Our analysis focuses on two questions within the field of racial and ethnic politics.
- Are Americans who believe that their race is important to their identity more likely to be interested in politics than those that are not?
- How do personal experiences with racism, in combination with demographic factors such as race, gender, and education, influence the importance respondents place on policies aimed at addressing racial and ethnic discrimination, controlling for other social and political factors? To investigate these two research questions, we used survey data from the 2020 Collaborative Multi-Racial Post-Election Survey (CMPS).
To model the outcome variable, interest in participating in politics, in the first research question, we will use an ordinal logistic regression with predictors including the importance of belonging to a racial group to one’s identity, race, age, educational attainment, and political party affiliation. The outcome variable has 4 ordered levels: “Not at all interested in politics”, “Not that interested in politics”, “Somewhat interested”, and “Very interested in politics”. We will also include an interaction term between gender and the importance of belonging to a racial group to one’s identity to assess whether the effect of the importance of belonging to a racial group on one’s identity differs based on one’s gender identity. To assess the model we will use a confusion matrix to compare predicted and observed values, and compare the model’s accuracy and no information rate. Finally, we will conduct a Brant test to provide additional evidence regarding whether the proportional odds assumption was violated.
For the second research question, which examines the importance placed on anti-discrimination policies, we will use a binomial logistic regression. The predictors include race, gender identity, sexual orientation, age, education, community type (i.e. rurality), experiences with racism, and perceptions of racism affecting different groups. To further explore intersectionality theory, we will develop a second model that includes an interaction term between gender and race. After identifying the better-performing model, we will evaluate its performance using a range of metrics, including the confusion matrix, accuracy, and other relevant indicators.
In this analysis, we investigated whether valuing belonging in a racial group can influence a respondent’s interest in politics. We found that the higher levels of importance an individual placed in belonging to their specific racial group (reporting that belonging to their group was “Very important” or “Extremely important”) had a significant positive effect on their interest in politics when compared to male respondents who did not place a high level of importance on belonging to their specific racial group. Further research is needed to understand how an individual’s racial belonging can impact their political participation. Following the resource model of participation, age and increases in educational attainment appeared to be associated with higher odds of reporting a higher level of interest in politics. Individuals who graduated from college had much higher odds of reporting a higher level of interest in politics than individuals with no college experience. This follows findings that educational attainment can lead to increases in political participation [8], and that older Americans vote at higher rates than younger Americans [9]. For our second question, we analyzed how personal experiences with racism influence respondents’ prioritization of policies addressing racial and ethnic discrimination. Our findings highlight the significant impact of race, gender, and lived experiences with discrimination, as well as the critical role of perceptions of racism against various groups in shaping public attitudes. These results underscore the complex social dynamics that drive policy preferences and provide valuable insights for designing targeted advocacy and policy efforts to address the unique experiences and needs of marginalized communities