The Joe Rogan Experience Podcasts

In short, Rogan repeatedly spread dubious coronavirus-related information. In December 2020, regular Rogan listeners’ intentions to vaccinate were 15 percentage points lower than those of non-listeners. By February 2021, they were 18 percentage points lower, both statistically significant effects.

We conducted demographically representative surveys once every two months, beginning in April 2020 and ending in February 2021, via Lucid Theorem’s online opt-in sampling service. Each survey sampled about 1,000 Americans, with the exception of the February 2021 survey, which sampled about 1,500. Lucid Theorem uses quota sampling to produce samples that resemble the U.S. adult population with respect to age, gender identity, racial identity, household income, educational attainment, political partisanship and geographic region. To account for any remaining deviations between the sample and U.S. adult population, we weight responses to U.S. census benchmarks on age, gender, race, household income and educational attainment. Each survey asked respondents to report whether they were “very likely,” “somewhat likely,” “not too likely” or “not likely at all” to be vaccinated against the coronavirus once a vaccine became widely available.

With this information, we scored respondents as more vaccine hesitant — meaning, intending to forgo a vaccine — if they indicated that they were “not too likely” or “not likely at all” to receive a vaccine. In the study’s February wave, those who said they already been vaccinated were scored as not hesitant. We also asked respondents to report how frequently, in the previous month, they had watched or listened to dozens of different programs. In our analysis, we compare the effects of Rogan listenership with how frequently respondents watch “local news” and “national news” broadcasts and listen to programs such as “The Daily” (a popular podcast) and NPR (on the radio).

Respondents could indicate that they watched or listened “never,” “just once or twice,” “about once a week” or “almost every day.” We considered respondents to be regular viewers or listeners of each program if they reported watching or listening to each program at least once a week that month. We also asked respondents about such information as their political partisanship, attitudes toward scientific experts and personal demographics.

In short, Rogan repeatedly spread dubious coronavirus-related information. In December 2020, regular Rogan listeners’ intentions to vaccinate were 15 percentage points lower than those of non-listeners. By February 2021, they were 18 percentage points lower, both statistically significant effects.

That’s also noticeably different than for listeners of comparable radio programs and podcasts. Regular NPR radio and “The Daily” podcast listeners, for example, were statistically neither more or less likely to intend to vaccinate throughout the duration of the study and in some survey waves were even significantly more likely than non-listeners to intend to vaccinate. For example, in December 2020, NPR and “The Daily” listeners’ intentions to vaccinate were respectively 18 and 19 percentage points higher than those of non-listeners.

Our data is correlational. We cannot determine that listening to Rogan causes someone to become skeptical about the vaccine; people who are skeptical about the vaccine may be more likely to listen to Rogan. But we can conclude that Rogan’s audience is more likely to hesitate to get the vaccine, compared with listeners of our set of other podcasts and radio programs.

This finding is consistent with the idea that Rogan listeners may be heeding his and his guests’ nonexpert medical advice. As someone who is able to garner a truly massive audience in a very fragmented media landscape, Rogan’s voice becomes very important — both because he is influential and because his influence is often ignored by researchers and public health planners who tend to focus on legacy and social media.

Did they account for age, who they follow on twitter, party affiliation, profession, etc etc etc.

We conducted demographically representative surveys once every two months, beginning in April 2020 and ending in February 2021, via Lucid Theorem’s online opt-in sampling service. Each survey sampled about 1,000 Americans, with the exception of the February 2021 survey, which sampled about 1,500. Lucid Theorem uses quota sampling to produce samples that resemble the U.S. adult population with respect to age, gender identity, racial identity, household income, educational attainment, political partisanship and geographic region.

To account for any remaining deviations between the sample and U.S. adult population, we weight responses to U.S. census benchmarks on age, gender, race, household income and educational attainment. Each survey asked respondents to report whether they were “very likely,” “somewhat likely,” “not too likely” or “not likely at all” to be vaccinated against the coronavirus once a vaccine became widely available.

With this information, we scored respondents as more vaccine hesitant — meaning, intending to forgo a vaccine — if they indicated that they were “not too likely” or “not likely at all” to receive a vaccine. In the study’s February wave, those who said they already been vaccinated were scored as not hesitant. We also asked respondents to report how frequently, in the previous month, they had watched or listened to dozens of different programs. In our analysis, we compare the effects of Rogan listenership with how frequently respondents watch “local news” and “national news” broadcasts and listen to programs such as “The Daily” (a popular podcast) and NPR (on the radio).

Respondents could indicate that they watched or listened “never,” “just once or twice,” “about once a week” or “almost every day.” We considered respondents to be regular viewers or listeners of each program if they reported watching or listening to each program at least once a week that month. We also asked respondents about such information as their political partisanship, attitudes toward scientific experts and personal demographics.