These are the headline metrics noted by the NZ Ministry of Health:
The metrics are updated daily and reported dutifully by the NZ news media, but what use are they, in fact? What information and knowledge can we glean from the data?
Here is the current summary (snapshot at 7am on April 2nd):
There are no detailed definitions of these data, and my beady eye spots little differences, for example whereas the headline says "how many people are (and have been) in hospital", the data actually provided are "Number of cases in hospital" showing the "Total to date" and "New in last 24 hours". If this is a cumulative total of the number of COVID-19 cases admitted to hospital each day, are they just the "confirmed" cases or does that include the "probables"? And what is a "case" anyway? What is a "hospital"? [Yes, I know, it's a place where sick people go, and that's not important right now ... but in the more remote parts of NZ where hospitals aren't, I guess COVID-19 patients may be treated at home or in makeshift facilities: do these qualify as "hospitals" or not?].
Another concern is that they are just numbers with no clear context. That 1 NZ death, for instance: is it above or below expectations? Is it significant? Are we even sure that it was a direct result of COVID-19, or merely a coincidence? And how does NZ compare to other countries? ...
... which is an important issue in these times of global comms. Aside from occasional mentions by the NZ news media of situations in China, Italy etc., vast rivers of information are flowing through the Interweb from all over the globe - some definitive, factual and reliable, some not. All the usual comms and information security issues apply e.g. integrity failures such as errors and omissions, particularly in how the information is interpreted and discussed by non-specialists (including me, no doubt!). The Web is a giant echo chamber.
There are other concerns with the "official" NZ COVID-19 metrics but I'll stop here because I have Stuff To Do, including thinking about the reasons for measuring and reporting "official" metrics at all. I'll have more to say on this tomorrow, maybe, once I've thunk.