were female. Participants represented a wide range of racial and ethnic backgrounds: White (54.7%), Hispanic or Latino/Latina (22.6%), Black or African American (2.8%), Asian (1.9%), Native Hawaiian or Pacific Islander (1.0%), American Indian or Alaska Native (< 1.0%), and 16.7% multiple or other.

Study Procedures

Students were sampled from a middle school located in the Southwestern United States. Students were eligible to participate in the study after they returned a signed parental consent form and provided their own informed assent. The participation rate for the study was 70%. Students completed surveys during two 45-minute class periods in the spring of 2009. Trained examiners read instructions aloud and provided individual assistance to students, as necessary. The study protocol was approved by the university’s institutional review board.

Measures

Self-report inventory

Self-reports of peer victimization and aggressive behavior were obtained from an adapted version of an established instrument (Little, Jones, Henrich, & Hawley, 2003). Both forms of victimization and aggression were measured with three items each: overt victimization (e.g., kids hit or kick me; α = .79), relational victimization (e.g., kids spread rumors about me; α = .75), overt aggression (e.g., I push others around; α = .82), and relational aggression (e.g., I try to keep others from being part of activities or in my group; α = .69). Each item was assessed on a 5-point scale (0=never to 4=always). Latent variables were created for each self-reported form of aggression and victimization using the three indicators for each construct, respectively, for use in structural equation modeling (SEM).

Peer nomination inventory

To obtain peer nominations of overt and relational forms of victimization and aggression, students marked the names of the peers in their grade (6th grade) or team (7th grade) who they perceived to meet the description for each item. Students were allowed to nominate an unlimited number of peers. Three items measured each form of aggression and victimization: overt victimization (e.g., these kids get hit or kicked by others; α = .90), relational victimization (e.g., others spread rumors about these kids; α = .88), overt aggression (e.g., these kids hit or kick others; α = .90), and relational aggression (e.g., these kids spread rumors about others; α = .89). Scores were computed as the percentage of peers who nominated a participant for each item, ranging from 0 to 100 percent. This percentage was then standardized by grade (6th grade participants) or team (7th grade participants). Latent variables were created for each form of peer-reported aggression and victimization using the three indicators for each construct, respectively, for use in SEM.

One item that assessed gender nonconformity was included on the inventory: “These boys act like girls / these girls act like boys.” This item is similar to others that assess gender nonconformity (e.g., GLSEN & Harris Interactive, 2012). Standardized percentages of nominations were calculated for each participant. Higher scores indicate more perceived gender nonconformity by peers. Given that only one item assessed gender nonconformity, a latent variable was not created for use in SEM, but rather the measured variable was utilized.

Results

Analytic Procedures

We examined associations of peer-reported gender nonconformity with self- and peer-reports of overt and relational victimization and aggression using structural equation modeling (SEM). We began with multi-group mean and covariance structures (MACS) models to evaluate whether our constructs had equivalent (i.e., invariant) structure and meaning across participant gender (Little, Card, Slegers, & Ledford, 2007). To test for measurement invariance, we fit a series of constrained models that examine configural, loading, and intercept equivalence (Little et al., 2007). After measurement equivalence was established, MACS modeling was used to examine the model for mean level differences in latent constructs and moderation of study variances and covariances by participant sex. We used Cohen’s d effect sizes (Cohen, 1988) to estimate effect sizes for all mean-level differences that emerged in the model. According to Cohen (1988), an effect size of d = .2 is considered small, d= .5 is considered medium, and d = .8 is considered large. Finally, to test whether there were meaningful sex differences in the latent associations between gender nonconformity and peer victimization and aggression, a series of sequential nested models were tested, such that each path of interest was sequentially constrained to be equal across groups.

For nested model comparisons, we examined the difference in chi-square (χ2) test, such that a significant difference in chi-square values indicates that the constraint led to a decrease in model fit and thereby is not tenable (i.e., moderation exists by participant sex). Because of the large number of comparisons made, a more conservative alpha value was observed (α = .01). To assess the adequacy of our models we used standard measures of practical fit: root mean square error of approximation (RMSEA) and the comparative fit index (CFI). Model fit is generally deemed good (or acceptable) when the RMSEA is less than .05 (.08) and the CFI is greater than .95 (.90) (Kline, 2011).

Gender differences

Table 1 displays the means, standard deviations, and correlations among all study constructs by participant sex. Consistent with prior research, high correlations were found between forms of victimization and forms of aggression within reporters (i.e., self-reports and peer-reports). Preliminary analyses revealed that the different forms of victimization and aggression could be distinguished (i.e., the correlation could not be constrained to 1.0): self-reported aggression (Δχ2 (Δdf = 1) = 61.806, p < .001), self-reported victimization (Δχ2 (Δdf = 1) = 13.829, p < .001), peer-reported aggression (Δχ2 (Δdf = 1) = 17.317, p < .001), and peer-reported victimization (Δχ2 (Δdf = 1) = 40.856, p < .001). These results suggest that, even though the bivariate correlations are high, there are meaningful distinctions between the different forms (i.e., overt and relational) for both reporters (self and peer). Measurement invariance across participant sex was established (see Table 2; Little et al., 2007), and the constrained factor loadings from these tests are presented in Table 3. Given the finding of measurement equivalence, we proceeded to explore whether the latent means, variances, and correlations differed by participant sex. Table 4 displays the unconstrained estimates for males and females (columns 2 and 3); the chi-square statistic for the constrained model with its associated degrees of freedom (columns 4 and 5); and the difference in chi-square statistic for the nested model comparison with the constraint for each path or mean across groups, its associated degrees of freedom change, and the significance level associated with the change statistic (columns 6-8).1

Table 1.

Correlations, Means, and Standard Errors of Key Study Constructs by Gender and Reporter

1 2 3 4 5 6 7 8 9 Mean (se)
Peer-Report 1. Gender Atypicality --- .28*** .33*** .22** .27*** .07 .14 .13 .16 0.02 (.09)
2. Overt Aggression .65*** --- .97*** .10 .05 .19* .14 .08 -.03 0.06 (.09)
3. Relational Aggression .54*** .99*** --- .08 .02 .17 .10 .08 -.06 -0.15 (.09)
4. Overt Victimization .37*** .37*** .35*** --- .97*** .00 .01 .22* -.62*** 0.12 (.09)
5. Relational Victimization .54*** .77*** .80*** .64*** --- -.10 .01 .16 .27** 0.03 (.09)
Self-Report 6. Overt Aggression .26*** .22** .18* .10 .13 --- .46*** .40*** .25** 0.53 (.10)
7. Relational Aggression .21** .35*** .39*** .06 .31 .61*** --- .38*** .69*** 0.60 (.12)
8. Overt Victimization .09 .30*** .31*** .28** .27** .28** .35*** --- .84*** 1.20 (.13)
9. Relational Victimization .19* .42*** .47*** .26** .38*** .32*** .52*** .99*** --- 0.98 (.12)
Mean (se) -0.41 (.08) -0.04 (.08) 0.13 (.08) -0.13 (.09) -0.03 (.09) 0.49 (.09) 0.39 (.09) 1.31 (.13) 1.27 (.12)
Open in a new tab

Note. Correlations for females are shown below the diagonal and correlations for males are shown above the diagonal. Means and standard errors for females are displayed on the bottom of the table, whereas means and standard errors for males are displayed on the left-hand side of the table.

***

p<.001,

**

p<.01,

*

p<.05.

Table 2.

Results of Factorial Invariance Tests

Model χ2 df p RMSEA NNFI CFI Constraint Tenable a
Configural Invariance 1149.70 480 < .001 .089 ∓ .007 .907 .925 ---
Weak Invariance 1218.93 496 < .001 .092 ∓ .007 .902 .919 Yes
Strong Invariance 1313.74 512 < .001 .0964 ∓ .007 .895 .911 Yes
Open in a new tab
a

Evaluated using criterion of decrease in CFI <.01 suggested by Cheung and Rensvold (2002).

Table 3.

Constrained Estimates of Unstandardized and Standardized Factor Loadings for Latent Variables Across Participant Sex

Factor Loadings
Constructs Unstandardized (Standard Error) Standardized
PR Gender Nonconformity 1.00 (0.05) 1.00/1.00
SR Overt Aggression
I hit or kick others 0.56 (0.06) 0.58/0.56
I push others around 0.57 (0.06) 0.57/0.57
I say mean things to others or call them names 1.00 (0.05) 0.99/0.99
SR Relational Aggression
I spread rumors about others 0.82 (0.06) 0.60/0.73
I try to keep others from being part of activities or in my group 0.72 (0.07) 0.57/0.65
I try to make my friends ignore others or stop talking to others 0.71 (0.07) 0.52/0.64
SR Overt Victimization
Kids hit or kick me 0.54 (0.06) 0.69/0.57
Kids push me around 0.69 (0.06) 0.82/0.73
Kids say mean things to me or call me names 0.77 (0.06) 0.87/0.84
SR Relational Victimization
Kids spread rumors about me 0.80 (0.06) 0.84/0.84
Kids try to keep me from being part of activities or in their group 0.58 (0.06) 0.68/0.59
Kids try to make their friends ignore me or stop talking to me 0.71 (0.06) 0.79/0.74
PR Overt Aggression
These kids hit or kick others 0.77 (0.06) 0.77/0.75
These kids push others around 0.88 (0.06) 0.86/0.86
These kids say mean things to others or call them names 0.95 (0.06) 0.89/0.93
PR Relational Aggression
These kids spread rumors about others 0.92 (0.06) 0.90/0.91
These kids try to keep others from being part of activities or groups 0.82 (0.06) 0.80/0.82
These kids try to make others ignore or not talk to certain children 0.87 (0.06) 0.88/0.87
PR Overt Victimization
These kids get hit or kicked by others 0.55 (0.05) 0.81/0.63
These kids get pushed around by others 0.67 (0.06) 0.94/0.80
Others say mean things to these kids or call them names 0.70 (0.06) 0.97/0.85
PR Relational Victimization
Others spread rumors about these kids 0.65 (0.05) 0.90/0.80
Others try to keep these kids from being part of activities or groups 0.65 (0.06) 0.94/0.75
Others try to get kids to ignore or not talk to these kids 0.66 (0.06) 0.93/0.79
Open in a new tab

Note. All factor loadings are significant at p < .001. SR = self-reported. PR = peer-reported. In column 3, standardized estimates for males are presented on the left and standardized estimates for females are presented on the right.

Table 4.

Results of Comparisons of Latent Means, Variances, and Correlations

Males Females χ2 df Δ χ2 Δ df p
Equality of Means M M 1532.97 556 219.23 44 < .001
PR Gender Nonconformity 0.06 0 1532.67 555 .30 1 ns
SR Overt Aggression 0.08 0 1532.08 555 .89 1 ns
SR Relational Aggression 0.08 0 1527.74 555 5.23 1 ns
SR Overt Victimization 0.34 0 1515.05 555 17.92 1 < .001
SR Relational Victimization -0.22 0 1509.90 554 5.15 1 ns
PR Overt Aggression 0.10 0 1480.34 554 34.71 1 < .001
PR Relational Aggression -0.31 0 1471.38 553 9.00 1 < .01
PR Overt Victimization 0.40 0 1459.70 552 11.68 1 < .001
PR Relational Victimization 0.09 0 1458.89 551 0.81 1 ns
Homogeneity of Variances and Covariances --- --- 1419.52 541 200.59 45 < .001
Equality of Variances Ψ Ψ 1245.63 505 26.70 9 < .01
PR Gender Nonconformity 1.00 1.00 1245.54 504 .09 1 ns
SR Overt Aggression 1.02 1.00 1245.48 504 .15 1 ns
SR Relational Aggression .56 1.00 1239.12 504 6.51 1 ns
SR Overt Victimization 1.45 1.00 1243.93 504 1.70 1 ns
SR Relational Victimization 1.22 1.00 1245.51 504 .12 1 ns
PR Overt Aggression .97 1.00 1245.53 504 .10 1 ns
PR Relational Aggression .98 1.00 1245.61 504 .02 1 ns
PR Overt Victimization 1.95 1.00 1242.13 504 3.50 1 ns
PR Relational Victimization 2.01 1.00 1239.49 504 6.14 1 ns
Equality of Correlations r r 1380.55 532 161.62 36 < .001
PR GNC ↔ SR Overt Aggression .09 .26 1377.76 531 2.79 1 ns
PR GNC ↔ SR Relational Aggression .18 .23 1379.36 531 1.19 1 ns
PR GNC ↔ SR Overt Victimization .12 .13 1379.25 531 1.30 1 ns
PR GNC ↔ SR Relational Victimization .17 .21 1380.43 531 0.12 1 ns
PR GNC ↔ PR Overt Aggression .29 .67 1359.03 531 21.52 1 < .001
PR GNC ↔ PR Relational Aggression .35 .56 1358.96 530 0.07 1 ns
PR GNC ↔ PR Overt Victimization .18 .31 1357.78 530 1.25 1 ns
PR GNC ↔ PR Relational Victimization .24 .56 1358.70 530 0.33 1 ns
Open in a new tab

Note. PR = peer-reports, SR = self-reports, and GNC = gender nonconformity. Male and female unstandardized estimates in columns two and three are from the fully unconstrained model. The chi-square test statistics and related difference in chi-square statistics resulted from the sequentially constrained, nested model comparisons.

Mean-level differences

Comparisons of latent means in nested MACS models (see Table 4) indicated that males self-reported higher levels overt victimization (d = .42) than females. Further, males were more likely to receive peer nominations of overt aggression (d = .33) and victimization (d = .24), and were less likely to receive nominations of relational aggression (d = -.39) than females. No sex difference emerged in the receipt of peer nominations of gender nonconformity.

Differences in associations

Peer-reported gender nonconformity was positively associated with self- and peer-reports of overt and relational aggression and victimization (see Table 4). No sex differences were found among the associations between peer-reported gender nonconformity and self-reports of aggression and self- and peer-reports of victimization, regardless of form; further, these associations were consistently positive and ranged in size from small (found for self-reports) to medium (found for peer-reports). Specifically, peer-reported gender nonconformity had small, positive associations with self-reported overt aggression (rconstrained = .19, p < .01) and relational aggression (rconstrained = .20, p < .01). Peer-reported gender nonconformity was not significantly associated with self-reported overt victimization (rconstrained = .10, p > .05); however, it was positively associated with self-reported relational victimization (rconstrained = .19, p < .01). Further, peer-reported gender nonconformity was moderately associated with peer-reported overt victimization (rconstrained = .32, p < .001) and relational victimization (rconstrained = .39, p < .001). Comparisons of latent correlations by sex (i.e., moderation by participant sex) yielded a significant difference in the association between peer-reported gender nonconformity and peer-reported overt aggression, but not peer-reported relational aggression (see Table 3). Specifically, the association between peer-reported gender nonconformity and peer-reported overt aggression was larger for females (r = .67, p < .001) than (r = .29, p < .001) males. The association between peer-reported gender nonconformity and peer-reported relational aggression was moderately positive (rconstrained = .46, p < .001).

Discussion

These findings provide additional support for the notion that early adolescents whose gender expressions do not conform to societal norms are at risk for negative peer relationships. In this study, early adolescents perceived as gender nonconforming by their peers self-reported higher amounts of overt and relational aggression and relational victimization in their peer relationships. Similarly, these individuals were also perceived by their peers as more overtly and relationally aggressive and victimized. These findings support converging evidence that early adolescents perceived as gender nonconforming may be a target for peer aggression (e.g., Aspenlieder, Buchanan, McDougall, & Sippola, 2009; D’Augelli, Grossman & Starks, 2006; Horn, 2007) and may enact aggression toward their peers (e.g., Carver, Yunger, & Perry, 2003).

We replicated and expanded upon the study by Aspenlieder and colleagues (2009) that documented associations between self- and peer-reported gender nonconformity and victimization among a sample of early adolescents. Specifically, we went beyond this initial study by Aspenlieder and colleagues by including multiple reports of overt and relational forms of peer victimization and included a focus on peer overt and relational aggression. In fact, this was the first study to examine the association between gender nonconformity and multiple forms of peer aggression among early adolescents, expanding on the work by Carver and colleagues (2003) and Yunger and colleagues (2004). Although others have examined peer-reported bullying (i.e., aggression) and victimization and differences among gender referred children (e.g., youth with gender identity disorder diagnoses; Wallien et al., 2010), the focus on early adolescence in the current study is a critical strength and contribution to the literature given that norms surrounding conformity to gender roles are typically lower in middle childhood compared to early adolescence (e.g., Alfieri, 1996; Hill & Lynch, 1983; Stoddart & Turiel, 1985).

Conformity to Gendered Norms and Peer Relationships

Consistent with prior research, we documented that young adolescents who do not conform to gendered norms are at greater risk for peer victimization (e.g., Aspenlieder, Buchanan, McDougall, & Sippola, 2009; D’Augelli, Grossman & Starks, 2006). Similar to the findings of Aspenlieder and colleagues, peer-reported gender nonconformity was associated with both self- and peer-reports of victimization for males; however, in contrast to their study, we did not find that gender nonconforming females were less likely to report victimization. This discrepancy requires further investigation, but may be related to the difference in measurement used to assess gender nonconformity in the two studies. That is, in the current study, we used a more global item to assess gender nonconformity perceptions by peers (e.g., the item may get at behaviors, style of clothing, choices of activities), while the measure used by Aspenlieder and colleagues assessed behavioral characteristics (e.g., willing to take risks as masculine). Therefore, the gender nonconforming girls in their study may have been seen as more assertive or risk-taking, which may be less associated with peer victimization, compared to a gender nonconforming female adolescent who typifies masculinity via clothing choices. Further, our findings were in contrast to the study by Wallien et al. (2010), given that they did not find that gender nonconforming youth were at greater risk for peer victimization; importantly, their study was conducted with children whereas the current study focused on early adolescents, suggesting that there may be an important developmental difference that emerges. Thus, these discrepancies point to the need for future research to prospectively examine the developmental associations between gender nonconformity and peer victimization across the transition from childhood to adolescence.

Beyond peer victimization, we found moderate to large associations between peer-reported gender nonconformity and peer- and self-reported overt and relational aggression. Further, gender nonconforming females were more likely than their male counterparts to be perceived as overtly aggressive by their peers. The larger association between overt aggression and perceived gender nonconformity for females may be understood as a proxy from which early adolescents perceive these female peers as possessing more male characteristics (i.e., overt aggression is typically associated with males; Card et al., 2008; Hyde, 1984; Maccoby & Jacklin, 1974), and provides some evidence that overt aggression may be perceived as non-normative for females. Interestingly, the sex difference in overt aggression was only found using peer-reports, and was not present in self-reports of aggression. Thus, it may be the case that females use of overt aggression is more saliently associated with gender nonconformity among peers, but not among adolescents themselves. While there was no sex difference in the association between gender nonconformity and peer-reported relational aggression, the association was significant suggesting that higher levels of perceived gender nonconformity are associated with higher levels of peer-reported relational aggression. Thus, even though previous research has found that males and females use this type of aggression equally (i.e., Card et al., 2008), male adolescents who enact relational aggression may be perceived as more feminine by their peers. Further, females who are perceived as more gender nonconforming by their peers may be perceived as more aggressive overall by their peers (both overtly and relationally). These findings are particularly important in light of other work that suggests that gender non-normative forms of aggression are associated with higher levels of negative social-psychological outcomes (Crick, 1997) and problematic peer relationships (e.g., Kochel et al., 2012).

Limitations and Future Directions

Although this study contributes to the emerging literature on peer relations of gender nonconforming early adolescents by including a focus on the multiple forms of aggression and victimization from the perspectives of multiple reporters, it is not without limitations. One limitation of this study is that the data are cross-sectional; that is, we cannot ascertain the directionality among gender nonconformity, aggression, and victimization. Although we have conceptualized gender nonconformity primarily as an antecedent of victimization and aggression, it is possible that experiences with peers also influence gender expression (e.g., Ewing Lee & Troop-Gordon, 2011a, 2011b). Second, shared method variance between our measures of peer-reported gender nonconformity and peer-reported aggression and victimization might have artificially inflated these correlations. That is, the associations were consistently higher between peer-reported gender nonconformity and peer reports of aggression and victimization, compared to self-reports of these constructs. Notably, these associations were stronger, but they were consistently related in the same direction; thus, future research is needed to ascertain the strength of these associations, possibly using observational methods, which may reduce self-reporting bias and potential issues with peer-reports (e.g., gender nonconformity may be an indicator of general dislike, rather than a true indicator of nonconformity). Third, given that a relatively small percentage of adolescents are consistently more gender nonconforming than their peers (e.g., 10% of boys and 20% of girls; Sandberg, Meyer-Bahlburg, Ehrhardt, & Yager, 1993), our sample size was relatively small to thoroughly examine the associations of interest; therefore, it will be important for future studies to examine these associations using larger, more representative samples. Fourth, our test of the unique associations between gender nonconformity and the different forms of aggression and victimization is limited given the high correlations found between forms within reporter; nonetheless, nested model comparisons did reveal that these forms were distinguishable. Future research is needed to create more distinguishable measures of the different forms of aggression and victimization, particularly among peer-reports of different forms of aggression and self-reports of victimization. Finally, our indicator of gender nonconformity was limited in that it was a single item and lacked specificity (i.e., it broadly assessed gender nonconformity without a formal definition). As noted above, it may be the case that our single item that assessed gender nonconformity actually assessed general dislike or negative evaluations of peers. Using post-hoc analyses, we evaluated this possibility by examining the association between gender nonconformity and an item that assessed social preference (i.e., “These kids are liked by other kids”), with the expectation that peer perceptions of gender nonconformity should be negatively associated with social preference if gender nonconformity is really an indicator of dislike. The bivariate correlation between gender nonconformity and this item was not significant for males (r = .16, p = .07) or females (r = .03, p = .65); thus, these correlations reduce some concern about the lack of specificity in our indictor of gender nonconformity. Nonetheless, future research should use multiple items to assess perceptions of gender nonconformity by peers.

Future research is also needed to understand the complex associations between adolescent gender nonconformity and overt and relational forms aggression. It also may be possible that gender nonconforming adolescents, especially females, report aggressive behavior and are viewed by their peers as aggressive because they are acting in self-defense (e.g., Wyss, 2004), given the consistent positive associations between peer-reported gender nonconformity and self- and peer-reports of overt and relational victimization. Similarly, previous research on the experiences of sexual minority youth has documented that they are more likely to engage in violent behaviors as a result of increased experiences of harassment (Russell, Franz, & Driscoll, 2001). Longitudinal research is needed that attempts to disentangle the predictive relations of gender nonconformity with forms of aggression (and victimization). Further, longitudinal research is also needed to examine whether or not the association between gender nonconformity and peer victimization and aggression changes developmentally. According to the previous research, conformity to gender norms is intensified in early adolescence (e.g., Hill & Lynch, 1983; Stoddart & Turiel, 1985); however, this research is based solely on attitudes and has not examined the stability of the relation between gender nonconformity and peer victimization developmentally.

Given that previous research has documented links between peer victimization based on gender nonconformity and psychosocial (e.g., internalizing problems; Yunger et al., 2004) and academic adjustment (e.g., Toomey et al., 2012), future research should examine whether or not overt and relational victimization based on gender nonconformity predict these outcomes differentially. Further, future studies should attempt to provide a more comprehensive examination to fully understand the conclusions raised by Crick (1997) about the association between gender non-normative aggression and poorer psychosocial adjustment. That is, given that the current study documented small to medium associations between gender nonconformity and peer victimization, one should also examine the possible interactions between the multiple forms of victimization and aggression, gender, and gender nonconformity to explain interindividual differences in psychosocial adjustment.

Conclusion

In conclusion, how gender is expressed (e.g., appearance, involvement in gender-normed activities) is an important characteristic to consider when understanding the complex phenomenon of adolescent peer relationships. This study builds on previous research that examines adolescent gender nonconformity and peer relationships and provides new information concerning the complex associations among gender, gender nonconformity, multiple forms of peer victimization and aggression, and the usefulness of considering different reporters. Specifically, the present study highlights the importance of the inclusion of multiple reports and attention to the multiple forms of peer victimization and aggression in understanding the experiences of gender nonconforming early adolescents. Further, perceived gender nonconformity was positively associated with all forms of peer victimization and aggression, a finding that warrants more attention to how to reduce heteronormativity in schools.

Acknowledgments

Support for the first author’s time was partially provided by a grant from the NIMH (T32 MH018387). Support for data collection was provided by a grant from NICHD (R21HD061345) to the second author.

Footnotes

1

Because the focus of the current study is on the associations of gender nonconformity with self- and peer- reported overt and relational aggression and victimization, we only report correlations that contain the gender nonconformity construct. Information about gender differences on correlations containing other latent constructs is available from the authors upon request (note: bivariate correlations among all constructs are presented in Table 1).

Contributor Information

Russell B. Toomey, Email: russell.toomey@asu.edu.

Noel A. Card, Email: ncard@email.arizona.edu.

Deborah M. Casper, Email: dcasper@email.arizona.edu.

References

  1. Archer J. Sex differences in aggression in real-world settings: A meta-analytic review. Review of General Psychology. 2004;8:291–322. [Google Scholar]
  2. Alfieri T. Gender stereotypes during adolescence: Developmental changes and the transition to junior high school. Developmental Psychology. 1996;32:1129–1137. [Google Scholar]
  3. Aspenlieder L, Buchanan CM, McDougall P, Sippola LK. Gender nonconformity in pre- and early adolescence. European Journal of Developmental Science. 2009;3:3–16. [Google Scholar]
  4. Brown BB. Peer groups and peer group cultures. In: Feldman SS, Elliott GR, editors. At the threshold: The developing adolescent. Cambridge, MA: Harvard University Press; 1990. pp. 171–196. [Google Scholar]
  5. Cairns RB, Xie H, Leung MC. The popularity of friendships and the neglect of social networks: Toward a new balance. In: Bukowski WM, Cillessen AH, editors. Sociometry then and now: Building on six decades of measuring children’s experiences with the peer group. San Francisco: Jossey Bass; 1998. pp. 25–53. [Google Scholar]
  6. Card NA, Isaacs J, Hodges EVE. Correlates of school victimization: Recommendations for prevention and intervention. In: Zins JE, Elias MJ, Maher CA, editors. Bullying, and peer harassment: A handbook of prevention and intervention. New York: Haworth Press; 2007. pp. 339–366. [Google Scholar]
  7. Card NA, Stucky BD, Sawalani GM, Little TD. Direct and indirect aggression during childhood and adolescence: A meta-analytic review of gender differences, intercorrelations, and relations to maladjustment. Child Development. 2008;79:1185–1229. doi: 10.1111/j.1467-8624.2008.01184.x. [DOI] [PubMed] [Google Scholar]
  8. Carver PR, Yunger JL, Perry DG. Gender identity and adjustment in middle childhood. Sex Roles. 2003;49:95–109. [Google Scholar]
  9. Cheung GW, Rensvold RB. Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling. 2002;9:233–255. [Google Scholar]
  10. Chesir-Teran D. Conceptualizing and assessing heterosexism in high schools: A setting-level approach. American Journal of Community Psychology. 2003;31:267–279. doi: 10.1023/a:1023910820994. [DOI] [PubMed] [Google Scholar]
  11. Cohen J. Statistical power analysis for the behavioral sciences. 2. Hillsdale, NJ: Erlbaum; 1988. [Google Scholar]
  12. Collaer ML, Hines M. Human behavioral sex differences: A role for gonadal hormones during early development? Psychological Bulletin. 1995;118:55–107. doi: 10.1037/0033-2909.118.1.55. [DOI] [PubMed] [Google Scholar]
  13. Crick NR. Engagement in gender normative versus nonnormative forms of aggression: Links to social-psychological adjustment. Developmental Psychology. 1997;33:610–617. doi: 10.1037//0012-1649.33.4.610. [DOI] [PubMed] [Google Scholar]
  14. Crick NR, Bigbee MA. Relational and overt forms of peer victimization: A multi-informant approach. Journal of Consulting and Clinical Psychology. 1998;66:337–347. doi: 10.1037//0022-006x.66.2.337. [DOI] [PubMed] [Google Scholar]
  15. D’Augelli AR, Grossman AH, Starks MT. Childhood gender atypicality, victimization, and PTSD among lesbian, gay, and bisexual youth. Journal of Interpersonal Violence. 2006;21:1462–1482. doi: 10.1177/0886260506293482. [DOI] [PubMed] [Google Scholar]
  16. Ewing Lee EA, Troop-Gordon W. Peer processes and gender role development: Change in gender atypicality related to negative peer treatment and children’s friendships. Sex Roles. 2011a;64:90–102. [Google Scholar]
  17. Ewing Lee EA, Troop-Gordon W. Peer socialization of masculinity and femininity: Differential effects of overt and relational forms of peer victimization. British Journal of Developmental Psychology. 2011b;29:197–213. doi: 10.1111/j.2044-835X.2010.02022.x. [DOI] [PubMed] [Google Scholar]
  18. Gay, Lesbian, and Straight Education Network (GLSEN) & Harris Interactive. Playgrounds and prejudice: Elementary school climate in the United States: A survey of students and teachers. New York: Gay, Lesbian, and Straight Education Network; 2012. [Google Scholar]
  19. Hill J, Lynch M. The intensification of gender-related role expectations during early adolescence. In: Brooks-Gunn J, Petersen A, editors. Girls at puberty. New York: Plenum; 1983. pp. 201–228. [Google Scholar]
  20. Horn S. Adolescents’ peer interactions: Conflict and coordination among personal expression, social norms, and moral reasoning. In: Nucci L, editor. Conflict, contradiction, and contrarian elements in moral development and education. Mahwah, NJ: Lawrence Erlbaum Associates; 2005. pp. 113–129. [Google Scholar]
  21. Horn S. Adolescents’ acceptance of same-sex peers based on sexual orientation and gender expression. Journal of Adolescence. 2007;36:363–371. doi: 10.1007/s10964-006-9111-0. [DOI] [PubMed] [Google Scholar]
  22. Hyde JS. How large are gender differences in aggression? A developmental meta-analysis. Developmental Psychology. 1984;20:722–736. [Google Scholar]
  23. Jackson S. Gender, sexuality, and heterosexuality: The complexity (and limits) of heteronormativity. Feminist Theory. 2006;7:105–121. [Google Scholar]
  24. Juvonen J, Nishina A, Graham S. Self-views versus peer perceptions of victim status among early adolescents. In: Juvonen J, Graham S, editors. Peer harassment in school: The plight of the vulnerable and victimized. New York: The Guilford Press; 2001. pp. 105–124. [Google Scholar]
  25. Kline RB. Principles and practice of structural equation modeling. 3. New York: The Guilford Press; 2011. [Google Scholar]
  26. Kochel KP, Miller CF, Updegraff KA, Ladd GW, Kochenderfer-Ladd B. Associations between fifth graders’ gender atypical problem behavior and peer relationships: A short-term longitudinal study. Journal of Youth and Adolescence. 2012;41:1022–1034. doi: 10.1007/s10964-011-9733-8. [DOI] [PubMed] [Google Scholar]
  27. Liben LS, Bigler RS. The developmental course of gender differentiation. In: Overton W, editor. Monographs of the Society for Research in Child Development. 2. Vol. 67. Boston: Blackwell; 2002. pp. vii–147. [PubMed] [Google Scholar]
  28. Little TD, Card NA, Slegers DW, Ledford E. Representing contextual effects in multiple-group MACS models. In: Little TD, Bovaird JA, Card NA, editors. Modeling contextual effects in longitudinal studies. New York: Routledge; 2007. pp. 121–148. [Google Scholar]
  29. Little TD, Jones SM, Henrich CC, Hawley PH. Disentangling the “whys” from the “whats” of aggressive behaviour. International Journal of Behavioral Development. 2003;27:122–133. [Google Scholar]
  30. Maccoby EE, Jacklin CN. The psychology of sex differences. Stanford, CA: Stanford University Press; 1974. [Google Scholar]
  31. Oswald RF, Blume LB, Marks SR. Decentering heteronormativity: A model for family studies. In: Bengston VL, Acock AC, Allen KR, Dilworth-Anderson P, Klein DM, editors. Sourcebook of family theory and research. Thousand Oaks: Sage; 2005. pp. 143–165. [Google Scholar]
  32. Russell ST, Franz BT, Driscoll AK. Same-sex romantic attraction and experiences of violence in adolescence. American Journal of Public Health. 2001;91:1276–1281. doi: 10.2105/ajph.91.6.903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sandberg DE, Meyer-Bahlburg HFL, Ehrhardt AA, Yager TJ. The prevalence of gender-atypical behavior in elementary school children. Journal of American Academic Child and Adolescence Psychiatry. 1993;32:306–314. doi: 10.1097/00004583-199303000-00010. [DOI] [PubMed] [Google Scholar]
  34. Smith TE, Leaper C. Self-perceived gender typicality and the peer context during adolescence. Journal of Research on Adolescence. 2006;16:91–103. [Google Scholar]
  35. Stoddart T, Turiel E. Children’s concepts of cross-gender activities. Child Development. 1985;56:1241–1252. [Google Scholar]
  36. Toomey RB, McGuire JK, Russell ST. Heteronormativity, school climates, and perceived safety for gender nonconforming peers. Journal of Adolescence. 2012;35:87–196. doi: 10.1016/j.adolescence.2011.03.001. [DOI] [PubMed] [Google Scholar]
  37. Wallien MSC, Veenstra R, Kreukels BPC, Cohen-Kettenis PT. Peer group status of gender dysphoric children: A sociometric study. Archives of Sexual Behavior. 2010;39:553–560. doi: 10.1007/s10508-009-9517-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Wyss SE. ‘This was my hell’: The violence experienced by gender non-conforming youth in US high schools. International Journal of Qualitative Studies in Education. 2004;17:709–730. [Google Scholar]
  39. Yunger JL, Carver PR, Perry D. Does gender identity influence children’s psychological well-being? Developmental Psychology. 2004;40:572–582. doi: 10.1037/0012-1649.40.4.572. [DOI] [PubMed] [Google Scholar]

RESOURCES