Factor Structure and Validity of the Lesbian, Gay, and Bisexual Knowledge and Attitude Scale for Heterosexuals (LGB-KASH)
Abstract
Heterosexual attitudes toward Lesbian, Gay, and Bisexual (LGB) individuals have become the focus of recent research as sexual minorities and allies have advocated for LGB human rights issues and have subsequently captured the attention of the media (Rimmerman, 2001, 2008; Rimmerman, Wald, & Wilcox, 2000). Multiple influences shape heterosexual attitudes including gender socialization, individual sexual identity exploration, religious beliefs, and systemic prejudicial attitudes (Altemeyer & Hunsberger, 2005; Kilanski, 2003; Worthington, Savoy, Dillon, & Vernaglia, 2002; Worthington, Becker-Schutte, & Dillon, 2005). Worthington, Dillon, and Becker-Schutte (2005) and colleagues proposed that heterosexual attitudes toward sexual minorities are one aspect of the individual’s sexual identity that is comprised of several dimensions. Worthington et al. developed an instrument titled the Lesbian, Gay, Bisexual Knowledge and Attitude Scale for Heterosexuals (LGB-KASH) to assess the proposed dimensions of heterosexual attitudes toward LGB individuals. Results of confirmatory factor analyses with primarily white college students and adults in the Midwest identified five factors that were consistent with the proposed dimensions. These factors were labeled: hate (violent homonegativity; avoidance of LGB people); LGB knowledge; attitudes toward LGB civil right issues; religious conflict (ambivalent and negative attitudes caused by religious beliefs); and internalized affirmativeness (degree of comfort of having friends who are identified as LGB; feeling comfortable of having feelings of attraction towards the same-sex). The LGB-KASH’s five dimensions correlated in the expected direction with scales assessing traditional homonegativity and religiosity. No other study was located that examined the factor structure and the validity of the LGB-KASH subscales. The purposes of the proposed study was to examine (a) the factor structure of the LGB-KASH with an ethnically diverse college sample, and (b) the relation of the LGB-KASH dimensions to scales assessing modern-homonegativity and religious fundamentalism. It was hypothesized that the LGB-KASH five-factor structure would be confirmed with ethnically diverse college students. It was expected that modern-homonegativity would correlate positively with the hate and religious conflict subscales, and correlate negatively with the LGB knowledge, LGB civil right and internalized affirmativeness subscales. It was expected that religious fundamentalism would correlate positively with the hate and religious conflict subscales, and correlate negatively with LGB knowledge, LGB civil right and internalized affirmativeness subscales. Spirituality experiences of participants were also assessed expecting to find that experiences of spirituality would be unrelated to LGB-KASH subscales. This study surveyed 701 heterosexual identified volunteer participants. Participants represented several major ethnic groups including African-Americans, Latino/a, Asian-Americans, and European Whites. The instruments that were used in the study include: a demographic questionnaire, the LGB-KASH, the Modern Homonegativity Scale (MHS; Morrison & Morrison, 2002), the Religious Fundamentalism Scale (RFS; Altemeyer & Hunsberger, 1992), and FACIT-Spirituality Scale (Peterman, Fitchett, Brady, Hernandez, & Cella, 2002). A confirmatory factor analysis was conducted with an oblique rotation using AMOS 17.0 to examine the factor structure of the LGB-KASH. Several indexes of fit were computed to assess how well the model fit the data including the chi-square, goodness-of-fit index (GFI), adjusted goodness-of-fit index (AGFI), comparative-fit index (CFI), root-mean-square residual (RMR), incremental fit index (IFI), parsimony comparative fit index (PCFI), and root-mean-square error of approximation (RMSEA). The CFA results indicated that the five factor-oblique model had a mediocre fit, and a comparable fit to the results found by Worthington and colleagues. Seven items with poor fit were identified and deleted from the scale in order to re-specify the five factor model. A CFA was conducted on the revised 21-item scale and results indicated that the model had a good fit. Convergent validity was evidenced as the LGB-KASH subscales scores were correlated in the expected direction to the measures of modern-homonegativity and religious fundamentalism; additionally, LGB-KASH subscales religious conflict and internalized affirmativeness were related to the construct of spiritual well-being. Limitations and implications for future research are discussed.