If you leave out the ‘beyond bare charts, numbers, graphs, and facts’ part I’d go along.
It’s pretty easy to tweak the numbers just so to get the charts and graphs you want. A more elegant approach is tweaking the collection of the data itself, i.e. make sure you don’t get data you don’t want in the first place.
If you want to gauge how reliable statistics are you need to see the raw data and you need to know how it has been collected and processed.
And that brings me to the ‘facts’. We can only work with whatever we find in the archives etc., i.e. with whatever has been recorded. We do not know what hasn’t been recorded. If there is more than one source and if there is background information on the source itself the information is more reliable.
Recorded facts are hard facts but they are proably not objectice facts.
The mere decision to record this bit of information but not that bit of information is already an interpretation.
2 Likes
