alt Hacker NewsThis is also a political problem. Most of the debates about racism for instance are between people who are illiterate in term of basic statistics, on both sides of the debate. They take extreme tails of distributions as benchmarks, think in term of stereotypes rather than distributions, infer rules from a handful of samples, etc. It makes the whole exercise sterile.
It is not just journalists and politicians that are guilty of that. I see that from many academics who should know better, particularly in humanities departments.
Replies
<https://en.wikipedia.org/wiki/Motivated_reasoning> affects us all. I'd like to believe education is an antidote, but really the only antidote is... not doing that.
This isn't new. "Lies. Damned lies and Statistics" as a phrase is already quite old.
Many people selectively shop around statistics that match their prejudice or their social circles social etiquettes.
Are you talking about actual academic work or the pop- version of it? Because most of the damage I see is in pop-psychology, pop-sociology. In my limited experience looking into the topic humanities researchers make specific balanced claims backed with some numbers and reasoning, and all the nuance gets lost when translated to news articles and pop-sci books or videos