Bi Now, Gay Later: Implicit and Explicit Binegativity Among Irish University Students

Abstract

To date, no assessment of explicit attitudes toward bisexual men and women has occurred in the Republic of Ireland. Further, there are no published studies examining implicit attitudes toward this group. The current research addressed these omissions by investigating bi-negativity, at explicit and implicit levels, within an Irish context. Various pen-and-paper measures were administered to 208 undergraduate students (170 females) enrolled in 2nd- and 3rd-year psychology courses at a large university in Western Ireland. Among the most pertinent findings: explicit binegativity was somewhat prevalent within the sample, overall implicit attitudes reflected a bias toward heterosexuals and against bisexuals, men evidenced greater levels of explicit bi-negativity than women and no support was obtained for a model of cognitive consistency, which aims to articulate the various conditions in which implicit and explicit attitudes are intercorrelated. Limitations of the current study and directions for future research are discussed.

KEYWORDS:

• bisexuality
• bi-negativity
• prejudice
• sexual minorities
• Ireland
• attitudes

Notes

^aStatistically significant correlations (p < .05).

^aReverse-coded items.

1. The implicit and explicit measures were not counterbalanced as the order of presentation does not seem to exert any meaningful effects (Nosek, 2005).

2. Counterbalancing of congruent and incongruent pairings was employed in accordance with Steffens’ (2005) recommendation that alternate versions be administered to avoid a “context effect,” in which administration of the first IAT influences the second. She posits this is plausible if we conceive of the IAT as measuring patterns of activation in a semantic network and of such patterns being influenced by previous activations.

3. Eight extreme scores were removed; however, inclusion of these scores (M _CAT = 1.57, SD = 4.33) did not change the results.

4. A paired samples t test revealed a significant difference between standardized tolerance (M = 1.74, SD = .64) and stability subscales (M = 2.5, SD = .61); t(192) = −19.24, p < .01, d = −1.28.

5. It should be noted that Eliason (1997) experienced a similar problem in her research: “In fact, the respondents themselves often used the ‘don't know’ option and indicated in additional written comments that they never encountered bisexuals or had any education on bisexuality” (p. 325).

6. Concerns about generalizability should be tempered with the realization that its importance may be secondary to testing the viability of a theory or the existence of a given phenomenon (see Mook, 1983, for an informative discussion about this matter).