The Index of Attitudes toward Homosexuals 30 Years Later: A Psychometric Study

Purpose: A cultural shift in attitudes towards gays and lesbians has occurred over the past several decades, yet negative attitudes toward homosexuality are far from eradicated as evidenced by continuing civil rights struggles. Assessing the degree to which these attitudes persist remains an important task—to improve our understanding of these negative attitudes and to examine factors that may lead to attitudinal change. A need continues for a current, useful, conceptually sound standardized measure. This validation study of the Index of Attitudes toward Homosexuals 30 years after its original development sought to examine its ongoing usefulness for current research. The IAH was designed to measure individual levels of negative attitudes toward homosexuality in terms of affective responses to and cognitions about gays and lesbians.

Methods: The data were collected from students (n=331) enrolled in a human sexuality class at a Southeastern university. Identities could not be linked to responses and participation was not a course requirement. The questionnaire included standardized measures and demographics, and the portion of the data related to investigating the psychometric properties of the IAH were utilized for this analysis.

Results: Participation in each class was over 90%. Data met the assumptions of the analyses, though four of the original 25 items were removed as too skewed and kurtotic. Correlation analysis revealed that all items, save one, were correlated with at least one other item at greater than .30, and none of the items correlated at over .75. All items were retained except the weakly correlating item.

A principal components factor analysis suggested a two-factor solution, theoretically and conceptually consistent with the two anticipated factors—cognitive responses to social distance, and affective responses to attraction or advances. Bartlett's Test of Sphericity and the Kaiser-Meyer-Olkin measure were excellent. Results of an orthogonal principal axis factor (PAF) analysis were consistent with the initial two-factor solution, but it was conceptually more logical that the two factors would be correlated. An oblique PAF revealed an excellent factor structure after two double loaded items were eliminated.

The overall Cronbach alpha coefficient was excellent at .92 for the resulting 19-item measure. The subscale factors were also quite good—Cronbach alpha of .91 for the 14-item cognitive/social distance factor, and .79 for the 5-item affective/attraction-advances factor.

Content validity was established conceptually, and preliminary discriminant validity was established statistically. Evidence of convergent validity was demonstrated by comparing IAH scores with homophobia subscale scores included in another standardized measure on the questionnaire (r=.77).

Implications: This scale demonstrated sound psychometric properties and was consistent with its original conceptualization. The index may be strengthened by reducing the number of items, as suggested in our analyses, and updating the language. Further confirmation of this streamlined measure on a different sample and utilizing a confirmatory factor analysis would be a good next step. In addition to its usefulness for research and intervention development, the IAH can be used to prompt discussion about sexual diversity, values, and beliefs among social work students in preparing them for culturally competent and nondiscriminatory practice.