Misinformation (and Knowledge and Other Belief States): The Case of France

• In Event: Misunderstanding and Misinformation In and About French Politics

Abstract

Political knowledge has been much studied in the U.S. and, to a lesser degree, abroad. Political misinformation is now receiving overdue attention, mainly, so far in the U.S. But these figure to be important variables elsewhere as well. What with populist parties, social media, disinformation campaigns, and, it seems, a waning or less pervasive cultural attachment to truth, misinformation appears to be on the rise in many countries, not just the U.S.
Here we consider another Western democracy, France, using data from the 2017 French Election Study (FNES). In the 2017 presidential campaign, many French observers saw a good deal of misinformation as being disseminated and held. Among other things, we consider how frequently misinformation is held, by whom, and with what attitudinal and electoral consequences.
Drawing on a more methodologically focused companion piece, we consider the differences between knowledge, mere correct belief, ignorance, mere incorrect belief, and misinformation (matters of the presence or absence of relevant cognition, the confidence with which that cognition is held, and its correctness or incorrectness) and sketch some instrumentation resting on questions asking how probably true or false the respondent considers given propositions to be, on a scale from 0 (definitely false) to 10 (definitely true). The responses can be fashioned into measures capable of capturing knowledge, mere correct belief, ignorance, mere incorrect belief, and misinformation more accurately than measures based on conventional multiple-choice knowledge/misinformation items can do.
The FNES contains measures of this sort. They permit us, first of all, to address methodological questions. Measures similar to those we employ have seemed to perform well on non-probability samples in the U.S., but their doing similarly well (as we expect) on the FNES’s national probability sample will buttress the case for their validity.
Substantively, we examine the prevalence of misinformation (and knowledge, mere correct belief, ignorance, and mere incorrect belief). We argue that actual misinformation (like actual knowledge) is much scarcer than conventional measures would suggest. But does mere incorrect belief (like mere correct belief in relation to knowledge) have much the same effect? Does merely suspecting or being inclined to believe a falsehood have the same effect as “knowing” it? More broadly, we compare the effects of knowledge to those of misinformation, of mere correct belief to those of knowledge and mere incorrect belief to those of misinformation, and of mere correct or incorrect belief to those of ignorance.
We also consider the relationship between misinformation and knowledge. With respect to any given proposition, misinformation and knowledge are mutually exclusive (though not exhaustive). But are people who are generally more knowledge better able to resist falsehoods flung their way? Or do those characteristics (partisan or ideological commitment, need for cognition, etc.) that make them more knowledgeable about some propositions also make them more misinformed about others? In that case, the more generally knowledgeable may also tend to be more misinformed. More broadly, we consider misinformation’s dependence on the individual’s partisan and ideological commitments, interest in politics, and personality traits. Among other things, we consider the question of partisan/ideological asymmetry: is misinformation more prevalent on the left or the right (and when—the answer may not be the same everywhere or under all conditions)?
This leads to a final set of questions concerning the effects of the proposition. First, is the proposition true or false? Denial (misinformation rejection consisting of confident rejection of a truth) may behave different from invention (misinformation consisting of a confident acceptance of a falsehood). Some of the propositions the FNES poses are true, others false. Second, to whom is the proposition congenial? Some of the propositions the FNES poses are things the left finds pleasing (and the right displeasing) to believe, others the reverse. Presumably misinformation (and mere incorrect belief) are more common on the left for left-congenial propositions, more common on the right for right-congenial ones.

Authors

• Robert C. Luskin, University of Texas, Austin

• Pavlos Vasilopoulos, CEVIPOF

• Martial Foucault, Sciences Po Paris

• Sylvain Brouard, Sciences Po