LAUSR

Abstract Should materials aimed at increasing Latino voter turnout be in English or bilingual? Credible, theoretical arguments can be made both ways: bilingual materials may be more effective if signaling cultural awareness or less effective if seen as pandering. We tested these competing hypotheses with two rounds of randomized field experiments in New Jersey and Virginia in 2015, and North Carolina in 2016. While some GOTV experiments have used bilingual mailers, previous scholarship has not tested whether bilingual mailers are more effective than English-language materials. In the 2015 elections, both treatments increased turnout, and the monolingual English version was more effective at increasing turnout than the bilingual version. These results are replicated in the high salience 2016 election in North Carolina. These results indicate that further research is needed about bilingual communication across political and demographic contexts and about how household composition may condition the effects of bilingual communication.