While there is a strong scholarly interest surrounding the content of political misinformation online, much of this research concerns misinformation in Western, Educated, Industrialized, Rich and Democratic (WEIRD) countries. Although such research has investigated the topical and stylistic characteristics of misinformation, its findings are frequently not interpreted systematically in relation to properties that journalists rely on to capture the attention of audiences, that is, in relation to news values. We close the gap on comparative studies of news values in misinformation with a perspective that emphasizes non-WEIRD countries. Relying on a dataset of URLs that were shared on Facebook in twenty-four countries and reported by users as containing false news, we compile a large corpus of online news items and use an array of computational tools to analyze its content with respect to a set of five news values (conflict, negativity, proximity, individualization, and informativeness). We find salient differences for almost all news values and regarding the WEIRD/non-WEIRD and flagged/unflagged distinction. Moreover, the prevalence of individual news values differs strongly for individual countries. However, while almost all differences are significant, the effects we encounter are mostly small.
