search

nzherald / nz-covid19-data

Repackaging the Ministry of Health's Covid-19 data
24 stars 9 forks source link

Running count of tests #2

Closed PetraOleum closed 4 years ago

PetraOleum commented 4 years ago

Is that published anywhere? A.B. just said 9780 so far, if I heard correctly

vizowl commented 4 years ago

I wish :) As far I it sometimes gets mentioned in the press conferences and sometimes doesn't - sometime I should go and listen to them all again ...

PetraOleum commented 4 years ago

Is there an archive of those?

On Wed, Mar 25, 2020 at 3:26 PM Christopher Knox notifications@github.com wrote:

I wish :) As far I it sometimes gets mentioned in the press conferences and sometimes doesn't - sometime I should go and listen to them all again ...

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/nzherald/nz-covid19-data/issues/2#issuecomment-603604260, or unsubscribe https://github.com/notifications/unsubscribe-auth/AENQKMF24B6G2ZVGFS3ZXP3RJFTVZANCNFSM4LTCH7VQ .

vizowl commented 4 years ago

There is - https://www.health.govt.nz/our-work/diseases-and-conditions/covid-19-novel-coronavirus/covid-19-novel-coronavirus-news-and-media-updates

But it is complicated by the language they use - today he said 1421 yesterday he said 'over 900' - I start adding the accurate numbers from today and if someone ;) wants to go back through the releases and look at what the said we could add tests with two columns - a number and an estimate/accurate flag...

vizowl commented 4 years ago

Actually I have started it with

testDates <- tribble(
  ~Date, ~Tests, ~`Total Tests`, ~`Test Count`,
  "2020-03-23", NA, 9780-1421-900, "Estimate",
  "2020-03-24", 900, 9780-1421, "Estimate",
  "2020-03-25", 1421, 9780, "Accurate",
  ) %>% mutate(Date=as.Date(Date))

I'm not sure whether it would be worth pushing further back - but I do recall numbers were mentioned occasionally so it could be a fun challenge...

vizowl commented 4 years ago

The test count summaries don't add up - they are out by a few hundred between yesterday and today - I have left them in my series but I'm not sure if the reported numbers will be reliable enough to do anything with.