The Role of Social Media in Polarizing Canadians

Executive Summary

Popular criticism holds that social media polarizes the mass public, yet a large and growing body of scholarly research casts doubt on this, finding that most social media users are exposed to a wide range of political information and that social media use and information campaigns generally have minimal impact on attitudes and behaviours. This essay reconciles the two views: while the majority of users experience little effect, a subset of users becomes drawn into limited, biased, and self-reinforcing information environments, and it is this group that experiences the strongest polarizing effects. Estimates across studies converge on roughly 5-10% of people in online spaces occupying echo chambers or experiencing heightened, largely algorithmically-driven polarization.

This small group disproportionately shapes political discourse because it is far more active than average. In a study of digital activists in Canada, 10% of the population produced approximately 80% of Canadian politics-related content on Twitter during the 2019 federal election, and this 10% showed stronger political conviction and much higher affective polarization than average Canadians. Similarly, 10% of users accounted for 98% of exposures to Russian Internet Research Agency content during the 2016 U.S. election, and removing roughly 15,000 hyperactive hate-speech producers from Reddit cut hate speech overall by 80%. This same group is also more likely to engage in embodied political participation - protests, blockades, harassment, and voting in low-turnout contests that select political leaders - illustrated by the 47 people arrested for blockading the Ambassador Bridge (which carries about 25% of U.S.-Canada merchandise trade), the fewer than 20,000 Canadians who shut down downtown Ottawa during the Freedom Convoy, and Canadian party leadership races decided by roughly 1-2% of eligible voters.

The essay concludes that it can be simultaneously true that most Canadians are not living in information bubbles or being radicalized by social media, and that a smaller but highly active and vocal population is - and that this second population poses a serious risk to democratic discourse and governance because of its outsized influence on political conversation, leadership selection, and key political moments.