tetano
Editor, Senior Moderator
PLoS One
. 2021 Oct 6;16(10):e0258189.
doi: 10.1371/journal.pone.0258189. eCollection 2021.
Will I die of coronavirus? Google Trends data reveal that politics determine virus fears
Joan C Timoneda[SUP] 1 [/SUP], Sebastián Vallejo Vera[SUP] 2 [/SUP]
Affiliations
Abstract
Is Google Trends (GT) useful to survey populations? Extant work has shown that certain search queries reflect the attitudes of hard-to-survey populations, but we do not know if this extends to the general population. In this article, we leverage abundant data from the Covid-19 pandemic to assess whether people's worries about the pandemic match epidemiological trends as well as political preferences. We use the string 'will I die from coronavirus' on GT as the measure for people's level of distress regarding Covid-19. We also test whether concern for coronavirus is a partisan issue by contrasting GT data and 2016 election results. We find strong evidence that (1) GT search volume close matches epidemiological data and (2) significant differences exist between states that supported Clinton or Trump in 2016.
. 2021 Oct 6;16(10):e0258189.
doi: 10.1371/journal.pone.0258189. eCollection 2021.
Will I die of coronavirus? Google Trends data reveal that politics determine virus fears
Joan C Timoneda[SUP] 1 [/SUP], Sebastián Vallejo Vera[SUP] 2 [/SUP]
Affiliations
- PMID: 34614032
- DOI: 10.1371/journal.pone.0258189
Abstract
Is Google Trends (GT) useful to survey populations? Extant work has shown that certain search queries reflect the attitudes of hard-to-survey populations, but we do not know if this extends to the general population. In this article, we leverage abundant data from the Covid-19 pandemic to assess whether people's worries about the pandemic match epidemiological trends as well as political preferences. We use the string 'will I die from coronavirus' on GT as the measure for people's level of distress regarding Covid-19. We also test whether concern for coronavirus is a partisan issue by contrasting GT data and 2016 election results. We find strong evidence that (1) GT search volume close matches epidemiological data and (2) significant differences exist between states that supported Clinton or Trump in 2016.