Grilled Cheese

ExploreLog inSign up

Berna Devezer

@devezer.bsky.social

999 Following11k Followers

Metascientist @ uidaho. Behavioral sciences, stats, modeling, phil of science. Cat slave, cook, sailor (see: https://bsky.app/profile/kumrusails.bsky.social), photographer, clarinet student, adventurer, nature lover. Allergic to unsolicited advice.

PostsRepliesMedia
@devezer.bsky.socialApr 30, 2026, 1:37 AM Pinned

πŸ“£ The difference between replicable and not replicable is not itself scientifically replicable. πŸ“£

New work with Erkan Buzbas, showing that verdicts such as "X% of results replicated" are based on an inferential machinery that doesn't work.

arxiv.org/abs/2604.26268

Screenshot of a paper's title page. Title: "The Difference Between 'Replicable' and 'Not replicable' is not Itself Scientifically Replicable". Authors: Berna Devezer and Erkan O. Buzbas, both at the University of Idaho β€” Devezer in the Department of Business and the Institute for Modeling Collaboration and Innovation, Buzbas in the Department of Mathematics and Statistical Science. Authors contributed equally. Corresponding author: bdevezer@uidaho.edu.
Abstract: Replication studies estimate the replicability rate of scientific results by aggregating binary verdicts of experiments. Exact replications are rarely attainable, so most replication sequences are non-exact. Experiments differ in ways that matter and do not share a single common data-generating process. We formalize two statistical interpretations of this non-exactness. In a shared latent rate model (benchmark), experiments are exchangeable and depend on a common random replicability rate. In a conditionally independent rates model (operational), each experiment has its own replicability rate drawn independently from a population distribution. Under the shared latent rate model, even small variability among replicability rates induces an irreducible variance floor on the estimated mean replicability rate that cannot be eliminated by adding more replications. Under the conditionally independent rates model, the degree of non-exactness is not identifiable from standard replication data, because one binary verdict per experiment contains no information about between-experiment heterogeneity. Researchers therefore cannot tell which precision regime they are operating in or whether high- and low-replicability sequences can be distinguished in principle. As a result, the usual data structure of one binary verdict per experiment cannot support reliable demarcation between "replicable" and "not replicable" results and systematically understates uncertainty, making high- and low-replicability sequences appear discrim…
15
@devezer.bsky.socialOct 12, 2026, 12:10 AM

It really is!!

@devezer.bsky.socialOct 11, 2026, 11:56 PM

was craving mom's yeast-free sheet pan pizza so asked her for a recipe and went right on to make it

Pizza in a rectangular sheet pan, browned on top with bits of black and green olives, turkey franks, peppers, and corn sticking out of the melted cheese
@saklein.bsky.socialOct 11, 2026, 7:01 PMReposted by @devezer.bsky.social

We should take the pressure off of that particular staging of writing and writerliness. Instead, we can look for thinking elsewhere; everywhere. I think that's how we reclaim thinking for ourselves.

@saklein.bsky.socialOct 11, 2026, 7:01 PMReposted by @devezer.bsky.social

If we want to help people find, cultivate, and reclaim intrinsic pleasure and power in thinking, it is helpful (and less elitist) to recognize how thinking is materialized through, with, and across many activities, practices and media.

@devezer.bsky.socialOct 11, 2026, 5:57 PM

so elegant, so demure

@devezer.bsky.socialOct 11, 2026, 4:38 PM

I find such diversity of experiences exciting as well. Wish I could read a good exploration about how this leads to different internal processes and different outputs (or not), and in what ways.

@devezer.bsky.socialOct 11, 2026, 4:17 PM

Oh wow I find the start of it a little too assertive and overly generalized hmmm

@devezer.bsky.socialOct 11, 2026, 4:15 PM

I tend to think the quality of writing may be orthogonal. I do believe it really is how some (many) people think through things. It's just not a universal fact. I find that the act of writing is rather disruptive to my thinking. Human experience varies a lot.

@devezer.bsky.socialOct 11, 2026, 4:12 PM

I think at this point the anti-AI hype has gotten quite as powerful as well so we keep seeing a lot of poorly designed studies and overhyped results doing the rounds because they confirm existing beliefs.

@devezer.bsky.socialOct 11, 2026, 3:56 PM

I had read it as note down anyway lol

@devezer.bsky.socialOct 11, 2026, 3:54 PM

πŸ˜‚πŸ˜­

@devezer.bsky.socialOct 11, 2026, 3:51 PM

I could use some of that right about now for my wrist

@devezer.bsky.socialOct 11, 2026, 3:44 PM

Oh I guess I don't. I find thinking a very distinct process from writing.

@devezer.bsky.socialOct 11, 2026, 3:39 PM

It's nice to be able to organize it. Mine just happens. I let it be.

@devezer.bsky.socialOct 11, 2026, 3:25 PM

My previous musings on the topic

@devezer.bsky.socialOct 11, 2026, 3:25 PM

Once again I find that I think better when I'm not writing. When I need to write something, I think about it throughout the day, while chopping onions, trying to fall asleep, reading something unrelated... It's only then that I know what to write. "Writing is thinking" is decidedly not me.

@devezer.bsky.socialOct 11, 2026, 3:21 PM

So relatable

@devezer.bsky.socialOct 11, 2026, 3:03 PM

Awwwww love that. Goes well with the yelling πŸ˜πŸ˜‚

@devezer.bsky.socialOct 11, 2026, 3:02 PM

πŸ’ž It's great when that happens, no? Still, Random used to give me attitude for days on our return. Pseudo is always absolutely nonchalant.

@devezer.bsky.socialOct 11, 2026, 3:00 PM

Oh no, you're in trouble!

Older posts
Terms of UsePrivacy PolicyCommunity StandardsHelpGet the app

Grilled Cheese is a product of Village Compute

Version devBuilt at: 2026-10-11 19:07:23 EDT