Religious and Racial/Ethnic Differences in Flu and COVID-19 Vaccination Intentions

ABSTRACT

Vaccination is an effective public health initiative to reduce severe illness and death due to COVID-19. Vaccine uptake in the United States has been uneven, however. One proposed mechanism to improve vaccine uptake is to engage religious groups to provide health information and encourage vaccination. It remains unclear though if Americans with differing racial/ethnic identities and religious affiliations are equally likely to receive health-related information in religious settings and endorse COVID-19 vaccination. We assessed data from the 2020–2021 National Politics Study, which utilized a national research panel of U.S. adults. Using logistic regression and a treatment effects approach, we found that having heard health-related messages in religious settings and endorsing progressive religious ideology were associated with increased COVID-19 and flu vaccine acceptance. This relationship was stronger among White American worship-goers. These findings suggest that partnerships with religious organizations may be important if additional barriers to vaccination are also addressed.

KEYWORDS:

• COVID-19
• race and ethnicity
• religion
• vaccination

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Our sample weights did not substantially alter our core findings.

2 Although Qualtrics relies upon an opt-in selection process to generate samples, a 2016 Qualtrics sample, collected by Boas and his colleagues (2020), generated the following social demographic variables: age, education, and income, which were quite similar to that of the General Social Survey (GSS) and the American National Election Survey, both national probability samples. The same holds true for our 2020 survey. In 2016, roughly the same proportion of Qualtrics respondents reported a religious affiliation as was the case in the GSS (79 percent) (Boas, Christenson, and Glick 2020). Similarly, in the 2020 NPS, 70 percent of respondents reported a religious affiliation, while 66 percent of respondents reported doing so in the July 2020 Pew Research Center Survey. By weighting their 2016 Qualtrics survey by political partisanship, Boas and colleagues (2020) generated responses to political knowledge questions that approximated the responses of the 2012 American National Election Study. Similarly, weighting our Qualitrics survey by vote choices generated responses to questions about social movements and health care policy attitudes that were not substantially different from the July 2020 Pew Research Center and the 2020 New York Times presidential vote exit poll surveys.

3 See Appendix A for variable question wording.

4 We employed Steensland et al.’s (2000) denominational classification method to determine Evangelical Protestants, Mainline Protestants, and Historically Black Protestants.

5 The probability estimates listed in Tables 3-4 are derived from logit regression analyses that assess the relationship between attending a congregation where individuals hear sermons about health care with vaccination attitudes while controlling religious affiliation, political partisanship, and social-demographic variables. That being said, the estimates for Tables 3–4 are based upon the following formula;

Pr(y=1| $\bar{X}$ , max xk) - Pr(y=1| $\bar{X}$ , min xk), in which Y represents vaccination attitudes and X represents attending a congregation where sermons about health care are heard.

6 Our interaction analyses indicate that the differences in probabilities between attending worship settings where one hears sermons about health care and those who do not attend with willingness to get the flu vaccine is statistically larger among Whites relative to Hispanics. However, our power analyses indicate that these differences are significant at the .7 level of power, far below the .8 level of power that social scientists rely upon to discern a high level of trust in mean differences based upon our sample size for each group and their confidence intervals. In short, although these analyses suggest that the relationship between health care sermons and willingness acquire the flu vaccine is statistically stronger among Whites than Hispanics, this test is not reliable.

7 Despite being statistically larger, however, at .105, the power analysis for these analyses is far below the .8 threshold that would provide us with the confidence to reject the null hypotheses.

8 While not displayed, these treatment effect analyses are available upon request.

9 See Appendix B for Sampling Methodology and Measures.

10 These analyses are available upon request.

11 The formula below expresses the average treatment effect, a counterfactual causal effect, of the association between attending health care sermon congregations and vaccination attitudes by estimating vaccination attitudes based on respondents being randomly assigned to congregations where they hear sermons about health care issues and not being assigned to those types of congregations. In the formula, E = average vaccination attitude of the sample, D_i = Dummy variable of whether the respondent attended a health care sermon congregation, Y_i = vaccination attitude of respondent who attended a health care sermon congregation, and Y_0i = vaccination attitude of respondent who did not attend that type of congregation.

$E[Y_{ii}\left|D_i=1\left]-E\right[Y_{ii}\right|D_i=0\left]=E\right[Y_{1i}-Y_{0i}|D_i=1\left]=E\right[Y_{1i}-Y_{0i}]$

Additional information

Notes on contributors

R. Khari Brown

R. Khari Brown is a sociology professor at Wayne State University and the 2021–2022 president of the Society for the Scientific Study of Religion. He teaches classes and does research on the sociology of religion. His research explores how race impacts the relationship between religion and social-political behaviors and attitudes. He is a co-investigator of National Politics Study, a project funded by the Louisville Institute, the Issachar Fund, and the University of Michigan. The National Politics Study is a bi-annual study that assesses American political attitudes and behaviors and religious life. He also served as a consultant for the Pew Research Center’s 2020–2021 survey on African American religion.

Karael Campbell

Karael Campbell is a fourth-year medical student at Ohio University Heritage College of Osteopathic Medicine pursuing Emergency Medicine. She holds an MS in Biological Sciences from Southern Illinois University. She is passionate about reducing health disparities, increasing health literacy, and promoting health equity. Karael is also a member of the Gold Humanism Honor Society.

Berkeley Franz

Berkeley Franz is a medical sociologist and health services researcher whose research and teaching focus on health disparities, community and population health, and substance use. Dr. Franz is an associate professor of community-based health at the Ohio University Heritage College of Osteopathic Medicine, in Athens, Ohio, and Osteopathic Heritage Foundation Ralph S. Licklider, DO, Endowed Faculty Fellow in Population Health Science.

Ronald E. Brown

Ronald E. Brown is a political science professor at Wayne State University. He teaches classes and does research on race, religion, and politics. His research explores how race impacts the relationship between religion and social-political behaviors and attitudes. He is a co-investigator of the 1993–1994 National Black Politics Study and the 2004–2020 National Politics Studies.