What inspired this Wish List Request?
In 2016, American Statistical Association issued the p-value statement. At #3 in the p-value statement, it is said that "Scientific conclusions and business or policy decisionsshould not be based only on whether a p-value passesa specific threshold. in the statement."
In 2019, the Moving to a World Beyond "p < 0.05" meeting also called "Woodstock for inference" was held. In the Editorial in The American Statistician for the meeting, it is said that "Don't Say "Statistically Significant"". (In 2021, the ASA Presidential Task Force reported that this Editorial is not issued by ASA, but by the editors.)
At the tutorial seminar at Japanese Joint Statistical Meeting this year, this topic was picked up again by Prof. Toshiya Sato. He picked up Prof. Greenland's opinions. Prof. Greenland is saying that we should stop dichotomization of P-values and testing only zero null hypotheses.
Please describe the new or improved feature or solution you would like to see.
My suggestions are...
(1) A Preference Option to output LogWorth (=-Log10(p)) or S-values (=-Log2(p)) instead of p-values.
(2) A Preference Option to surpress the coloring and astarisk formats based on p<0.05 and p<0.01.
(3) Otions to draw p-value functions and S-value functions.
What would the value be to you and/or other users if this new feature is provided?
As far as I know, many statisticians know about the ASA 2016 p-value statement, and I guess that many of them agree on the statement.
Is this an external customer request?
No, this is my personal request. But when I talked informally with Prof. Shuntaro Sato at Nagasaki University Hospital (,who translated Modern Epidemiology,) he said that JMP should stop the formatting for p<0.05 and p<0.01. (He is a JMP Student Edition user.)
Is there a competitive software package that has this feature or solution that we should be aware of?
I do not know. But I guess R must have some functions to calculate S-values from p-values.
Will this impact a renewal, expansion, or adoption at a customer, and does this put an account at risk if we do not address this?
No.
