Stereotype Deduction About Bisexual Women

Abstract

Bisexuals are an invisible sexual minority. However, at the same time, bisexuals are stereotypically associated with confusion and promiscuity. Stereotype learning theories suggest that individuals who are unfamiliar with a social group are less likely to have stereotypical beliefs about its members. In contrast, it has been recently hypothesized that stereotypes about bisexuality are not necessarily learned but rather deduced based on common conceptualizations of sexuality. Because stereotypes are suppressed only if they are recognized as offensive, lack of knowledge regarding bisexual stereotypes should actually enhance their adoption. To assess the strength of the two competing accounts, we examined the relationship between explicit knowledge of bisexual stereotypes and stereotypical evaluation of bisexual individuals. Heterosexual participants (N = 261) read a description of two characters on a date and evaluated one of them. Bisexual women were evaluated as more confused and promiscuous relative to nonbisexual women. Moreover, the stereotypical evaluations of bisexual women were inversely related to knowledge about these stereotypes. The findings support the notion that bisexual stereotypes are not learned but rather deduced from shared assumptions about sexuality. Consequently, public invisibility not only exists alongside bisexual stereotypes but might also exacerbate their uninhibited adoption.

Acknowledgments

We thank Tatyana Schwartzman-Zivony for her able assistance in proofreading.

Notes

¹ The illustrations are available from the corresponding author upon request.

² To make sure that the arbitrary 10% cutoff point did not affect the reported results, we repeated the statistical analyses reported below with a 5% cutoff point with no change in the results.

³ Effects with p values below .05 are reported. However, a conservative Bonferroni correction for multiple ANOVAs would suggest that only p values below .0055 should be considered significant. In that case, the effect of target’s bisexuality on neuroticism (p = .007), the interaction between target’s bisexuality and gender match on extraversion (p = .01) and on openness to experience (p = .029) should be considered to be approaching statistical significance.

⁴ To avoid possible problems from dichotomizing continuous data (MacCallum, Zhang, Preacher, & Rucker, 2002), we also examined the correlation between bisexual stereotype knowledge and stereotypical evaluations. The correlation was highly significant among participants who evaluated a bisexual target, r (129) = −.245, p = .005, but not among participants who evaluated a nonbisexual target r (127) = .016, p = .85. Analysis with Fisher’s r-to-z transformation confirmed that the difference between the two correlations was significant, z = 2.1, p = .036.

⁵ Note that the bisexual stereotype knowledge score included knowledge of stereotypes that were not evaluated in this study (e.g., masculine). However, all the analyses yielded the same conclusions when we calculated the participants’ stereotype knowledge based on their indication of the four most stereotypically related categories (promiscuous, confused, not really bisexual, and nonmonogamous).

⁶ We included untrustworthiness into the new measure of stereotypical evaluation, even though it did not significantly differ between bisexuals and nonbisexuals, as it was part of the originally hypothesized pattern. Removing untrustworthiness from the analysis did not change any of the reported results.