Media bias exposure and the incidence of COVID-19 in the USA

The role of the media as a source of reliable health information during the COVID-19 pandemic has come under intense scrutiny, with claims of misinformation and partisanship coming from all sides of the political divide. This paper seeks to understand the relationship between exposure to biased media outlets and the likelihood of testing positive for COVID-19 in the USA. I use detailed household data extracted from the 2020 American National Election Study in order to gauge media consumption patterns, coupled with data on media bias scores for different outlets and programmes. I combine these variables to compute media bias exposure values for each respondent, and relate these to the likelihood of a positive COVID-19 test within each respondent's household, controlling for a variety of other factors including partisanship, social media use, trust in the media and several socioeconomic and demographic variables. The results indicate that media bias exposure is significantly related to COVID-19 incidence, and in particular the coefficients show that a 1% increase in exposure to left-wing media is associated with a 0.2% decrease in the probability of a positive COVID-19 test. Conversely, I find no significant relationship between right-wing media exposure and COVID-19 infection rates. I also find a significantly higher likelihood of contracting COVID-19 among low socioeconomic status households, suggesting a disproportionate impact of the pandemic on such cohorts. These findings are robust to a number of tests, and emphasise the importance of aligning media messages with those advocated by leading medical experts during public health crises.