Is Homophobia Associated with an Implicit Same-Sex Attraction?

Abstract

Some theorists propose that homophobia stems from underlying same-sex attraction. A few studies have tested this hypothesis, yet without a clear measure of implicit sexual attraction, producing mixed results. For the first time, we test this attraction-based account of homophobia among both men and women using an implicit measure of sexual attraction. No evidence of an attraction-based account of homophobia emerged. Instead, implicit same-sex attraction was related to positive evaluations of gay men and lesbians among female participants. Even in targeted analyses examining the relation between implicit same-sex attraction and homosexual evaluations among only those theoretically most likely to demonstrate an attraction-based homophobic effect, implicit same-sex attraction was not associated with evaluations of homosexuals or was associated with more positive evaluations of homosexuals. In addition, explicit same-sex attraction was related to positive evaluations of gay men and lesbians for male participants. These results are more in keeping with the attitude-similarity effect (i.e., people like, rather than dislike, similar others).

Acknowledgments

The research was sponsored by a Social Sciences and Humanities Research Council of Canada grant (410-2007-2133) to the second author.

Notes

¹Historically, negativity toward homosexuals was termed homophobia (Weinberg, 1972). Many contemporary researchers prefer less clinical terms (e.g., sexual prejudice, Herek, 2000; antigay bias, Mahaffey, Bryan, Ito, & Hutchison, 2011). Each term shares a negative evaluation of gay men and/or lesbians. We use the terms homophobia and negative evaluations interchangeably.

²Reviews of the Oedipus complex are beyond the scope of the present article and are available elsewhere, e.g., Hartocollis, 2005.

³Meier et al. (2006) suggested two subgroups of homophobic men: defensive (those high in homophobia and self-defensive motives) and nondefensive (those high in homophobia and low in defensive motives).

⁴See Snowden et al., 2008, for specific pictures and words used. All pictures depicted nudity or partial nudity. Note that picture 4332 is cited in Snowden et al.'s study in error. Rather, picture 4232 was employed in both the Snowden et al. study and the current study (Robert J. Snowden, personal communication, September 1, 2010).

Note. Heterosexual participants (79 men, 139 women). Attraction measures are based on difference scores; negative or positive scores indicated opposite-sex and same-sex attraction, respectively. Standard deviations are provided in parentheses. IAT means for both men and women significantly differ from 0 at p < .01.

⁺ p ≤ .08. ***p ≤ .001.

Note. Heterosexual participants (79 men, 139 women).

⁺ p ≤ .10. *p ≤ .05. **p ≤ .01.

⁵Controlling for measurement error (see Nunnally, 1978), support for an attraction-based account of homophobia weakened rather than strengthened among women and produced no changes among men.

⁶Note that when limiting analyses based only on categorically indicated heterosexuality, results are very similar. In addition, conservatively limiting the sample in this way for the main analyses did not significantly alter results.