Developing and validating the ambivalent prejudice towards gay men scale (APGS): Initial evidence of hostile and benevolent prejudice towards gay men in the United Kingdom

This paper describes the development and initial validation of the ambivalent prejudice towards gay men scale (APGS) – a multidimensional 19-item measure of hostile and benevolent prejudice towards gay men – in two British samples. In Study 1 (N = 801), exploratory factor analysis produced a four-factor (repellent, romanticised, paternalistic, and adversarial) scale, with men scoring higher in repellent and adversarial prejudice and women scoring higher on romanticised prejudice, supporting known-groups validity. The APGS further evidenced convergent and discriminant validity relative to hostile and benevolent prejudice measures and ideologies associated with prejudice, and good internal and test-retest reliability. Romanticised (benevolent) and adversarial (hostile) subscales were positively correlated, and used to create an individual differences index of ambivalent prejudice towards gay men associated with system justifying beliefs and insensitive to self-presentational concerns. In Study 2 (N = 300), confirmatory factor analysis showed that the four-factor model demonstrated superior fit compared to alternative one-factor (generalised ambivalence) and two-factor (hostility and benevolence) models after minor re-specifications to parameters. The four-factor model demonstrated full scalar invariance between men and women and between data collected in 2016 and 2023, suggesting superior psychometric properties compared to currently available alternatives. Implications and directions for prejudice, intergroup relations, and scale validation research are presented.