Scientism redux

Stick or switch?

Some years ago, reading Case Studies in Sample Design by A.C. Rosander and some other books on sampling practice by W. Edwards Deming, TOF was offered this particular nugget of information.  On any particular problem there will be statistical aspects and non-statistical aspects and the practicing statistician's task is to provide advice on the former.  The statistics is not the be-all and end-all of the problem in view.  There may be a wide number of other aspects to it.  A statistical analysis may shed light on some of these, and that is all to the good.  But the important thing to remember is that not all problems are simply statistical problems.

Consider the Monte Hall problem, which puzzled even pros because they forgot that the problem was not probabilistic.  Monte Hall did not use random chance to choose which door he would open.  Let us call this tendency to apply the rules of probability even in non-random situations probabilism .  Which serves as today's lead-in.
Read more »