The Relation Between Childhood Gender Nonconformity and ...

Sex Roles (2008) 59:684?693 DOI 10.1007/s11199-008-9476-5

ORIGINAL ARTICLE

The Relation Between Childhood Gender Nonconformity and Adult Masculinity?Femininity and Anxiety in Heterosexual and Homosexual Men and Women

Richard A. Lippa

Published online: 2 August 2008 # Springer Science + Business Media, LLC 2008

Abstract Recalled childhood gender nonconformity (CGN) and adult traits--e.g., masculinity?femininity of occupational preferences (MF-Occ), self-ascribed masculinity? femininity (Self-MF), self-reported anxiety--were assessed in an ethnically diverse US convenience sample of California college students and gay pride attendees, including 238 heterosexual men, 127 gay men, 343 heterosexual women, and 197 lesbian women. CGN showed large heterosexual? homosexual and gender differences, with homosexual individuals higher than heterosexual individuals and women higher than men. CGN was significantly associated with both men's and women's MF-Occ and Self-MF, both within and across sexual orientations. CGN was associated with self-reported anxiety for men but not for women, a finding true for both heterosexual and gay men, suggesting that CGN had a more negative impact on men than women.

Keywords Childhood gender nonconformity . Femininity . Masculinity . Gender . Personality stability . Gender differences . Sexual orientation

Introduction

The research reported here investigated the relation between childhood gender nonconformity (CGN) and adult personality. Its purpose was to see if gender conformity and nonconformity show consistency from childhood to adulthood and if CGN has implications for adult adjustment,

R. A. Lippa (*) Psychology Department, California State University, Fullerton, Fullerton, CA 92834, USA e-mail: rlippa@fullerton.edu

both in heterosexual and homosexual individuals. Although the data reported here were collected in the USA, the current results may apply to other countries and cultures as well. CGN seems to be a universal phenomenon and it likely shows associations with adult sexuality, personality, and adjustment in many cultures (see Lippa 2008).

Few studies have directly assessed the relation between childhood gender nonconformity and adult personality. The most firmly established adult correlate of CGN is not personality per se, but rather sexual orientation (Lippa 2005b). Gender nonconforming children are relatively more likely to become homosexual adults, and gender conforming children are relatively more likely to become heterosexual adults. In a meta-analytic review of 41 studies, Bailey and Zucker (1995) reported large mean effect sizes (d=1.31 for men and .96 for women) for homosexual? heterosexual differences in recalled CGN. More recent studies continue to document strong links between CGN and sexual orientation (e.g., Cohen 2002; Dunne et al. 2000; Loehlin and McFadden 2003; Phillips and Over 1995; Safir et al. 2003; Skidmore et al. 2006; Whitam et al. 1998).

Sexual orientation, in turn, is reliably associated with a number of adult personality traits (Lippa 2005b, in press). It is most strongly associated with gender-related interests (e.g., occupational and hobby preferences, with d values for homosexual?heterosexual differences generally greater than 1.0). Heterosexual men and women tend to have interests typical of their gender, whereas homosexual men and women tend to have interests that are relatively more typical of the other gender. Sexual orientation is also associated with selfascribed masculinity?femininity (Self-M?F): Gay men describe themselves as moderately more feminine and less masculine than heterosexual men, on average, and lesbian women describe themselves as strongly more masculine and less feminine than heterosexual women, on average.

Sex Roles (2008) 59:684?693

685

Sexual orientation is more weakly associated with a number of Big Five personality traits and the Big Fiverelated traits of instrumentality and expressiveness (Lippa 2005b, in press). Both gay men and lesbian women score moderately higher on openness to experience than same-sex heterosexuals do. On a number of other Big Five traits, gay men and lesbian women show reliable, small-to-moderate cross-gender shifts. For example, gay men tend to score higher but lesbians tend to score lower on neuroticism than same-gender heterosexuals do, and gay men tend to score higher than heterosexual men do on expressiveness and agreeableness (traits considered to be stereotypically feminine), whereas lesbian women tend to score higher than heterosexual women do on instrumentality (a trait considered to be stereotypically masculine).

In sum, CGN is strongly associated with adult sexual orientation, which in turn is associated with a number of adult personality traits. If these relations are transitive, then CGN should also be associated with adult personality traits. The most straightforward prediction for men is that CGN will be associated with more feminine interests, higher selfascribed femininity, and possibly with Big Five traits such as neuroticism. The corresponding prediction for women is that CGN will be associated with more masculine interests, higher self-ascribed masculinity, and possibly with Big Five traits such as neuroticism. Because sexual orientation is most strongly associated with adults' gender-related interests and Self-MF, CGN is similarly predicted to be most strongly associated with these two traits. Although the previous predictions seem most likely to hold true for groups of men and women that are unselected for sexual orientation, they may also hold within sexual orientation groups--i.e., among just heterosexual men, homosexual men, heterosexual women, and homosexual women. Indeed, examining within-group associations between CGN and personality may help answer the question: Does sexual orientation mediate the relation between CGN and adult traits, or is the relation between CGN and adult traits, to some degree, independent of sexual orientation?

Broader Research on the Stability of Personality

Research on CGC and adult personality can be framed more broadly in terms of research on the stability of personality over the lifespan. Longitudinal research has focused most often on the stability of Big Five traits over periods of time ranging from a few years to many decades. Overall, such research has documented considerable temporal stability to personality, particularly during adulthood (Caspi and Roberts 2001; Roberts and DelVecchio 2000; Terracciano et al. 2006). Not surprisingly, Big Five traits prove to show more stability over shorter than over longer periods of time. Perhaps more surprising is the finding that some traits (e.g.,

extraversion and openness) tend to show stronger temporal stability than others (e.g., agreeableness). A small number of studies have investigated associations between childhood and adult assessments of Big Five traits. For example, Hampson and Goldberg (2006) reported the following crosstime correlations for 799 individuals who were assessed on Big Five traits both in elementary school and as adults, 40 years later: .27 for extraversion, .25 for conscientiousness, .17 for openness, .09 for agreeableness, and .00 for neuroticism (all significant except for the last).

Longitudinal studies of masculinity?femininity (M?F) have been less common than longitudinal studies of Big Five traits. A few studies have investigated the stability of adults' M?F, typically as assessed by omnibus personality inventories such as the California Psychological Inventory (e.g., see Helson and Moane 1987; Helson and Wink 1992). These studies show that M?F measures often display significant temporal consistency, as do other personality traits. The meaning of this stability is made ambiguous, however, by the fact that such M?F scales often have heterogeneous content, and thus the nature of the traits that are demonstrating stability is not clear (see Lippa 2001, 2005a).

Longitudinal studies of associations between CGN and adult personality are rare. However, the recent initiation of large-scale longitudinal studies of gender-related behaviors in infants and children promises to generate new findings on this topic (e.g., Hines et al. 2002). In a classic longitudinal study of personality, Block (1971) reported the temporal consistency of California Q Set ratings of 84 males and 87 females as assessed by clinical judges who had access to rich and varied information about participants. For the Q Set items, "behaves in a masculine style and manner" (for men) and "behaves in a feminine style and manner" (for women), Block reported the following cross-time correlations between junior high school and senior high school assessments: .37 for boys and .52 for girls. The corresponding correlations between senior high school and middle-adulthood assessments were .39 for men and .38 for women (all four correlations were significant at p ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download