When I was involved in product development, I was terribly enraged by the pragmatic designers - the ones who tried to do everything only on the basis of statistical research.
So I want the button to be green, just because I like it better. And the designer says - “it doesn't matter, AB tests have shown that the button of a diarrhea color is clicked 0.2% more often”. Lord, buddy, you have been pumping your taste and experience for ten years, so what? To make our product look like bird poop? But business says - since there are numbers, then we will cover everything with it.
I understand people want to make money. They don't want to trust their taste when it comes to crowd satisfaction. But now I know that the problem may not be in numbers, but in people who do not know how to use statistical tests.
Last week in our podcast was Andrey Akinshin, PhD in Physics and Mathematics and an expert in performance analysis. He told us why he, too, is bombed by modern mathematical statistics.
. — .
, . « ». , , , . – . , , , .
« » — . . , , P-value, . P-value , , . , ( « » ).
- , . — . - . . P-value , . , .
, -, , . , – , : , , , . , ! .
– 0,05. ? , 30- , , – , — . 20 , , , .
0,05. — , . Qwerty, , . Qwerty .
. 80- , , . – . , , - , , – «false positive». . , , – «false negative».
0.2. . : «, , , , . – 0.05. ». , , 0.2 – , , .
: «, , , , , . , , 0.2, ». , . , . -, . , ?
, , - .
. , , , , , . . .
, – . , . P-value. – P-value , 0.05. 0.049 0.051, : «! !».
0.9, , . «» , 20 , P-value , . , .
, . , – -, , . , . - , , , .
— -. — , -.
: ? -, -, P-value . : «, - , , ». , . , – .
– , .
, , , .
, : « ?» ( ), : « !».
– 0 1. . , . , , , , , . . – . : « » « - » — .
. — ? -, , . ; , – P-value – .
, .
, , - — « , . ». , . . . — -, , . , , P-value, — , ! - .
, . , , .
, , . , . . « - , , , , ».
— -, , . , . , — - .
Harlow, Lisa L., Stanley A. Mulaik, and James H. Steiger, eds. What if there were no significance tests? Routledge, 2016.
Cohen, Jacob. "The earth is round (p<. 05)." In What if there were no significance tests?, pp. 69-82. Routledge, 2016.
Amrhein, Valentin, Fränzi Korner-Nievergelt, and Tobias Roth. "The earth is flat (p> 0.05): significance thresholds and the crisis of unreplicable research." PeerJ 5 (2017): e3544.
Wasserstein, Ronald L., Allen L. Schirm, and Nicole A. Lazar. "Moving to a world beyond “p< 0.05”." (2019): 1-19.
Winder, W. C. "What you always wanted to know about testing but were afraid to ask." American dairy review (1973).
Grieve, Andrew P. "How to test hypotheses if you must." Pharmaceutical statistics 14, no. 2 (2015): 139-150.
Krawczyk, Michał. " The search for significance: a few peculiarities in the distribution of P values in experimental psychology literature ." PloS one 10, no. 6 (2015).
And one more speech by Andrey, where he talks about the problem in more detail: