Public Health Communication and Engagement on Social Media during the COVID-19 Pandemic
Executive Summary
Social media gives governments a direct channel to communicate with citizens, and during a pandemic reaching as many people as possible with health messaging is critical to reducing the spread of disease. This study evaluates efforts by Canadian local, provincial, and federal governments to spread health-related information on Facebook and Twitter during the first five months of the COVID-19 pandemic. The authors collected all COVID-19-related communications from government accounts on both platforms between January 1 and May 24, 2020, ultimately analyzing 14,674 tweets and 4,510 Facebook posts after filtering to organic, health-related content. Using a nested mixed-method approach, they first identify quantifiable features linked to citizen engagement through large-N regression analysis, then perform qualitative content analysis on outlier (over- and under-performing) posts.
The analysis finds significant cross-platform differences in engagement strategy effectiveness. Despite posting far less frequently on Facebook, government accounts achieved greater total engagement there (approximately 4.9 million likes and shares) than on Twitter (approximately 2.9 million favorites and retweets). On Twitter, video content produced the largest engagement gains (a coefficient translating to roughly a 121 percent increase over text-only tweets), followed by images (about 35 percent), while hashtags added roughly a 12 percent increase per tag; more frequent daily tweeting reduced engagement per tweet by about 4.5 percent, a penalty not observed on Facebook. Who posts also mattered as much as what was posted: tweets from the Prime Minister's account generated 727 percent more engagement than those from large city government accounts, premiers' accounts 702 percent more, and mayors of large cities only 196 percent more.
Qualitative analysis of the 300 most over- and under-performing posts showed that concise messages with clear behavioral directives, high-quality media, and the use of influencers or celebrities tended to overperform, while negative news, low-quality media, and posts focused narrowly on politics or policy announcements tended to underperform. The study concludes that governments should pair national and sub-national leaders with health authorities in messaging, favor simple and direct calls to action, invest in creative and high-quality media, and tailor content differently to each platform rather than posting near-identical messages on both.
Related Publications

Building a Media Ecosystem Observatory from Scratch: Infrastructure, Methodology, and Insights
Read more
Can-PolNews: A Multi-Platform Dataset of Political Discourse in Canada
Read more
Polarization as the Technological Goal – Not the Error
Read more
