Thread by @evpatora

I am teaching intro to epidemiology concepts to research year students at @DCEAarhus. One of the topics I mention is retiring null hypothesis testing. Here are some of my favorite materials on abolishing null hypothesis testing/p-value misinterpretation⤵️

1/n #IYKYK #epitwitter

Work smart: one amazing collection already exists

https://twitter.com/_MiguelHernan/status/1476928329794027522?s=20&t=lM-euC6Sz1jN6LE56U6Fnw

2/n

“Scientists rise up against statistical significance” @vamrhein This paper (thankfully) got >1.5 citations 👏 nature.com/articles/d4158…

3/n

51% misinterpretation proportion based on p-values dichotomization🙈

4/n

But what to do: use p-value functions for one. “A p-value function allows us to visualize results continuously by examining how the p-value varies as we move across possible effect sizes”

https://twitter.com/vamrhein/status/1540078850545225729?s=20&t=9MRrO0JzfUhoVA6qRYZCzQ

5/n

One great practical example of using the p-value function sciencedirect.com/science/articl… @ken_rothman @Morten__Schmidt @DCEAarhus

6/n

Tired of P-values? Meet S-values (if you like games, you’ll like them):

https://twitter.com/dailyzad/status/1174744433025265665?s=20&t=y5XKa-BlkzGZ5ZcjtOUW1w

and tandfonline.com/doi/pdf/10.108…

7/n

One of the best papers I have ever read. Not easy to digest, especially when first introduced to it. Wondering if there ever will be a time when I remember all 24 p-value misconceptions. Note to self: time to reread 🤓

https://twitter.com/ken_rothman/status/862110114953744385?s=20&t=bvuUFifdGVHsYUUtLwpRjw

8/n

Another favourite #epitwitter reading suggestion:

https://twitter.com/MaartenvSmeden/status/1052701609178718208?s=20&t=A5lqHr9Ts-Y7CY19Sk7uMA

Brushing up on logic: absence of evidence != evidence of absence.

9/n

Finally, maybe, the best preface to "what's wrong with p-values" talk and why we should and can do better.

https://twitter.com/cecilejanssens/status/1443885400590782466?s=20&t=lM-euC6Sz1jN6LE56U6Fnw

n/n