Logically incompatible conspiracy beliefs may show a high correlation in subjects

Imhoff, Roland, and Pia Lamberty. 2020. “A Bioweapon or a Hoax? The Link Between Distinct Conspiracy Beliefs About the Coronavirus Disease (COVID-19) Outbreak and Pandemic Behavior.” PsyArXiv. April 14. doi:10.31234/osf.io/ye3ma

Abstract: Exploring the association of distinct conspiracy beliefs (COVID-19 is a hoax; SARS-Cov-2 was human-made) with pandemic-related behavior.

General Discussion

We observed similarly problematic correlates of two distinct conspiracy beliefs concerning the coronavirus pandemic. Depending on whether COVID-19 was believed to be a hoax or the SARS-Cov-2 human-made participants indicated less compliance with selfreported infection-reducing containment-related behavior and more engagement in selfreported self-centered prepping behavior targeting not a reduction of the infection rate but personal benefits in the crisis. Although these associations seem relatively straight-forward, it is important to note that previous research has pointed to the danger of conspiracy beliefs but has dedicated less attention to potentially distinct relations of different kinds of conspiracy beliefs. These distinct associations notwithstanding, our results also provide strong support for the general notion that even logically incompatible conspiracy beliefs show a high correlation and are both positively associated with a general mindset of conspiracy mentality.

Adding to the robustness of the findings, another study conducted within the German context (reported in the Supplement) closely replicated this general pattern. This seems noteworthy as another dataset from the German context failed to find strong relations between conspiracy endorsement and hygiene measures (Pummerer & Sassenberg, 2020).


Limitations and further research

As arguably the most important limitation of our research, all findings are crosssectional correlations and thus mute with regard to causality. Although it seems plausible that people adapt their behavior according to how they see and perceive the world, it is also conceivable that people behave in a certain way (for no or other reasons) and adapt their worldview as a justification after the fact. Another clear limitation is that these studies were conducted in a time of rapidly changing world events and thus might not have undergone the amount of planning and detailed pre-registration as generally desirable for any kind of research question (Scheel, 2020). Applying a stricter alpha level (e.g., p = .005; for a discussion see Benjamin et al., 2018) would yield the negative impact of hoax belief on selfreported containment-related behavior in the UK non-significant after including all control variables. Trusting in the readers’ intuition to interpret the results we refrain here from forcing a binary significant- non-significant decision on these data but merely point to the substantially weaker data pattern in the UK compared to the US.