844 PAULETTL COOPER.AND PERRY diffe entiating beliefs and identities h nder-atv cal other night.individually and in combination with other yariables.affec ats to their se gender-atyp der favorably th the other) der-atypical ecting other people to recipr sec atment of nd gen with one's gender anxiety over one's adegu as a member of one be gender fit s of ende wel to the esire t th Tha he o th Felt d Eisler,Moore 2001;Kimme 1994:Parro 2001 is par ce adisturbin eater use of and b a gap betwec what one feels By guilt self-rejection ange 助d iated with elational difficultie 198( cr. (for oys:P.Ro 20:P. 1992).Males with a strongly gen 200 eithe lack self-effi d male c-typica two dimensions of gender identit gcnder to inhibit feminine behavi nder identity because they re children' ve own gen ption of salien and valued attribute 8.De and ng that the gender inadequacy reside nsions of gender ide ntity are usuall mentof both the o fiden ure cogni nent in gender-diff ng by et al Egan Perry,2001;Smith Leaper,2006 emulate the the result is an people hold the ail to one anoth (e.g..Egan persons to har ender-nonco nstruct (e.g.,s hr oh a pern der lens:Bem The Present Study he betw -gender and the within-gender forms of gender identity pt.desc ribed be ment of ger nder-atypical peers arch to be associated with maltre tment of aibltidim ollec e (As eers. There grounds for expecting th patters to apply to preadolescent times dislike and are angered by gender-atypical females (Parrott & Gallagher, 2008). Presumably, people with strong gender￾differentiating beliefs and identities harass gender-atypical others because they perceive them as threats to their worldviews, as morally disgusting, as challenges to the positive distinctiveness of their gender collective, as disloyal and untrustworthy members of the collective, or as expecting other people to reciprocate their gender-atypical behavior. A second cognitive pattern associated with maltreatment of gender-atypical persons reflects insecure gender identity— uncer￾tainty and anxiety over one’s adequacy as a member of one’s gender collective. Insecurity over one’s gender fit is especially conducive to attacks on gender-atypical others for individuals who also possess strong gender-differentiating stereotypes and attitudes (in this regard, they resemble persons with overconfident gender identity). That is, males who subscribe to rigid standards of hy￾permasculinity but perceive themselves as failing to live up to the standards tend to treat gender-atypical others harshly (Bernat, Calhoun, Adams, & Zeichner, 2001; Eisler & Skidmore, 1987; Franchina, Eisler, & Moore, 2001; Kimmel, 1994; Parrott, 2009; Vandello, Bosson, Cohen, Burnaford, &Weaver, 2008). Such males presumably experience a disturbing ought self-discrepancy (a gap between what one feels one should be and what one really is). Ought self-discrepancies are conducive to frustration, agitated distress, guilt, shame, self-rejection, anger, and loss of self￾regulatory control (C. S. Carver, Lawrence, & Scheier, 1999; Higgins, 1987; Swann & Bosson, 2008; Tangney, Wagner, Fletcher, & Gramzow, 1992). Males with a strongly gendered ought self-discrepancy are said to experience masculine gender role stress (Levant, 2011). Although they strive to be male typical, they either lack self-efficacy for self-prescribed male-typical be￾haviors (e.g., dominance, risk taking) or fear that in certain situ￾ations they will be unable to inhibit feminine behavior. Presum￾ably, they perceive in gender-atypical others what they fear in themselves, and this perception activates latent feelings of insecu￾rity (fears of gender inadequacy, of being perceived as gay, of rejection by same-sex others) that motivate them to lash out defensively at gender-atypical persons, thereby removing the threat and demonstrating that the gender inadequacy resides in other people and not themselves. Note that a common element of both the overconfident and insecure cognitive patterns is investment in gender-differentiating stereotypes and attitudes. When people adopt such standards and successfully emulate them, the result is an overconfident gender identity. When people hold the standards but fail to fulfill them, the result is an insecure gender identity. Perhaps either pattern disposes persons to harass gender-nonconforming others. The Present Study To our knowledge, no study has investigated the factors that lead preadolescent children to single out gender-nonconforming peers for abuse. The principal purpose of the present study was to investigate the role of gender identity in preadolescents’ maltreat￾ment of gender-atypical peers. Gender identity is a multidimensional construct encompassing a person’s felt compatibility with, quality of motivation to fit in with, and evaluation of a gender collective (Ashmore, Deaux, & McLaughlin-Volpe, 2004; Egan & Perry, 2001; Tobin et al., 2010). Using a model of gender identity proposed by Egan and Perry (2001), we examined how four dimensions of gender identity might, individually and in combination with other variables, affect children’s tendencies to abuse gender-atypical peers. These dimen￾sions of gender identity were intergroup bias (perceiving one’s own gender more favorably than the other), felt pressure for gender differentiation (internalized pressure to avoid other-sex behaviors), felt gender typicality (perceived similarity to same-sex others), and gender contentedness (satisfaction with one’s gender). The first two dimensions—intergroup bias and felt pressure for gender differentiation—may be considered between-gender facets of gender identity in that they depend on perceiving, and likely exaggerating, differences between the sexes. Strong forms of these facets of gender identity may well contribute to the desire to interact with, fit in with, and emulate others of one’s gender, but they probably also encourage avoidance and disparagement of other-gender activities and persons. Felt pressure for gender dif￾ferentiation is in fact assessed as the expectation of punishment (e.g., ridicule, alienation) for cross-gender behavior (Egan & Perry, 2001). Intergroup bias is part of a set of intergroup cogni￾tions that also includes exaggeration of differences between the sexes, greater use of stereotypes, and homogenization of the other sex (Powlishta, 1995; Tajfel & Turner, 1979). By preadolescence, these gender-polarizing forms of identity are relatively immature and are associated with peer-relational difficulties, low self￾esteem, internalizing problems (for girls), and externalizing prob￾lems (for boys; P. R. Carver, Egan, & Perry, 2004; P. R. Carver, Yunger, & Perry, 2003; Corby, Hodges, & Perry, 2007; Egan & Perry, 2001; Smith & Leaper, 2006; Yunger, Carver, & Perry, 2004). The other two dimensions of gender identity—felt gender typ￾icality and gender contentedness—may be considered within￾gender facets of gender identity because they capture children’s feelings of compatibility with, and acceptance by, their own gen￾der collective. They are largely a function of children’s self￾perception of salient and valued same-gender attributes (e.g., same-sex friendships) and are not necessarily undermined by self￾perception of a few other-gender attributes (Egan & Perry, 2001; Spence, 1993). These dimensions of gender identity are usually positively correlated with self-esteem, social skills (e.g., prosocial behavior), and peer acceptance (P. R. Carver et al., 2003, 2004; Corby et al., 2007; Egan & Perry, 2001; Smith & Leaper, 2006; Yunger et al., 2004). Because the four dimensions of gender identity typically show low correlations with one another (e.g., Egan & Perry, 2001; Yunger et al., 2004), they cannot be considered interchangeable reflections of a single, common underlying construct (e.g., gender schematicity, or a generalized tendency to view self and others through a pernicious gender lens; Bem, 1981). Moreover, as noted, the between-gender and the within-gender forms of gender identity usually relate to adjustment in opposite ways. Using measures of the four gender identity dimensions (and of other potentially relevant aspects of self-concept, described be￾low), we examined whether the cognitive patterns found in previ￾ous research to be associated with maltreatment of gender￾nonconforming people by older persons (overconfident and insecure gender identity) also motivate preadolescents to attack gender-nonconforming peers. There are grounds for expecting both patterns to apply to preadolescents. This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. 844 PAULETTI, COOPER, AND PERRY