Facebook

We study the effect of Facebook and Instagram access on political beliefs, attitudes, and behavior by randomizing a subset of 19,857 Facebook users and 15,585 Instagram users to deactivate their accounts for 6 wk before the 2020 U.S. election. We report four key findings. First, both Facebook and Instagram deactivation reduced an index of political participation (driven mainly by reduced participation online). Second, Facebook deactivation had no significant effect on an index of knowledge, but secondary analyses suggest that it reduced knowledge of general news while possibly also decreasing belief in misinformation circulating online. Third, Facebook deactivation may have reduced self-reported net votes for Trump, though this effect does not meet our preregistered significance threshold. Finally, the effects of both Facebook and Instagram deactivation on affective and issue polarization, perceived legitimacy of the election, candidate favorability, and voter turnout were all precisely estimated and close to zero.

The question of how social media usage impacts societal polarization continues to generate great interest among both the research community and broader public. Nevertheless, there are still very few rigorous empirical studies of the causal impact of social media usage on polarization. To explore this question, we replicate the only published study to date that tests the effects of social media cessation on interethnic attitudes (Asimovic et al., 2021). In a study situated in Bosnia and Herzegovina, the authors found that deactivating from Facebook for a week around genocide commemoration in Bosnia and Herzegovina had a negative effect on users’ attitudes toward ethnic outgroups, with the negative effect driven by users with more ethnically homogenous offline networks. Does this finding extend to other settings? In a pre-registered replication study, we implement the same research design in a different ethnically polarized setting: Cyprus. We are not able to replicate the main effect found in Asimovic et al. (2021): in Cyprus, we cannot reject the null hypothesis of no effect. We do, however, find a significant interaction between the heterogeneity of users’ offline networks and the deactivation treatment within our 2021 subsample, consistent with the pattern from Bosnia and Herzegovina. We also find support for recent findings (Allcott et al., 2020; Asimovic et al., 2021) that Facebook deactivation leads to a reduction in anxiety levels and suggestive evidence of a reduction in knowledge of current news, though the latter is again limited to our 2021 subsample.