Abstract
This study used a correlational design with a sample of university students to clarify the relationships between attitude functions and homonegativity with respect to gender. Classic work on attitude functions posits that attitudes serve psychological needs for the attitude holder. Herek (1986b) adapted this theory to explain attitudes toward homosexuality. Herek (1987) identified four functions: ego-defensive (defense of threats to the self), value-expressive (expression of key values), social-expressive (expression of important social norms), and experiential (based on past experiences). Results suggested that men were more likely to attribute their attitudes to the ego-defensive function. Men and women were equally likely to attribute their attitudes to the experiential function. The ego-defensive function was the best predictor of homonegativity for men and women, whether they held generally positive or generally negative attitudes toward homosexuality. The experiential function did not predict homonegativity. Participants tended to be neither very homonegative nor very ego-defensive.
G. J. Meaney is now at Department of Health Studies and Gerontology, University of Waterloo.
A version of this article was presented as a poster at the 28th Annual Guelph Sexuality Conference, June 2006. This research was supported in part by a grant from the Dean's Research Committee at St. Jerome's University. The authors would like to acknowledge Krystal Kellington for help with data collection for this study.
Notes
1. We are unsure as to whether Cronbach's alpha is an appropriate measure of reliability for two-item scales. Therefore, we also report the inter-item correlation for each subscale of the AFI. Cronbach's alpha is given primarily for comparison with other studies.
2. A midpoint split was chosen because we wanted to compare people with positive attitudes and people with negative attitudes. The midpoint of the scale provided an absolute cut-off point. However, this resulted in groups with vastly different ns. Alternatively, we could have split on the median, which would create groups of similar ns, but a more relative definition of “positive” and “negative.” For lack of better terminology, we wanted to examine true homophobes and true homophiles.