Reasoning under uncertainty sounds intriguing. Brings up images of logic, philosophy, and artificial intelligence.
Statistics sounds boring. Brings up images of tedious, opaque calculations followed by looking some number in a table.
But statistics is all about reasoning under uncertainty. Many people get through required courses in statistics without ever hearing that, or at least without ever appreciating that. Rote calculations are easy to teach and easy to grade, so introductory courses focus on that.
Statistics is more interesting than it may sound. And reasoning under uncertainty is less glamorous than it may sound.
I’ve been using the term “estimation” when informally discussing applied statistics. Thanks to the media coverage of Coronavirus, I’ve also become able to use “modeling” as a technically incorrect by functionally valid replacement.
My favorite SARS-CoV-2/COVID-19 sites and videos to share:
https://www.worldometers.info/coronavirus/country/us/
https://aatishb.com/covidtrends/
https://www.youtube.com/watch?v=gxAaO2rsdIs
https://www.youtube.com/watch?v=54XLXg4fYsc
I like the phrase “reasoning under uncertainty.”
Statistics textbooks sometimes describe statistics as “decision making under uncertainty,” but that always bothered me, because there’s very little about decision making in statistics textbooks.
Clearly I need to redouble my efforts in promoting this idea. If only to convince my undergraduate Poindexters that statistics calculated to 7-digit accuracy are misleading when the sample size is 11.
Michael:
It’s not just undergraduates who give too many digits. The major news media do it too, perhaps to sell the idea that they are being super-precise and quantitative; see here: https://statmodeling.stat.columbia.edu/2012/10/22/is-it-meaningful-to-talk-about-a-probability-of-65-7-that-obama-will-win-the-election/