datasets Health needs more dimensions.

Maybe join in demographics? Where'd we get this from?

Nov 10 '16 15:11 acbart

I have no clue where we got this, but if we were to include the demographic data for this set then it would only be one year as opposed to the range we have right now, 1928 - 2011. So I'm not sure if it's worth keeping, even if we had the demographic info added

Jan 11 '17 19:01 RyanWhitcomb-VT

Apparently this comes from project Tycho: https://www.tycho.pitt.edu/

Why can't we get demographic data going back that far? Seems like the info is out there, just needs a lot of interpolation. For example this has populations from census for the past while (http://www.infoplease.com/ipa/A0004986.html). We can probably get a number of other factors too.

Jan 15 '17 21:01 acbart

They make the data available at the weekly level. One interesting option is to include data for the entire year, but also for individual months or seasons.

Further, they report Cases and Incidence Rate, so we can do something very similar to what we do with crime data.

Feb 11 '17 04:02 acbart

I believe the "Increase" field in this dataset is actually meant to be "Incidents"! Goodness, what an error.

Apr 01 '18 20:04 acbart