kevinmking.bsky.social
He/him. Something something Self control and quant methods. Professor of Psychology at University of Washington. All opinions are my own and correct. Co-host of https://thatimplementationsciencepodcast.podbean.com/
https://faculty.washington.edu/kingkm
1,325 posts
5,792 followers
718 following
Prolific Poster
Conversation Starter
comment in response to
post
Congratulations Brandon and Michael!
comment in response to
post
You seem pretty angry to me Colette
comment in response to
post
I'm sorry Sylvia. That is just awful.
comment in response to
post
Yeah, it's a great example of statistical myths/rituals/heuristics.
comment in response to
post
Hmm. I disagree. So, I'll bite.
1. Sample size isn't just about "larger" it's about right size and understanding how much power you have. This is supported by openly reporting sample size determination and power analyses.
1/5
comment in response to
post
If you can’t produce the code/output you can’t prove you did the study. No exceptions. No patience for any arguments to contrary.
This is a low bar, but it’s not been implemented and could be easily.
3/8
comment in response to
post
But people are so allergic to what they misunderstand as post-hoc power, they fail to consider anything that their study might be able to inform anything beyond statistical significance.
comment in response to
post
3. Report whether effects smaller than your MDES could be meaningful.
If none of those apply, then maybe your study isn't worth reporting.
But it still might be worth doing! It could establish feasibility of recruitment, procedures, measurement, & analysis.
Still a worthy goal!
2/3
comment in response to
post
As an editor of a clinical journal charged with implementing TOP guidelines, and someone who has tried to be as open and transparent as I can be in my work since about 2017, I’d venture that I know the challenges of making these shifts better than most.
Here are some thoughts I have.
1/8
comment in response to
post
comment in response to
post
Other times they're quite big and broad and based on broad observations across disciplines. But those observations are often only barely connected at the empirical level (see lots of "we see parallels in behavior, self-report, and neurobiology") so it looks more rigorous than it is.
2/2
comment in response to
post
Agreed. I think both that we have too many theories for too many things. Sometimes the theories are really narrow and vague (theory dressed up as speculation).
1/2
comment in response to
post
You just don't appreciate my psychological jazz, baby!
comment in response to
post
I find study section invigorating. People are so smart, careful, and thoughtful about the grant applications. They really work hard to understand the proposals and evaluate them fairly and equitably. In my experience, anyway.
comment in response to
post
I find this one to be my go to. It's quite easy in terms of steps and the bagels come out great.
I did make bagel boards and use a pizza steel, but I think you can still make awesome bagels without either.
thia.codes/newbagels.html
comment in response to
post
Are you making or buying them? I've really gotten into making bagels lately and it is so satisfying
comment in response to
post
This feels like a hate crime. In that I hate I couldn't eat this.