Do attitude functions and perceiver demographics predict attitudes towards asexuality?

ABSTRACT

Research indicates asexual individuals experience stigma. Addressing this phenomenon, this study examined attitude functions – experiential, social-expressive, ego-defensive, and value-expressive – in the prediction of attitudes towards asexuality. As well, demographic variables – participant gender, religiosity, and sexual orientation – were examined vis-à-vis asexuality attitudes. Herek’s Function of Attitudes Inventory assessed asexual attitude functions. General attitudes were assessed using the Attitude towards Asexuality scale, feeling thermometers, and semantic differential scales. Participants were asked to imagine developing a relationship with an asexual person; attitudes towards the asexual target were assessed by belief statements specific to the person, a feeling thermometer, and target-specific semantic differential items. On average, all asexuality attitudes measures were rated favourably. Men, religious individuals, and exclusively heterosexual participants were generally less positive in their asexual attitudes. While statistically significant, these demographic differences were quite weak. Participants generally denied the attitude functions as the basis for their asexuality attitudes. The ego-defensive attitude function was strongly predictive of all asexual attitudes measures. The value-expressive function was a significant but small multiple regression predictor of some asexual attitudes. Understanding attitudes towards asexuality would be advanced by further consideration of how the attitude serves the social perceiver.

KEYWORDS:

• Asexuality
• attitudes
• attitude functions
• ego-defensive

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Participants of this study were not asked for permission for their data to be shared publicly, so supporting data is not available

Notes

1. Exclusion criteria available by request.

2. We use the term queer to refer to the meta-category of those with a sexual orientation other-than-heterosexual. We follow the Learning for Justice project (2021), who state, ‘[w]e recognise the complicated history of the word “queer” and that its reclamation as a positive or even neutral term of identity isn’t universally accepted … we use queer as an inclusive term to refer to those who fall outside of cisgender or heterosexual identities – not as a pejorative’ (p. 6).

3. Includes 87 who indicated Man plus 2 who indicated Transgender Man.

4. Because the Box M tests were significant (Asexuality Attitudes: Box M = 86.41, F(21,83329) = 4.00, p < .0001; Attitude Function Inventory: Box M = 24.32, F(10,114017) = 2.39, p < .01), homogeneity of covariances assumptions were violated. However, the Box M test may be detecting normality departures because of large sample sizes as the Box M test is overly sensitive for larger samples. The Pillai’s Trace statistic is reported given that it is the most powerful and robust vis-à-vis assumption violations; the closer Pillai’s Trace is to 1, the stronger the evidence of an effect. Further, when univariate test assumption violations were detected, non-parametric tests were also reported.

5. While it is less than desirable to amalgamate all minority sexual orientations into one group, there were not enough participants in the various sexual orientation categories to allow for a separate analysis. Bisexual/pansexual women were the largest sexual minority within the queer group, constituting two-thirds of the queer group overall and 72% of all queer women. While statistically questionable because of small sample size (i.e. bi/pansexual, n = 72; mostly homosexual, (n = 7); completely homosexual, (n = 10); queer (n = 4); and other, (n = 6)), we explored the differences in asexual ratings as a function of sexual orientation of female participants. There were no multivariate differences between the groups (F(24,364) = 1.17, ns) and the variances were equal (Levene’s test). This suggests the different orientations, for the purposes of these analyses, can be grouped together.

6. This set of analyses supports the idea that ‘mostly heterosexual’ participants should be considered separately from ‘exclusively heterosexual’ participants yet should not be considered the same as ‘queer’ (Vranglova & Savin-Williams, 2012).

7. Griffiths and Pedersen’s (2009) indirect-experiential could arguably be labelled social-expressive as some items involved ‘what people I know have told me’ along with media coverage (i.e. valued or trusted others) about the target groups.

8. The study contained some comparable general feeling thermometers that suggested feelings towards asexuals were the same as feelings towards gays, lesbians, and bisexuals, but significantly less positive than feelings towards heterosexuals.

Additional information

Funding

This work was supported in part by St. Jerome’s University under faculty research grant IIRG430.