AGGRESSION TOWARD GENDER-NONCONFORMING PEERS 845 These self-appraisals do not necessarily reflect felt inc patibilit ad ender-atypicalpeerso ective.but at t tend to juc veness).Ov identity is perhap nder others (Ha 2006 hias or felt er diffe Mo ith one Tgender (st e strong ocu ch f orm of p ing ridicule) and they f within-gen identity to spu betwen-endo of gender identiy and a ugender-nonspeci peer group but p Frick. ng sh0 engths,goals, den )it ssion.forboth adultsand hildrer r identity.low felt gender typicality.gende 996:Mo 200 self-effica lthof.2008).Thus.high m might en Howeve like onfident er identity.n ent)probab 's ability to satisfy e and att with s the exacerbated they Thus em or l self-effic y fo children with strone r-nolar tity to s their gender adequacy by assling gender gende and in well he alt identity may not equir preadolescents sment nonconforming withm-g ight also som s take on s ades of tity.We allowed for this possibility in our analytic s dentity is nother cognitiv riable (co teracts with self m or a measure of self-efficacy to promo hildren with low intergroup s).malt cal peer ren with hi mSuc由a ny of th pattem would constitute evidence for both pathways. their inner conflict atte ing to show that the deficienc elat identity (int bias or felt diffe Our ach to h ent of g stitute clear-cut gend elf-dis self-dis incapable of displaying dominance.popularity.or good looksChildren with overconfident gender identity might aggress to￾ward gender-atypical peers for many of the same reasons that older persons do (e.g., moral outrage, distrust, threat to ingroup cohe￾siveness). Overconfident gender identity is perhaps most clearly represented when a strong form of between-gender identity (strong intergroup bias or felt pressure for gender differentiation) com￾bines with strong felt compatibility with one’s gender (strong felt gender typicality or gender contentedness). Thus, we examined whether each form of between-gender identity interacts synergis￾tically with each form of within-gender identity to spur children to attack gender-nonconforming peers. However, we also examined whether each between-gender form of gender identity combines with each of four gender-nonspecific self-appraisals to stimulate attacks on gender-nonconforming peers. These other self￾appraisals were global self-esteem and perceptions of self-efficacy for dominance, popularity, and physical attractiveness. High self￾esteem motivates people to act confidently on their self-perceived strengths, goals, identities, and values (Baumeister, 1998). When high self-esteem is paired with self-serving ideation (e.g., narcis￾sism), it sometimes fosters aggression, for both adults and children (Baumeister, Smart, & Boden, 1996; Morf & Rhodewalt, 2001; Thomaes & Bushman, 2011; Thomaes, Bushman, Stegge, & Olthof, 2008). Thus, high self-esteem might encourage children with strong ingroup favoritism or felt pressure for gender differ￾entiation to attack gender-nonconforming others. High self￾efficacy for dominance, popularity, or attractiveness might as well. High self-efficacy in these domains reflects confidence in one’s social status and in one’s ability to satisfy demonstration goals, or desires to be admired by peers; children with such goals tend to display adjustment difficulties, including aggression (O’Moore & Kirkham, 2001; Rodkin, Ryan, Jamison, & Wilson, 2013; Ru￾dolph, Abaied, Flynn, Sugimura, & Agoston, 2011; Salmivalli, 1998). It is conceivable that high self-efficacy in these domains empowers children with strong gender-polarizing identity to strive for admiration and status by discriminating against gender￾nonconforming peers. Overconfident gender identity may not always require strong between-gender identity, however. Strong within-gender identity— high felt gender typicality or gender contentedness—might also some￾times take on shades of overconfidence or narcissistic arrogance and promote attacks on gender-nonconforming peers. Perhaps this occurs when strong within-gender identity is accompanied by high self￾esteem or self-efficacy for dominance, popularity, or attractiveness. Thus, we also examined whether each form of within-gender identity interacts with self-esteem or a measure of self-efficacy to promote harassment of gender-atypical peers. Children with insecure gender identity might also discriminate against gender-nonconforming others for many of the same rea￾sons that older persons do (e.g., defensive lashing out at those who arouse their inner conflict, attempting to show that the deficiency lies in others and not themselves). Insecure gender identity is perhaps most clearly represented when a strong form of between￾gender identity (intergroup bias or felt pressure for gender differ￾entiation) combines with strong felt incompatibility with one’s gender (low felt gender typicality or low gender contentedness). Such pairings constitute clear-cut gender self-discrepancies. How￾ever, children with strong between-gender identity might also feel self-discrepant and anxious if they have low self-esteem or feel incapable of displaying dominance, popularity, or good looks. These self-appraisals do not necessarily reflect felt incompatibility with one’s gender collective, but at this age, children tend to judge their self-worth and competencies by comparing themselves to same-gender others (Harter, 2006), and self-appraisals on these dimensions are therefore likely to reflect how well children feel they are stacking up against same-sex others. Moreover, children with strong between-gender identity might focus on satisfying demonstration goals (e.g., feeling superior, being admired, avoid￾ing ridicule), and if they view themselves as valueless or as incapable of fulfilling these goals, their frustration and insecurity may be acute. Although research on self-discrepancies in children is scant, children who ardently strive for status and admiration in the peer group but perceive themselves as falling short do tend to be aggressive (Barry, Frick, & Killian, 2003; Harter & McCarley, 2004; Pauletti, Menon, Menon, Tobin, & Perry, 2012); it is plau￾sible that between-gender identity channels the aggression of these frustrated children toward gender-atypical targets. Thus, the inse￾cure pathway to aggression toward gender-nonconforming peers would be supported if for children with a strong form of between￾gender identity, lower felt gender typicality, gender contentedness, self-esteem, or self-efficacy for dominance, popularity, or attrac￾tiveness were to motivate attacks on gender-nonconforming peers. However, like overconfident gender identity, insecure gender identity may not always require strong between-gender identity. Children who feel incompatible with their gender (who feel gender atypical or discontent) probably experience some uncertainty over their adequacy as members of their gender collective even if they lack strong gender-differentiating stereotypes and attitudes; more￾over, their anxiety may be exacerbated if they have low self￾esteem or feel incapable of displaying attributes needed for status in their peer group. Thus, low self-esteem or low self-efficacy for dominance, popularity, or attractiveness might potentiate a ten￾dency on the part of children who feel gender incompatible to demonstrate their gender adequacy by hassling gender￾nonconforming peers. The overconfident and insecure pathways may well be alternate routes to preadolescents’ harassment of gender-nonconforming peers: Some children might harass gender-atypical peers because of overconfident gender identity, others because of insecure gender identity. We allowed for this possibility in our analytic strategy by probing for a quadratic effect of one cognitive variable (e.g., self-esteem) at different levels of another cognitive variable (con￾ceptualized as the moderator; e.g., intergroup bias). For example, it might be that for children with high intergroup bias (but not for children with low intergroup bias), maltreatment of gender￾atypical peers is greater for children with either high or low self-esteem than for children with intermediate self-esteem. Such a pattern would constitute evidence for both pathways. The Present Study in Relation to Contemporary Personality Theory Our approach to conceptualizing and evaluating cognitive path￾ways to harassment of gender-atypical others is grounded in con￾temporary cognitive metatheories of personality (Cervone, 2004; Mischel & Shoda, 1995, 2008; Read et al., 2010; Zakriski, Wright, & Underwood, 2005). These theories offer conceptual strategies for generating hypotheses linking people’s stable cognitive pro￾cessing structures—their values, expectancies, goals, beliefs, iden￾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. AGGRESSION TOWARD GENDER-NONCONFORMING PEERS 845