Effects of Misinformation on Election Beliefs: Disentangling Motivated Reasoning from Selective Exposure

The proliferation of misinformation, especially during critical periods such as election seasons, challenges democratic processes worldwide. Combining a two-wave panel study with web-browsing behavior (N = 21M) of 1,194 American adults around the 2020 U.S. Presidential Election, we find that those exposed to at least one misinformation website are 17.3% more likely to believe that the election was fraudulent in a post-election survey. However, one challenge in assessing the effects of misinformation is disentangling the behavior of individuals selectively exposing themselves to misinformation congenial to an existing belief versus prior political identities heterogeneously leading to differing effects on beliefs. To disentangle selective exposure from motivated reasoning, we apply a causal machine learning method that controls for the propensity to be exposed to a misinformation website. Using this approach, we quantify the association between exposure to misinformation websites and false election beliefs. Using this method, we find exposure to misinformation is associated with the likelihood of false election beliefs by 4.2%. This effect, however, is asymmetric and consistent with a motivated reasoning account: for conservatives, whose candidate lost the election, the association is 12.6%, while for liberals, the association is negligible at -0.2%. Further, we identify a dosage effect, where each additional exposure to misinformation websites increases the association between misinformation website exposure and false election beliefs among conservatives by .035% but only .010% for liberals. These findings contribute to our understanding of misinformation's real-world effects, demonstrating a significant impact on election validity beliefs that vary by political orientation and exposure frequency. Our results underscore the importance of addressing misinformation not as a uniform influence but as a factor whose effects are contingent upon individual predispositions and exposure volume.