Reducing Social Media Usage During Elections: Evidence from a Multi-Country WhatsApp Experiment

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Social media messaging platforms are central to worldwide communication, but are also major hubs of misinformation and toxic content. On these platforms, informationspreads through interpersonal and group-based chats rather than feed-based recommendations. We argue that introducing barriers to usage can increase the costs ofconsuming low-quality content and promote more deliberate engagement, shaping information consumption and downstream attitudes. We evaluate our argument throughthree coordinated online field experiments in Brazil, India, and South Africa. We incentivize participants to either avoid multimedia content on WhatsApp or to limit theirusage to 10 minutes per day for four weeks ahead of each country’s elections. Our interventions significantly reduced participants’ exposure to uncivil political discussionsand misinformation—but at the expense of keeping up with political news. However, political attitudes did not shift, although treated participants did report improved wellbeing, particularly when they substituted WhatsApp usage with more offline activities.

Article activity feed