______
Are Religious Adherents More Likely to Buy Into COVID-19 Conspiracy Theories?
DIANA ORCÉS, IAN HUFF, AND NATALIE JACKSON
With the ongoing spread of COVID-19 in the United States and the world, global misinformation about the pandemic is also spreading. Since the start of the pandemic, a number of conspiracy theories have surfaced, including beliefs about the nonexistence of the new virus, its creation at a Chinese lab or as a biological weapon, and blaming 5G networks or Bill Gates for the spread. Unfortunately, belief in conspiracy theories can prevent people from taking appropriate health-related behaviors (Islam et al. 2020; Jolley and Douglas 2014). Previous research has found that willingness to believe in unseen, intentional forces is associated with support for conspiracy theories (Oliver and Wood 2014), a cognitive style of thinking—better known as magical thinking—more likely to be present among certain religious believers (Oliver and Wood 2018; Dyrendal, Roberston, and Asprem 2019). Both religion and conspiracy theory are usually understood as “involving specific patterns of thought and ideas” that are related in intricate ways to social power (Dyrendal, Roberston, and Asprem 2019), making it imperative to evaluate systematically how beliefs of particular religious groups are ideologically in line with conspiracy beliefs.
Are religious adherents more likely to buy into COVID-19 conspiracy theories, and if they are, why? To answer this question, we use the 2020 American Values Survey data collected by the Public Religion Research Institute (PRRI) to investigate the relationship between religious adherence and support for the belief that COVID-19 is lab made, while accounting for key several sociodemographic and political factors.
This chapter hopes to illuminate, through a quantitative approach, whether religious adherence accounts for belief in COVID-19 conspiracy theories alongside political factors. With the current polarized nature of U.S. politics, increasing distrust in government and other institutions, and a growing skepticism of mainstream media outlets, better understanding of who is more likely to believe in COVID-19 conspiracy theories during the pandemic is vital given its health-related implications (Islam et al. 2020).
Why Do People Believe in COVID-19 Conspiracy Theories?
People are drawn to conspiracy theories due to many factors. The literature suggests that partisanship, ideology, distrust in political leaders, scientists, or other authorities, as well as individuals’ predispositions and general views about the world, have an impact on how people think about conspiracy theories (e.g., Douglas and Sutton 2011; Oliver and Wood 2014; Uscinski and Olivella 2017).
President Donald Trump’s administration was an era of conspiracy theories, which complicated the ability to accurately communicate about the spread of the coronavirus (Hellinger 2019). Recent research finds that the belief in COVID-19 conspiracy theories is tied to support for Donald Trump and the rejection of information from experts and other authorities, in addition to conspiracy thinking, party affiliation, and political ideology (Uscinski et al. 2020). Belief in COVID-19 conspiracy theories is also associated with lower levels of education, more negative attitudes toward government responses, and broader beliefs in conspiracy theories (Georgiou, Delfabbro, and Balzan 2020). In fact, Miller (2020a) shows that support for COVID-19 conspiracy theories is a form of monological belief system better understood as a “mutually supportive network of beliefs,” adding evidence to a growing body of research that suggests that if people believe in one conspiracy theory, they tend to believe in other conspiracy theories too (e.g., Goertzel 1994; Lewandowsky, Oberauer, and Gignac 2013; Swami et al. 2011; Wood, Douglas, and Sutton 2012).
There has also been a problem with denial of science related to COVID-19. Lewandowsky, Oberauer, and Gignac (2013) find that believers of conspiracy theories are more likely to reject scientific propositions. The literature on the sources of science denialism argues that individuals who deny science lack scientific knowledge (Landrum and Olshansky 2019). However, science denialism is complex. Beliefs in science depend on individuals’ prior attitudes and values.1 Studies on the relationship between religion and science suggest that individuals see both religion and science as in conflict or that religion exists to answer questions about meaning and morality, whereas science answers questions about how the natural world works, or that both religion and science completement, influence, and guide each other (Ecklund 2010). In essence, most religious belief systems encapsulate a set of beliefs about perceptions of how a society works and a set of values of how society should work. In this study, we explore how religion is associated with belief in COVID-19 conspiracy theories, especially as religion has come to the forefront of American politics (Jones 2020).
The Role of Religion in the Belief of Conspiracy Theories
Religion and conspiracy theory are usually understood as “involving specific patterns of thought and ideas” that are related in intricate ways to social power (Dyrendal, Roberston, and Asprem 2019). In this section, we specifically seek to understand how conspiracy beliefs are connected to the beliefs of religious adherents.
Leveraging previous research, Oliver and Wood (2014) argue that belief in conspiracy theories is the function of two psychological predispositions. The first, “magical thinking,” relates to “an unconscious cognitive bias to draw causal connections between seemingly related phenomena.” This way of thinking is associated with religious beliefs because they project feelings of control when in situations of uncertainty. Beliefs in conspiracy theories are positively related to intuitive rather than analytic thinking (Swami et al. 2014; Oliver and Wood 2018), and religion reinforces intuitionist thinking through beliefs of magical thinking understood as “thinking that relies on omnipotent fantasy to create a psychic reality that the individual experiences as ‘more real’ than external reality” (Ogden 2010, 318). Religious concepts present in most religions, such as miracles, healing powers, and supernatural manifestations, are easily remembered because they are uncommon and challenge the way things happen typically. For example, magical thinking is present among evangelical Christians who believe in the Prosperity Gospel, a belief system that suggests that individuals who are faithful to God and God’s church will be blessed with health and wealth (Burge 2017).
Belief in religious concepts often helps to make sense of nonsensical occurrences or offers comfort from a perceived higher power. In the same way, conspiracy theories are easily remembered because they involve concepts that are uncommon but can offer explanations for what might seem inexplicable (Andrade 2020).2 Both religion and conspiracy theories provide structure to uncertainty due to their reliance on magical thinking. In this sense, we expect the following hypotheses:
H1: Religious adherents will be more likely to believe in COVID-19 conspiracy theories than religiously unaffiliated individuals.
H2: Individuals who exhibit magical thinking will be more likely to believe in COVID-19 conspiracy theories than those who do not exhibit magical thinking.
Oliver and Wood’s (2014) second predisposition to conspiracy theories relates to Manichean thinking, specifically the belief in a clash between good and evil, which is common in religious rhetoric and “comports with how some people process political information” and how they explain ambiguous events. Most religions subscribe to the idea that good things happen to good people and bad things to bad people (Furnham 2003), thus making them more receptive to messaging that benefits “good” people and punishes “bad” people. For example, evangelical Christians who believe in the Prosperity Gospel may think that those who are faithful will be blessed with health and wealth, while those who are unfaithful will be punished.
Dualistic-Manichean views in which human beings are believed to be composed of a body and a soul are closely related to religious dogmatic views (Del Rio and White 2012), and dogmatism is associated with lower critical reasoning skills. While confirmation bias is a universal human trait (Nickerson 1998), we suspect that religious individuals marked by Manichean worldviews tend to cling to specific beliefs that resonate with their moral views and affirm their thinking at greater rates. In short, high levels of dogmatism make individuals less likely to look at issues from others’ perspectives and more receptive to messaging that strengthens their own views (Friedman and Jack 2018). Conspiracy theories also typically utilize Manichean thinking by creating an enemy—someone or something is doing something bad—that good people must contend with. Thus, we expect the following hypothesis:
H3: Individuals who exhibit Manichean thinking will be more likely to believe in COVID-19 conspiracy theories than those who do not exhibit Manichean thinking.
Moreover, when experiencing anxiety-provoking events, individuals tend to resort more to magical or Manichean thinking as a way to control their surroundings, and religious thinking also operates this way, finding explanations for incomprehensible phenomena (Malinoswki 1992). In fact, conspiracy theories become especially attractive under societal crises, such as natural disasters, because they help reduce anxiety, uncertainty, or feelings of lack of control (Van Prooijen and Douglas 2017; Abalakina-Paap et al. 1999; Green and Douglas 2018). Thus, during the pandemic, we expect to see the following hypothesis corroborated:
H4: The effects of both magical and Manichean thinking on the belief in COVID-19 conspiracy theories will be stronger among religious adherents than religiously unaffiliated individuals.
COVID-19 in the United States
The case of the United States and the Trump administration serve as a good test case for an analysis of the relationship between religious adherents and the belief in COVID-19 conspiracy theories. The Trump administration’s response to the first year of the pandemic was incoherent and left plenty of space for conspiracy theories to take hold. Trump made his first public comment on January 22, 2020, downplaying the impact of the virus and suggesting that the United States had it under control. On January 30, 2020, the World Health Organization (WHO) declared the coronavirus a global health emergency, and a day later Trump prohibited entry to the United States for select individuals coming from China (Peters 2020). Recent studies suggest the coronavirus was already in the United States at that point and came from Europe (Zimmer 2020).
On February 25, 2020, Trump appointed then vice president Mike Pence to lead the coronavirus task force and a few days later suggested that the virus would miraculously disappear. On March 19, Trump alleged that the Food and Drug Administration (FDA) approved the drug hydroxychloroquine for treating COVID-19 and continued to support its use as a COVID-19 treatment after it was demonstrated to be less helpful than initially thought (Gittleson, Phelps, and Cathey 2020). On April 30, Trump said that he had reason to believe that the virus had originated in a laboratory in Wuhan, China (Mangan and Lovelace 2020). Trump himself was diagnosed with COVID-19 on October 2, and despite spending three days in the hospital, he continued downplaying the threat of the virus and its spread, even though COVID-19 cases were increasing everywhere and worsened throughout the fall of 2020 (Burns 2020).
Misinformation appears to influence people’s response to the virus (Islam et al. 2020). Many religious groups, especially evangelical Christians, remain firmly supportive of and receptive to Trump’s message (Strang 2020). When party leaders and other government officials misinform the public and promote conspiracy theories, likeminded individuals who perceive these leaders as credible and are exposed to this type of rhetoric are more likely to accept these ideas (Swire et al. 2017).
Data and Method
To test our hypotheses, we use the 2020 American Values Survey by PRRI, a random sample of 2,538 adults (age eighteen and up) living in the United States, including all fifty states and the District of Columbia. Interviews were mostly conducted online using a self-administered design in both Spanish and English between September 9 and September 22, 2020.3
This survey provides a few items that allow for the systematic assessment of the relation between religion and a belief in conspiracy theories. The survey includes one COVID-19 conspiracy theory question applied to half of the sample (1,285 cases) and serves as our dependent variable: “Which of these two statements do you think is most likely to be true? (1) The coronavirus was developed intentionally by scientists in a lab (2) The coronavirus developed naturally.”4 Americans are evenly divided on this question (50 percent vs. 49 percent, respectively). Because of the binary nature of our dependent variable, we use logit regression models and control for political variables as well as personal demographic and socioeconomic characteristics.5
Our key independent variable is religious affiliation. PRRI uses an identity measure to distinguish among Protestants, asking if they are born-again or evangelical Protestants or not. This is not the same as classifying their denominational affiliation, but analyses using either measure tend to return quite similar results (Burge and Lewis 2018). PRRI data show that the largest shares of religiously affiliated Americans are Christians, with 22 percent evangelical Protestants (13 percent white, 9 percent of color), 25 percent nonevangelical Protestants (17 percent white, 8 percent of color), 17 percent Catholics (10 percent white, 6 percent Hispanic, 1 percent all other of color), and 1 percent other Christians, including Latter-day Saints and Orthodox Christians. The rest of religiously affiliated Americans belong to non-Christian groups, which make up 9 percent of all religious groups, including Jewish, Muslim, Buddhist, and Hindu Americans as well as Americans who identify with other religions. One in four (25 percent) Americans identify as religiously unaffiliated. The survey also asks for religious service attendance. About one in four (27 percent) Americans reported they attend religious services either at least once a week or more, 21 percent indicated once or twice a month or a few times a year, and 52 percent said seldom or never.
Luckily, the survey includes a measure of magical thinking, asking if Americans completely agree, mostly agree, mostly disagree, or completely disagree with the following statement: “God always rewards those who have faith with good health and will protect them from being infected by the coronavirus.” This measure allows us to capture one of the ways magical thinking is associated with religious beliefs, that of divine intervention and protection, which projects feelings of control when in an uncontrollable situation—the pandemic (Ogden 2010).6 The belief that God will protect individuals from the coronavirus (magical thinking) could help them make sense of illogical occurrences, such as the possibility that a virus was manufactured in a lab. We also incorporate a proxy of Manichean thinking: “It is necessary to believe in God in order to be moral and have good values.” Manichean thinking is closely related to dogmatic thinking, particularly regarding what is good and what is not, and this measure captures the belief in a dogmatic view of requiring a specific belief to qualify as good. Those who agree with this statement will suggest black-and-white thinking. The more moral correctness religious individuals see in something, the more likely this affirms their thinking of God rewarding those who have faith and punishing those who do not (Friedman and Jack 2018). We acknowledge the limits of both measures of magical and Manichean thinking as they capture part of multidimensional concepts. About one in four (23 percent) Americans either completely or somewhat agree that God always rewards those who have faith and will protect them from the virus, and 39 percent completely or somewhat agree that it is necessary to believe in God in order to be moral and have good values.
Figure 2.1 Predicted Probability of Believing COVID-19 by Magical Thinking and by Manichean Thinking. Note: Results based on model 4 in the online appendix. Differences in mean predicted probabilities for magical thinking between completely agree and completely disagree as well as for Manichean thinking between all categories and completely disagree are statistically significant at p < 0.05. (Source: 2020 American Values Survey.)
In addition, we add variables to assess how support for Trump impacts conspiracy beliefs. The survey asks, “Do you strongly approve, somewhat approve, somewhat disapprove, or strongly disapprove of the job Donald Trump is doing as president?” Since most of Trump’s COVID-19 messaging to his supporters has been through Fox News, we add another variable that captures Americans who trust Fox News the most among television news sources to provide accurate information about politics and current events.
We also include an index of mistrust based on “not at all” responses to how much Americans trust various government officials and institutions to provide accurate information and advice regarding the ongoing coronavirus pandemic, including university research centers, Dr. Anthony Fauci, and then senator Joe Biden. We exclude trust in Donald Trump as this variable is highly correlated with Trump’s job approval (r = 0.80).
We control for political ideology, partisanship, racial/ethnic identity, education, income, gender, age, and region. Following the literature on the sources of belief in conspiracy theories, we expect that those with lower levels of education and income will be more likely to believe that the virus was developed in a lab. Less educated individuals tend to attribute agency and intentionality where it is not present (Van Prooijen 2016).
Results
The first step in the analysis confirms (H1) that religious adherents are more likely to believe that COVID-19 is lab made than those who are religiously unaffiliated.7 With respect to our second hypothesis (H2), we find support for the idea that individuals who exhibit magical thinking will be more likely to believe that COVID-19 is lab made (models 2 and 4 in Table A2.1 in the appendix). We find the degree of agreement with the statement “God always rewards those who have faith with good health and will protect them from being infected by the coronavirus” increases the mean predicted probability of believing the virus is lab made by 22 percentage points as one moves from completely disagreeing (49 percent) to completely agreeing (71 percent) with this statement (Figure 2.1).
Moving on to our third hypothesis (H3), we also find evidence that individuals who exhibit Manichean thinking are more likely to believe that COVID-19 is lab made (model 3 and 4 in Table A2.1 in the appendix). The mean predicted probability increases by 29 percentage points as one moves from completely disagreeing (38 percent) to completely agreeing (66 percent) with the statement “It is necessary to believe in God in order to be moral and have good values” (Figure 2.1).
Our last step in the analysis partially confirms our fourth hypothesis (H4). We find evidence in support of the joint impact between magical thinking and religious affiliation on the belief that COVID-19 is lab made, even after controlling for everything else.
Figure 2.2 (model 5 in Table A2.1 in the appendix) shows that among evangelical Protestants, the mean predicted probability of believing COVID-19 is lab made increases by 43 percentage points (p < 0.05), moving from completely disagreeing (51 percent) to completely agreeing (94 percent) with the statement “God always rewards those who have faith with good health and will protect them from being infected by the coronavirus.” Because of the growing political significance of evangelical Protestants under the Trump administration (Jones 2020), these results add additional evidence to the corrosive impact of misinformation.
Further, while there are no significant differences in COVID-19 beliefs between nonevangelical Protestants who completely agree and completely disagree with the magical thinking statement, we do find that the mean predicted probability of believing COVID-19 is lab made among nonevangelical Protestants increases by 50 percentage points (p < 0.05) moving from somewhat disagreeing (35 percent) to completely agreeing (85 percent) with the statement (Figure 2.2).
Figure 2.2 Predicted Probability of Believing COVID-19 Is Lab Made by Religious Tradition and Magical Thinking. Note: Results based on model 5 in the online appendix. Differences in mean predicted probabilities for magical thinking among evangelical Protestants between completely agree and completely disagree as well as among non-evangelical Protestants between completely agree and completely disagree are statistically significant at p < 0.05. (Source: 2020 American Values Survey.)
Because the goal of this chapter is to also evaluate whether the effect of magical thinking on the belief in COVID-19 conspiracy theories is stronger among religious adherents than nonreligious adherents, we show graphically this relationship among religious unaffiliated Americans, who do not differ significantly between those who completely agree and those who completely disagree with the statement (Figure 2.2). The same is true among Catholics and members of other religious groups (not shown graphically).
In addition, when examining the joint impact between Manichean thinking and religion on COVID-19 beliefs (Figure 2.3), we only find significant results among Catholics: the mean predicted probability of believing COVID-19 is lab made increases by 54 percentage points among Catholics moving from somewhat disagreeing (25 percent) to completely agreeing (79 percent) with the statement “It is necessary to believe in God in order to be moral and have good values.” Even though there is some variation in Manichean thinking among unaffiliated Americans, these differences are not statistically significant. These results suggest that Catholics may be particularly attuned to Manichean beliefs compared to other religious groups. Manichean thinking could be one mechanism that helps explain Catholics’ COVID-19 conspiracy beliefs.
Figure 2.3 Predicted Probability of Believing COVID-19 Is Lab Made by Religious Tradition and Manichean Thinking. Note: Results based on model 5 in the online appendix. Differences in mean predicted probabilities for Manichean thinking among Catholics between completely agree and somewhat disagree are statistically significant at p < 0.05. (Source: 2020 American Values Survey.)
Other findings from our logit regression models show that those who identify as conservative, mistrust political officials and institutions, and those who approve of President Trump and watch Fox News are more likely to believe that COVID-19 is lab made.
An intriguing and unexpected finding in our models is how attending religious services seems to undermine belief in the COVID-19 conspiracy theory. We find evidence that Americans who attend religious gatherings at least once a week, compared to those who seldom or never attend religious gatherings, are less likely to believe that COVID-19 is lab made, contrary to our expectations. We expected that individuals who engage more with their faith groups would be more inclined to exhibit similar thinking, as likeminded individuals have been found to be exposed to specific rhetoric and more likely to accept these ideas (Swire et al. 2017). However, this could also be an artifact of the regression due to the fact that we are already accounting for religious identification. If the key theoretical linkage between conspiracy theories and religion is the magical and Manichean nature of their beliefs, perhaps more frequent attendance at religious services is simply not an indicator of those particular aspects of religion.
Finally, age, racial identification, and income matter, adding support to previous research suggesting that stigmatized minority groups believe more strongly in conspiracy theories than majority group members (Goertzel 1994; Thorburn and Bogart 2007).
Conclusion
The United States was ripe for conspiracy theories to take hold when the pandemic hit. There was no coherent, unified national message, and the former president—who could have provided that message—instead often repeated misinformation and conspiracy theories regarding the novel coronavirus. But this is not the complete story.
The other unique aspect of the United States is its religiosity. Despite a decrease in religious affiliation among Americans, Americans still remain among the most religious people in the industrialized world. More than half of Americans (55 percent) say they pray daily, compared to 25 percent of Canadians, 18 percent of Australians, and 6 percent of the British (Fahmy 2018). That matters for conspiracy theories because religious beliefs and conspiracy beliefs share two common features: magical thinking, the willingness to believe stories in order to explain or add stability to uncertain situations, and Manichean thinking, the need to sort the world into forces of good and evil. Both are associated with believing that the virus causing COVID-19 was intentionally manufactured in a laboratory.
In addition to the direct linkage between religiously based magical and Manichean thinking and the COVID-19 conspiracy theory, we have shown in these analyses that religious affiliation itself is directly associated with stronger belief in the conspiracy theory. For evangelical and nonevangelical Protestants, there is an interactive effect as well, which shows that stronger magical thinking combined with their religious affiliation makes them more likely to believe in the conspiracy theory.
Layering all of these pieces together—the religiosity of the U.S. population, the connection between religious thinking and conspiracy thinking, and the environment in which U.S. leadership espoused conspiracy theories—created a perfect storm for the U.S. public to believe conspiracy theories. In this light, it becomes less surprising that 50 percent of the country would believe a conspiracy theory that says the virus that causes COVID-19 was intentionally manufactured in a laboratory. And, since believing in COVID-19 conspiracy theories could influence health behaviors and possibly have detrimental health outcomes (Jolley and Douglas 2014), in addition to widespread beliefs that God would protect the faithful, it is likely these mechanisms of thinking have contributed to the United States coping quite badly with the pandemic overall.
____________
Material referencing an appendix in this chapter can be found online available here: https://dataverse.harvard.edu/dataverse/epidemic_among_my_people.