Measurement Matters: Prevalence of COVID-19 Conspiracies
COVID-19 conspiracies spread rapidly but they may not be as common as we think.
Posted Sep 28, 2020
Discussion of COVID-19 conspiracy theories has proliferated in an attempt to keep up with the rapid dissemination of these theories. Indeed, my most recent blog post was about the psychology behind COVID-19 conspiracy theories. Since the outbreak of the pandemic, it has become increasingly apparent that even seemingly innocuous conspiracy theories, such as those falling under the #FilmYourHospital thread, can have large consequences and reach many people. Hence, as a field, it is essential that we strive to examine COVID-19 conspiracy theories, their consequences, and ways in which we might stop their spread.
Nevertheless, we need to approach research on COVID-19 conspiracy theories with the same skepticism and critical-thinking as we do with any other topic. The push to get research on COVID-19 conspiracy theories out there may have its own consequences. We cannot and should not accept any published study on COVID-19 conspiracy theories as fact. Herein, I will review an empirical study of COVID-19 conspiracy theories that not only received considerable attention in the popular press but also in academic circles. I will provide an overview of these academic debates, concluding with some “lessons learned” for future research.
Freeman et al. (2020)
A recent study published in Psychological Medicine examined the prevalence of COVID-19 conspiracy theories in a large (N > 2,000) sample of British community members. They also examined the relations between endorsement of COVID-19 conspiracy theories and a range of psychological variables, such as paranoid thinking and mental well-being. Freeman and colleagues (2020) report several important findings, including that the endorsement of COVID-19 conspiracy theories is positively associated with a tendency to endorse conspiracy theories generally and is robustly linked with paranoid thinking. Moreover, COVID-19 conspiracy theories were associated with decreased compliance regarding COVID-19 policies.
Perhaps most notably, the study found that a considerable minority of the British population endorsed alarming conspiracy theories about COVID-19. For instance, they found that approximately 19 percent of the sample agreed that Jewish individuals were responsible for COVID-19 and approximately 20 percent agreed that Muslim individuals were responsible for COVID-19. Freeman et al. (2020) conclude that “fringe beliefs may now be mainstream” (p. 12). Moreover, a whopping 45 percent of the sample agreed that China developed COVID-19 as a bioweapon to attack the West.
Results from this study garnered international media attention. National Public Radio (NPR), for instance, published an article entitled “1 In 5 Adults In England Think The Coronavirus Is A Hoax.” The Jerusalem Post also published an article along these lines, and it was entitled “One in five English people believe COVID is a Jewish conspiracy.” These reports and results from the Freeman et al. (2020) study paint a gloomy portrait indeed.
In the weeks and months that followed the publication of the Freeman et al. (2020) study, however, academics responded with a healthy dose of skepticism and critical feedback, with two of these responses being published in Psychological Medicine (McManus, D’Ardenne, & Wessely, 2020; Sutton & Douglas, 2020).
McManus et al. (2020) note that the results don’t “quite pass the smell test” (p. 1). They emphasize that the prevalence rates diverged, in some instances dramatically, from those identified in other studies. Although surprising results about human nature can be incredibly important in the field of psychology, we should perhaps be most skeptical of results that are surprising and inconsistent with results from other respected papers.
In addition, when taking a closer look at the methodology Freeman et al. (2020) employed, scholars observed a potentially concerning oversight related to measurement. In the study, participants had the option to disagree (do not agree) or agree (agree a little, agree moderately, agree a lot) with the accuracy of conspiracy statements. This response scale is what scholars call imbalanced – there is one option to disagree with several options to agree. An imbalanced response scale can give rise to an acquiescence bias, meaning that participants tend to agree with the statements due largely in part to the response options. Acquiescence bias can dramatically inflate agreement with survey items. As Sutton and Douglas (2020) note, “participants who disagree with a conspiracy theory but are willing to admit that it might have some merit, may feel that they have no option but to select one of the ‘agree’ responses” (p. 1).
Sutton and Douglas (2020) tested whether acquiescence bias influenced results from the Freeman et al. (2020) paper. They recruited more than 500 participants, again from Britain, to rate their agreement with COVID-19 conspiracy theories about Jewish and Muslim individuals in addition to a Chinese bioweapon. They used three different response scales, one of which was the same used in the Freeman study and the other two were balanced response scales. Using the balanced response scales, the prevalence of endorsement ranged from 1 percent-10 percent, which is in stark contrast to the 19 percent-45 percent estimates identified in the Freeman study. Although it is not possible to conclude that the differences in scales caused these differences in prevalence estimates, it is likely to have played a significant role.
Freeman and colleagues responded to these criticisms in a separate 2020 paper also in Psychological Medicine. They describe that they selected the imbalanced rating scale to “test whether there was any degree of endorsement” (p. 1) of COVID-19 conspiracy theories. They contend that their results may be surprising but this fact does not render them inaccurate.
Still, an accurate result can be a misleading result. Just as COVID-19 conspiracy theories spread like wildfire, so can surprising research results that may, in fact, be overstated. Scientists, reporters, and community members alike should put on their critical-thinking caps when reading information on COVID-19 conspiracy theories and remind themselves of what McManus et al. (2020) described: “misleading evidence can be more damaging than no evidence at all” (p. 1).
Freeman, D., Waite, F., Rosebrock, L., Petit, A., Causier, C., East, A., ... & Bold, E. (2020). Coronavirus conspiracy beliefs, mistrust, and compliance with government guidelines in England. Psychological Medicine, 1-30.
McManus, S., D'Ardenne, J., & Wessely, S. (2020). Covid conspiracies: Misleading evidence can be more damaging than no evidence at all. Psychological Medicine, 21, 1-2.
Sutton, R. M., & Douglas, K. M. (2020). Agreeing to disagree: reports of the popularity of Covid-19 conspiracy theories are greatly exaggerated. Psychological Medicine, 1-3.
Freeman, D., Waite, F., Rosebrock, L., Petit, A., Causier, C., East, A., ... & Bold, E. (2020). We should beware of ignoring uncomfortable possible truths (a reply to McManus et al). Psychological Medicine, 1-1.
Edmunds, D. R., & Tercatin, R. (2020, May 25). One in five English people believe COVID is a Jewish conspiracy. The Jerusalem Post. https://www.jpost.com/diaspora/antisemitism/one-in-five-english-people-believe-covid-is-a-jewish-conspiracy-survey-629187
Langfitt, F. (2020, May 22). 1 In 5 Adults In England Think The Coronavirus Is A Hoax. National Public Radio (NPR). https://www.npr.org/sections/coronavirus-live-updates/2020/05/22/860947708/1-in-5-in-england-think-the-coronavirus-is-a-hoax