Objectives: The present study examined the measurement invariance of the Symbolic Racism Scale (SRS) and the Modern Sexism Scale (MSS) across racial/ethnic and gender groups. Previous psychometric evaluations of the SRS and MSS scores have not examined the equivalence across racial/ethnic and gender groups or have been otherwise statistically inadequate. Therefore, this study sought to fill this gap. Method: To establish measurement equivalence across racial/ethnic (Black, Latinx, and white) and gender (women and men) groups, we conducted a measurement invariance analysis of the SRS and the MSS in a large, diverse sample (N = 719). Results: We found that the SRS and MSS were invariant across gender, and the SRS was invariant across racial/ethnic groups. However, the MSS was noninvariant across racial/ethnic groups. Partial invariance testing revealed nonequivalent factor loadings between Black and Latinx participants compared to white participants on an item of the MSS that referenced "unwarranted" attention that women receive from the government and media. Conclusions: Researchers should consider reevaluating the item that reads: "Over the past few years, the government and news media have been showing more concern about the treatment of women than is warranted by women's actual experiences." Future research is needed to assess how the item is interpreted by Black and Latinx people so it can be modified for use in these communities. Our findings underscore the importance of assessing the validity of the scores in commonly used scales across diverse groups.
Public Significance StatementThe Symbolic Racism Scale (SRS) and the Modern Sexism Scale (MSS) are measures used to assess covert prejudice toward Black people and women, respectively. These scales have been used to understand groups' attitudes toward social movements, social issues, and political candidates and have previously been found to predict attitudes toward race-and gender-related policies. However, to compare group scores on psychological measures, researchers must know if the measured construct has the same structure or meaning across groups. We examined this question in the SRS and MSS by assessing the invariance of these measures across diverse gender and racial/ethnic groups. Measurement invariance indicates that the same construct was measured in each group, while noninvariance suggests that there may be measurement error for at least one of the groups we examined. We found that the MSS is noninvariant between racial/ethnic groups; specifically, an item regarding special favors for women may not be a reliable indicator of modern sexist beliefs among Black and Latinx individuals. Our findings highlight the need for more work on how modern sexism manifests within diverse communities.