Is it just me, or does this widely-shared figure from Monday's @nytimes cover story on herd immunity WAY overstate actual COVID-19 vaccine hesitancy in the US?
: https://twitter.com/AdamUren/status/1389188406731706371

The figure just cites Health and Human Services, but the data source seems to be the same one used in this @MPRnews article by @dhmontgomery: https://bit.ly/3tiECHP . Note that at the state level, vaccine hesitancy never exceeds 31% (2/)

The underlying HHS data is at https://aspe.hhs.gov/pdf-report/vaccine-hesitancy. There were only 4 survey responses—the NYT map is supposed to capture the first 2 ('def/prob'), and the HHS definition of "hesitancy" includes the last 2 ('prob/def not'). That is, NYT definition = 1 - hesitancy... (3/)

... but this isn't borne out in the data! HHS estimates a max of 32% vaccine hesitancy across all US counties.

Take Wyoming: no counties should fall in the lightest color bin. The data only matches if you double-count "strongly hesitant" (already included in "hesitant"). (4/)

You can take a look at the original (often less grim) HHS county map of vaccine hesitancy here: https://data.cdc.gov/stories/s/Vaccine-Hesitancy-for-COVID-19/cnd2-a6zw And for more info on the limitations behind these estimates, see this thread: https://twitter.com/dhmontgomery/status/1389223136189501444 (5/5)

Update: I remade the map using the same color scales and data source. Still lots of interesting patterns to dig into, but Minnesota is no longer an island of vaccine acceptance (6/)