Mattan S. Ben-Shachar
@mattansb.msbstats.info
Statistics lecturer | Freelance statistical consultant & research analyst | #rstats dev @easystats.github.io
home.msbstats.info
(He/Him)
home.msbstats.info
(He/Him)
Pinned
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!
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!
Reposted by Mattan S. Ben-Shachar
Look all I’m saying is that Cursor has access to a million lifetimes of knowledge, so the fact that *I* had to tell it how to incrementally try my code piece by piece to figure out where the error is coming from after it went in circles for 20 minutes trying to debug on its own….😒
a woman in a black dress is looking at a man in a room .
ALT: a woman in a black dress is looking at a man in a room .
media.tenor.com
February 17, 2026 at 1:37 AM
Look all I’m saying is that Cursor has access to a million lifetimes of knowledge, so the fact that *I* had to tell it how to incrementally try my code piece by piece to figure out where the error is coming from after it went in circles for 20 minutes trying to debug on its own….😒
Reposted by Mattan S. Ben-Shachar
If you write a paper, you OWN it.
If you can't account for all aspects of the paper, that's on you. It's not complicated.
Agree that current human authorship is also complicated!
If you can't account for all aspects of the paper, that's on you. It's not complicated.
Agree that current human authorship is also complicated!
I've heard this position articulated by a lot of people, but I think this is hard to make precise. Much of this tracks a problem we already know in science: standards of authorship are not stable and there are many difficult edge cases.
Consider these cases...
Consider these cases...
Seeing ppl debate how to deal with paper submissions that are partly written by “AI”. Why are we having that conversation?
I would refuse to review or edit any paper that was not 100% written by its authors (i.e., the people who have authorial responsibility for the creation of a papers content).
I would refuse to review or edit any paper that was not 100% written by its authors (i.e., the people who have authorial responsibility for the creation of a papers content).
February 16, 2026 at 9:46 PM
If you write a paper, you OWN it.
If you can't account for all aspects of the paper, that's on you. It's not complicated.
Agree that current human authorship is also complicated!
If you can't account for all aspects of the paper, that's on you. It's not complicated.
Agree that current human authorship is also complicated!
Reposted by Mattan S. Ben-Shachar
Starting to dig into updates for my empirically supported treatments course for next year (will be an undergrad and a PhD version)
Want to expand on mechanisms of psychotherapy (emo reg, stress generation, physical activity, social, etc)
Any suggestions for papers of comprehensive frameworks?
Want to expand on mechanisms of psychotherapy (emo reg, stress generation, physical activity, social, etc)
Any suggestions for papers of comprehensive frameworks?
February 16, 2026 at 6:23 PM
Starting to dig into updates for my empirically supported treatments course for next year (will be an undergrad and a PhD version)
Want to expand on mechanisms of psychotherapy (emo reg, stress generation, physical activity, social, etc)
Any suggestions for papers of comprehensive frameworks?
Want to expand on mechanisms of psychotherapy (emo reg, stress generation, physical activity, social, etc)
Any suggestions for papers of comprehensive frameworks?
In this post, Gelman gives 4 "sources" or "types" of priors.
Can anyone shed some light in the 4th type? "Prior as unconditional distribution"
#stats
statmodeling.stat.columbia.edu/2025/05/21/p...
Can anyone shed some light in the 4th type? "Prior as unconditional distribution"
#stats
statmodeling.stat.columbia.edu/2025/05/21/p...
Prior as data, prior as belief, prior as soft constraint, prior as unconditional distribution in a generative model | Statistical Modeling, Causal Inference, and Social Science
statmodeling.stat.columbia.edu
February 16, 2026 at 3:12 PM
In this post, Gelman gives 4 "sources" or "types" of priors.
Can anyone shed some light in the 4th type? "Prior as unconditional distribution"
#stats
statmodeling.stat.columbia.edu/2025/05/21/p...
Can anyone shed some light in the 4th type? "Prior as unconditional distribution"
#stats
statmodeling.stat.columbia.edu/2025/05/21/p...
Reposted by Mattan S. Ben-Shachar
Picard management tip: Humility without confidence is sheepishness. Confidence without humility is arrogance.
February 15, 2026 at 4:04 PM
Picard management tip: Humility without confidence is sheepishness. Confidence without humility is arrogance.
Reposted by Mattan S. Ben-Shachar
I find it baffling that people want AI to summarise papers for them. The point is to read the papers in order to become a domain expert. Better tools for education are great. But you need a little friction to learn. Also why tests and assessment are important.
February 11, 2026 at 9:16 PM
I find it baffling that people want AI to summarise papers for them. The point is to read the papers in order to become a domain expert. Better tools for education are great. But you need a little friction to learn. Also why tests and assessment are important.
I'll take an ugly excel plot made from actual data over a beautiful AI generated figure kinda sorta based on data - any day!
Why plot actual data, when you can have PaperBanana vibe plot your data - beautifully?
Why read papers when DeepResearch can write a summary or a whole lit review?
Why not upload a CSV file and bot generate a professional looking report for you?
Why read papers when DeepResearch can write a summary or a whole lit review?
Why not upload a CSV file and bot generate a professional looking report for you?
February 14, 2026 at 6:55 AM
I'll take an ugly excel plot made from actual data over a beautiful AI generated figure kinda sorta based on data - any day!
Vibe research and vibe science have always been bad.
AI now makes them easier than ever to do, in new ways. It's so tempting. I'm not even talking about pure slop.
I've overheard conversations between grad students that should make us all uneasy... 😟
AI now makes them easier than ever to do, in new ways. It's so tempting. I'm not even talking about pure slop.
I've overheard conversations between grad students that should make us all uneasy... 😟
February 14, 2026 at 6:53 AM
Vibe research and vibe science have always been bad.
AI now makes them easier than ever to do, in new ways. It's so tempting. I'm not even talking about pure slop.
I've overheard conversations between grad students that should make us all uneasy... 😟
AI now makes them easier than ever to do, in new ways. It's so tempting. I'm not even talking about pure slop.
I've overheard conversations between grad students that should make us all uneasy... 😟
How I feel about random intercept models 😨
February 14, 2026 at 5:30 AM
How I feel about random intercept models 😨
Reposted by Mattan S. Ben-Shachar
#statstab #485 Bayesian ANCOVA and the ATE
Thoughts: Still grappling with the implications of using the causal inference approach to randomized experiments. But it's interesting.
#ATE #causalinference #ancova #ANOVA #rstats #estimand #counterfactuals
solomonkurz.netlify.app/blog/2025-07...
Thoughts: Still grappling with the implications of using the causal inference approach to randomized experiments. But it's interesting.
#ATE #causalinference #ancova #ANOVA #rstats #estimand #counterfactuals
solomonkurz.netlify.app/blog/2025-07...
February 13, 2026 at 4:58 PM
#statstab #485 Bayesian ANCOVA and the ATE
Thoughts: Still grappling with the implications of using the causal inference approach to randomized experiments. But it's interesting.
#ATE #causalinference #ancova #ANOVA #rstats #estimand #counterfactuals
solomonkurz.netlify.app/blog/2025-07...
Thoughts: Still grappling with the implications of using the causal inference approach to randomized experiments. But it's interesting.
#ATE #causalinference #ancova #ANOVA #rstats #estimand #counterfactuals
solomonkurz.netlify.app/blog/2025-07...
Reposted by Mattan S. Ben-Shachar
Where do parents find the time to parent? Less sleep, work and screens.
Amazing chart feat. in @alphaville.ft.com Further Reading.
homeeconomics.substack.com/p/where-do-p...
Amazing chart feat. in @alphaville.ft.com Further Reading.
homeeconomics.substack.com/p/where-do-p...
February 13, 2026 at 8:11 AM
Where do parents find the time to parent? Less sleep, work and screens.
Amazing chart feat. in @alphaville.ft.com Further Reading.
homeeconomics.substack.com/p/where-do-p...
Amazing chart feat. in @alphaville.ft.com Further Reading.
homeeconomics.substack.com/p/where-do-p...
Reposted by Mattan S. Ben-Shachar
Considering the following table:
| Sig | Non|
Prec | A | B |
Impr | C | D |
where Sig = significant, Non = nonsignifcant, Prec = precise, Impr = imprecise
The paper seems to say "too much A and C; more B and D!" but I would say "too much C; less C and more B!"
| Sig | Non|
Prec | A | B |
Impr | C | D |
where Sig = significant, Non = nonsignifcant, Prec = precise, Impr = imprecise
The paper seems to say "too much A and C; more B and D!" but I would say "too much C; less C and more B!"
February 11, 2026 at 10:40 PM
Considering the following table:
| Sig | Non|
Prec | A | B |
Impr | C | D |
where Sig = significant, Non = nonsignifcant, Prec = precise, Impr = imprecise
The paper seems to say "too much A and C; more B and D!" but I would say "too much C; less C and more B!"
| Sig | Non|
Prec | A | B |
Impr | C | D |
where Sig = significant, Non = nonsignifcant, Prec = precise, Impr = imprecise
The paper seems to say "too much A and C; more B and D!" but I would say "too much C; less C and more B!"
Reposted by Mattan S. Ben-Shachar
Compositional data (proportions that sum to 1) behave in ways standard models aren’t built for
I walk through why Dirichlet regression is often the right tool & what extra insight it gives using a real ex of eyetracking
#Dirichlet #r #brms #guide #eyetracking
open.substack.com/pub/mzlotean...
I walk through why Dirichlet regression is often the right tool & what extra insight it gives using a real ex of eyetracking
#Dirichlet #r #brms #guide #eyetracking
open.substack.com/pub/mzlotean...
February 9, 2026 at 4:05 PM
Compositional data (proportions that sum to 1) behave in ways standard models aren’t built for
I walk through why Dirichlet regression is often the right tool & what extra insight it gives using a real ex of eyetracking
#Dirichlet #r #brms #guide #eyetracking
open.substack.com/pub/mzlotean...
I walk through why Dirichlet regression is often the right tool & what extra insight it gives using a real ex of eyetracking
#Dirichlet #r #brms #guide #eyetracking
open.substack.com/pub/mzlotean...
Reposted by Mattan S. Ben-Shachar
The Perspectives on Scientific Error (#PSE8) conference was great - I will be coming back.
It was nice meeting so many new people, and getting the opportunity to discuss statistical, philosophical, and meta-scientific issues of scientific error.
It was nice meeting so many new people, and getting the opportunity to discuss statistical, philosophical, and meta-scientific issues of scientific error.
February 13, 2026 at 9:04 AM
The Perspectives on Scientific Error (#PSE8) conference was great - I will be coming back.
It was nice meeting so many new people, and getting the opportunity to discuss statistical, philosophical, and meta-scientific issues of scientific error.
It was nice meeting so many new people, and getting the opportunity to discuss statistical, philosophical, and meta-scientific issues of scientific error.
Reposted by Mattan S. Ben-Shachar
#rstats tidypolars 0.17.0 is available!
tidypolars provides the tidyverse syntax while using polars for better perf.
In this release:
- support new functions from dplyr 1.2.0 (filter_out, when_any...)
- pivot_wider with lazyframe
- bug fixes
and more
News: tidypolars.etiennebacher.com/news/
tidypolars provides the tidyverse syntax while using polars for better perf.
In this release:
- support new functions from dplyr 1.2.0 (filter_out, when_any...)
- pivot_wider with lazyframe
- bug fixes
and more
News: tidypolars.etiennebacher.com/news/
Changelog
tidypolars.etiennebacher.com
February 12, 2026 at 12:57 PM
#rstats tidypolars 0.17.0 is available!
tidypolars provides the tidyverse syntax while using polars for better perf.
In this release:
- support new functions from dplyr 1.2.0 (filter_out, when_any...)
- pivot_wider with lazyframe
- bug fixes
and more
News: tidypolars.etiennebacher.com/news/
tidypolars provides the tidyverse syntax while using polars for better perf.
In this release:
- support new functions from dplyr 1.2.0 (filter_out, when_any...)
- pivot_wider with lazyframe
- bug fixes
and more
News: tidypolars.etiennebacher.com/news/
Reposted by Mattan S. Ben-Shachar
Once you start thinking about implausibly large effect sizes, you can't stop spotting them around you or wondering how others aren't doing so.
Harkin et al.'s (2016) meta analysis has 740+ citations, but it reports Cohen d values as large as 14. 17 cases of d>4.
psycnet.apa.org/record/2015-...
Harkin et al.'s (2016) meta analysis has 740+ citations, but it reports Cohen d values as large as 14. 17 cases of d>4.
psycnet.apa.org/record/2015-...
APA PsycNet
psycnet.apa.org
February 12, 2026 at 3:06 PM
Once you start thinking about implausibly large effect sizes, you can't stop spotting them around you or wondering how others aren't doing so.
Harkin et al.'s (2016) meta analysis has 740+ citations, but it reports Cohen d values as large as 14. 17 cases of d>4.
psycnet.apa.org/record/2015-...
Harkin et al.'s (2016) meta analysis has 740+ citations, but it reports Cohen d values as large as 14. 17 cases of d>4.
psycnet.apa.org/record/2015-...
February 10, 2026 at 1:37 PM
Funny and educational! Double whammy!
February 9, 2026 at 8:22 PM
Funny and educational! Double whammy!
Reposted by Mattan S. Ben-Shachar
Chapter 23 openeclass.panteion.gr/modules/docu...
openeclass.panteion.gr
July 18, 2025 at 1:50 PM
Chapter 23 openeclass.panteion.gr/modules/docu...
No surprises here
February 9, 2026 at 6:16 AM
No surprises here
"Is my sample large enough for Bayesian statistics?" is a weird question coming from someone who only ever uses frequentist stats that are only approximately correct for large samples.
February 8, 2026 at 3:26 PM
"Is my sample large enough for Bayesian statistics?" is a weird question coming from someone who only ever uses frequentist stats that are only approximately correct for large samples.
This show was too good.
Past/Present/Future
YouTube video by Brandon Roberts - Topic
www.youtube.com
February 8, 2026 at 1:04 PM
This show was too good.
Reposted by Mattan S. Ben-Shachar
The role of perceived sandwich origin on gustatory satisfaction: a study protocol – Ben Harrap
benharrap.com
February 8, 2026 at 9:15 AM
Reposted by Mattan S. Ben-Shachar
but why is the save icon a vending machine
February 8, 2026 at 3:17 AM
but why is the save icon a vending machine
Reposted by Mattan S. Ben-Shachar
oooooooof
February 8, 2026 at 3:41 AM
oooooooof