Is improperly used statistics to blame for poor reproducibility?

[Joseph Malkevitch]

Dear Colleagues,

There has been a lot of attention devoted recently to why various studies when repeated don't show the same results of the original study.

One concern is that many use "p-tests" in a way that gives the impression that results are more "iron clad" than is truly the case.

Here are some recent comments on this debate.

https://www.nature.com/magazine-assets/ ... 7522-z.pdf

Regards,

Joe