Mattan S. Ben-Shachar 🎗️🇮🇱🇺🇦
@mattansb.msbstats.info
3.1K followers 320 following 1.1K posts
Statistics lecturer | Freelance statistical consultant & research analyst | #rstats dev @easystats.github.io home.msbstats.info (He/Him)
Posts Media Videos Starter Packs
Pinned
mattansb.msbstats.info
1k followers! Calls for a re-introduction:

Statistics lecturer, freelance stats consultant, & #rstats dev @easystats.bsky.social 📊

I try to help social scientists make better inferences from their data & communicate their findings 👨‍🏫

Hope to bring the #stats twitter/R community vibes over to bsky!
mattansb.msbstats.info
Yes, see second graph here:
bignardi.bsky.social
We ran 3 simulation studies to evaluate RMU's bias, accuracy and coverage (% of times the 95% credible intervals included the right answer) across linear factor, signal detection theory and reinforcement learning models. Results were good-to-great across studies and simulation conditions.
Reposted by Mattan S. Ben-Shachar 🎗️🇮🇱🇺🇦
bignardi.bsky.social
New preprint with @rogierk.bsky.social @paulbuerkner.com - we introduce "relative measurement uncertainty" - a reliability estimation method that's applicable across a broad class of Bayesian measurement models (e.g., generative-, computational- and item response theory-models osf.io/h54k8
OSF
osf.io
mattansb.msbstats.info
The paper gives SDT as an example. What's new is the ability to get uncertainty about the reliability for "free".
mattansb.msbstats.info
This is super interesting!

I would be interested in seeing how this can also be applied to designs where ICCs and other G coefficients (e.g. Rc, R1Rand other for repeated measures) and used.

ICCs
doi.org/10.1037/met0...

Retreated measures G coefficients:
cran.r-project.org/web/packages...
APA PsycNet
doi.org
mattansb.msbstats.info
A whole course for CFA! We have that as part of a broad multivariate course (CFA, SEM, EFA, missing data, ...)

What topics do you cover?
mattansb.msbstats.info
Well, now that I've been properly primmed...
Reposted by Mattan S. Ben-Shachar 🎗️🇮🇱🇺🇦
arynn.bsky.social
AI company alignment chart.
mattansb.msbstats.info
Isn't that also data driven variable selection?
mattansb.msbstats.info
Angering everyone? Now THAT'S being a Bayesian!
mattansb.msbstats.info
There was an excellent tweet years ago that I can't find, someone announced their departure:

"I'm leaving academia to do the things I love most: research and teaching"
mattansb.msbstats.info
I think that in that regard, perhaps there's a difference between """applied""" and """pure science""" research?
mattansb.msbstats.info
Interestingly, like Mike, I also still live in the academic (or academia adjacent) space doing what I really like - teaching!
mattansb.msbstats.info
This seems to echo a lot of the fears I had when I decided not to pursue an academic career.

Glad I wasn't off.
But also sad I wasn't off.
mikexcohen.bsky.social
Why I left academia and neuroscience.

This post on Substack has gained a lot of traction. I think many people identify with it.
(Most of my posts are technical tutorials on machine-learning and LLM-mechanisms.)
mikexcohen.substack.com/p/why-i-left...
Why I left academia and neuroscience
Don't worry, this isn't yet another story of rage-quitting.
mikexcohen.substack.com
Reposted by Mattan S. Ben-Shachar 🎗️🇮🇱🇺🇦
mikexcohen.bsky.social
Why I left academia and neuroscience.

This post on Substack has gained a lot of traction. I think many people identify with it.
(Most of my posts are technical tutorials on machine-learning and LLM-mechanisms.)
mikexcohen.substack.com/p/why-i-left...
Why I left academia and neuroscience
Don't worry, this isn't yet another story of rage-quitting.
mikexcohen.substack.com
Reposted by Mattan S. Ben-Shachar 🎗️🇮🇱🇺🇦
konsta.happonen.eu
Relatedly, ages ago @betanalpha.bsky.social wrote "Remember that using Bayes' Theorem doesn't make you a Bayesian. Quantifying uncertainty with probability makes you a Bayesian". I still think about this regularly.
Reposted by Mattan S. Ben-Shachar 🎗️🇮🇱🇺🇦
mzloteanu.bsky.social
#statstab #432 PsychOpen Gold

Thoughts: instead of submitting to greedy and unhelpful publishers, try this list of fully open and free journals in psychology.

#OpenScience #openaccess #apcs #goldaccess #pedagogy

psychopen.eu
PsychOpen GOLD: Open Access Publishing
We are a Diamond Open Access platform for psychology research. Peer-reviewed, free-to-read journals with no publication fees, promoting open science.
psychopen.eu
mattansb.msbstats.info
Running, hanging out here.
I should read things unrelated to work...
mattansb.msbstats.info
Idk, it's still an estimate of the conditional sd across (pooled? What is language?) all sub populations captured by the model. I agree this is a bad idea in terms of "what standardized effect size is interesting to inspect".
mattansb.msbstats.info
I feel like that's the "extension" part of what I said. Are we not talking about the same thing?
mattansb.msbstats.info
The (pooled) sd in Cohen's d can also be thought of as the residual sd when predicting y ~ group.

Compare:
m <- lm(mpg ~ am, data = mtcars)
coef(m)[2] / sigma(m)

effectsize::cohens_d(mpg ~ am, data = mtcars)