We need more frequentism, not less.

Two applications:

1️⃣ The best critiques of NHST practices are frequentist critiques (e.g., significance filter)

2️⃣ The best justifications for Bayesian estimators are frequentist justifications (e.g., improvement in RMSE from regularization)

2 points

1️⃣ discussion is much harder to follow as an OP and a reader when folks QT

2️⃣ I use QTs when a reply would derail discussion of the original point—a “sorry for hijacking the thread” kinda post

The algorithm is good at boosting replies appropriately, so default to that

I like to write papers.

🚨🚨New Post🚨🚨

'Statistical Power Is for You, Not for Reviewer Two'

We focus on the consequences of low power for the community (eg. Type M errors), but low power is immediately relevant for researchers themselves.

Low Power → Wasted Opportunity

carlislerainey.com/blog/2023-05-2…

We haven't heard from the 15% in the quotes and replies.

I'd like to hear from the anti-interdisciplinarians--what is the downside?

Just posted a minor update to the preprint version of this paper.

osf.io/preprints/soca…

What do you think about boundaries between disciplines?

E.g., the boundary between political science and psychology

Would we be better off if the boundaries were stronger (less interdisciplinary work)? Or would we be better off if they were weaker (more interdisciplinary work)?

Sticky notes on my wall.

They go in 3 bins.

1.) Active: arranged closest to furthest from publication

2.) Waiting: underway, but I’m waiting for someone else (under review, with a collaborator)

3.) Future: maybe I’ll do this later

Then work on the top project in bin 1 daily.

Luskin's 'Robert's Rules: Suggestions for Writing' has some excellent, pithy advice.

PDF: github.com/soodoku/on-wri…

Serious question: under what definition of “Bayesian” is this researcher a Bayesian?

Is a Bayesian just someone who regularizes?

Oh no