Identifying important individual- and country-level predictors of conspiracy theorizing: A machine learning analysis

Psychological research on the predictors of conspiracy theorizing—explaining important social and political events or circumstances as secret plots by malevolent groups—has flourished in recent years. However, research has typically examined only a small number of predictors in one, or a small number of, national contexts. Such approaches make it difficult to examine the relative importance of predictors, and risk overlooking some potentially relevant variables altogether. To overcome this limitation, the present study used machine learning to rank-order the importance of 115 individual- and country-level variables in predicting conspiracy theorizing. Data were collected from 56,072 respondents across 28 countries during the early weeks of the COVID-19 pandemic. Echoing previous findings, important predictors at the individual level included societal discontent, paranoia, and personal struggle. Contrary to prior research, important country-level predictors included indicators of political stability and effective government COVID response, which suggests that conspiracy theorizing may thrive in relatively well-functioning democracies.

Duke Scholars

Author Claudia Fernandes Nisa DKU Faculty

Author Gavan J. Fitzsimons Fuqua School of Business