Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Oh the piles of bodies. Please check the details behind such claims before trotting them out.

By the same US federal data COVID brought someone back to life. I'm not joking the obvious mistake is recorded and is a wonderful example of how the data needs to be cleaned checked and examined more closely.



Hundreds of millions of entries of data input by a large group of people across a wide region will always have errors and inconsistencies. This is true of data related to COVID-related deaths, homicides, car accidents, daycare enrollment rates, and everything in between.

Showing that individual data points are inconsistent or wrong does not by itself indicate that conclusions gleaned from a data set as a whole are incorrect.

Read the accounts from funeral home staff some time. We were very literally running out of places to put the actual, physical bodies. It turns out when you have a mortuary system that’s set up to process a reasonably consistent and predictable number of deceased humans, adding an sudden, unexpected, and sustained additional load of 10–15% (with spikes up to 33% and 45%) system stretches it beyond its limit.


No but it's a glaring example of referring to a non consistent dataset.

The data should be cleaned first to remove such problems which impacts totals.

Such work then has to look at labelled as having COVID at time of death Vs sole cause which is _missing_ from most Western datasets. And I'd be willing to believe it's missing from totals that large, unfortunately raising a question as to how many actually were killed by despite the fact we know >80% were likely over 80yr...

Not to forget the nice example, if you died from a heart attack due to a racy TV advert did the TV kill you?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: