The Covid-19 pandemic has upended normal life. Concerns abound about the economy, public health, and the chances for a return to normalcy. However, while the impact of the virus affects everybody, not everybody reacts to it in the same way. Our research partnership with Utah Policy and KUTV reveal the extraordinary importance of partisanship in the way in which people respond to the challenges posed by the virus. We find the partisan reaction extraordinary for a number of reasons.
“In the wake of 9/11, Americans rallied together and forged a remarkable consensus on what to do. Similar results followed Hurricanes Katrina and Sandy.”
First, in a nationwide or global crisis, it is not uncommon for citizens -- partisans and independents alike -- to set aside politics and come together in pursuit of a collective good. For example, in the wake of 9/11, Americans rallied together and forged a remarkable consensus on what to do. Similar results followed Hurricanes Katrina and Sandy. Partisanship typically seems inappropriate in response to such events because the tragedy calls for cooperation rather than conflict.
Second, the partisan reaction affects a broad array of institutions, such as trust in the government, politicians, and the scientific establishment. Citizens split along partisan lines when assessing the very institutions needed to create consensus and mobilize resources to address the crisis. Such divisions risk undermining any positive reaction to the crisis and continue to generate opportunities for partisans to contend with each other.
An examination of the data from the March survey indicate that Utah, like the rest of the nation, seems to view the crisis through a partisan lens. The next section explores the nature of this peculiar facet of public opinion in the wake of a pandemic. We then pivot to the ways in which these differences may affect American institutions and the possibility of a diminished role for scientific authority in American society.
Analysis and Graphics
We begin by considering a few obvious hypotheses. It is no secret that Utah has a large share of members of the Church of Jesus Christ of Latter-day Saints. Members have a history of collectively working together to solve problems. It seems reasonable to conclude that members of this organization would respond differently to the crisis than other groups. The data indicate that respondents who are members of the LDS Church, particularly respondents who are active in their faith, are different from other sub-groups in the sample, but not in ways that we expected. While they are more likely to trust the federal government in their response to the coronavirus, they are less trustful of the American news media, and they are more worried about the pandemic’s effects on the economy as opposed to its effects on public health.
However, as we stipulated up above, the pandemic seems to have called forth different ways for citizens to react. For this reason, we control for partisanship in the analysis. In other words, we look at differences between the various subgroups while holding the partisanship of the respondent constant. We find that controlling for partisanship makes the effect of religiosity go away. In terms of opinions on COVID19 as well as other elements of public opinion, the factor of partisanship is the most important factor.
This should not surprise anybody who knows anything about Utah politics. Members of the LDS Church tend to be more conservative and to be more Republican. This analysis indicates that when respondents consider their answers to the questions about the crisis, partisanship plays a more important role than religion. However, the correlation between religion and partisanship covered up the importance of the partisan variable. For this reason, the rest of the analysis will focus on the importance of the partisan variable.
First, levels of worry about COVID19 vary slightly but noticeably along partisan lines. When we asked respondents to rank on an ordinal scale how worried they feel about COVID19, small differences were visible with regards to how worried certain respondents felt as shown in the image below:
Again, while the differences in worry by partisan identity are not large, an overall trend emerges: the proportions of ‘very’ and ‘somewhat’ worried respondents increase as party identification changes from Republican to Democrat.
Worry about COVID19 can be examined at an even more granular level by examining which specific institutions or groups respondents do or do not trust. We asked respondents which institutions they approve or disapprove of with regard to their response to the pandemic. As we thought, partisan differences emerge in the ways in which respondents assess the various institutions.
Of particular interest to us were trust or approval in President Trump and his administration, and healthcare services in the United States. As the next figure shows, the difference in approval in these institutions is particularly pronounced, and the differences are almost due completely to partisanship.
Another institution worth considering is the news media. While the news may not technically be termed an institution, many Americans rely heavily on the institution for information. Perceptions of the news media differ, once again along partisan lines.
Specifically, respondents who responded as being or leaning more Democrat were more likely to report trust in the news media, or were at least less likely to say that they believed the news media had exaggerated the seriousness of COVID19 in the United States. In contrast, Republicans were far more likely to report a belief that the news media had exaggerated the seriousness of COVID19, even though many Republicans were split on the issue.
Lastly, we explored perceptions of the possible effects of COVID19 and how these perceptions may vary by partisanship. We asked our respondents whether they were more worried about the effects of COVID19 on public health or its effects on the economy. Generally, those who identified or leaned Democrat were far more likely to report being worried about the effects of COVID19 on public health, whereas Republicans were more likely to report worry about the effects on the economy.
Discussion
While these partisan differences may seem unremarkable to some, they highlight serious substantive dynamics at the heart of American society.
First, differences in trust in the preparedness and effectiveness of institutions (such as the executive branch or healthcare services) that are informed by partisanship threaten the effectiveness of those institutions. Consider that if a voter decides how they feel about an institution based on their party, then the institution itself, as separate from the political party that operates it, will become politicized. That voter will perceive the strength and effectiveness of that institution in a unidimensional way, thereby constraining the good that the institution can pursue.
Second, imagine for a moment that being more worried about the effects of COVID19 on the economy tracks with an opinion that government social distancing measures should be lifted, and, conversely, that being more worried about the effects of COVID19 on public health is congruent with an opinion that those measures should remain in place. To say nothing of which policy stance is objectively correct (whether America should or should not lift social distancing restrictions), it is sobering to imagine a world where a decision this important could only achieve legitimacy if it satisfied partisan preferences. Such an outcome further restricts the domains in which people can come together and form coalitions for the public good.
Conclusion
In yet another area of American life, partisanship has manifested itself as a critical feature for understanding the distribution of public opinion. Crucially, the power of partisanship to color public debate persists even in the throes of a national and worldwide pandemic. These dynamics, produced by partisan preferences, pose challenges to the credibility of the institutions on which American democracy rests.