Do Twitter's Science Stars Get a Citation Premium?

We analyze whether the social media popularity of Twitter star scientists, who were identified by Science in a 2014 report, pays off in terms of an increased number of citations. To establish a causal relationship, we use the COVID-19 global pandemic as a quasi-natural experiment exogenously increasing public attention and the demand for expertise. Using Twitter science stars' and their coauthors' publications on COVID related topics prior to the break out of the pandemic, we run a difference-in-differences analysis for annual incoming citations of the two groups. We find that the Twitter star status added about 1.07 extra citations following the breakout of COVID-19 per year per article, corresponding to about 70% of the already existing citation gap between Twitter science stars and their coauthors. Moreover, we also document that the publication of the Science list on Twitter science stars in 2014 per se caused an increase in citations, i.e. the publication of the supposed celebrity status by Science already benefited the stars, which meant 1.06 more citations per year per article compared to their coauthors. Treatment based on scientists' Kardashian indexes yields no robust effects, implying that unjustified social media popularity does not pay off in terms of citations.