Charly Marie, PhD
@charlymarie.bsky.social
Psychologist, working on the stigma of unemployment.
(Re)lecteur assidu d’Astérix.
https://charlymarie.github.io/
(Re)lecteur assidu d’Astérix.
https://charlymarie.github.io/
The more logistic regression I do, the more I like linear probability models 🙌
The more deep learning I do, the more I like logistic regression
November 3, 2025 at 10:48 PM
The more logistic regression I do, the more I like linear probability models 🙌
Great post, as always!
I'd just add that multicollinearity is *almost* never an issue... unless your predictors are so heavily correlated that it becomes statistically spaghetti (i.e., that it becomes algebraically impossible to disentangle their unique effects, leading to unstable estimates).
I'd just add that multicollinearity is *almost* never an issue... unless your predictors are so heavily correlated that it becomes statistically spaghetti (i.e., that it becomes algebraically impossible to disentangle their unique effects, leading to unstable estimates).
October 13, 2025 at 3:15 PM
Great post, as always!
I'd just add that multicollinearity is *almost* never an issue... unless your predictors are so heavily correlated that it becomes statistically spaghetti (i.e., that it becomes algebraically impossible to disentangle their unique effects, leading to unstable estimates).
I'd just add that multicollinearity is *almost* never an issue... unless your predictors are so heavily correlated that it becomes statistically spaghetti (i.e., that it becomes algebraically impossible to disentangle their unique effects, leading to unstable estimates).
Reposted by Charly Marie, PhD
Lecornu II : Matignon Drift
Lecornu 3
Lecornu 4Ever
Lecornu V Infinite
Lecornu Origins
The Lecornu Tales
Legend of Lecornu
Lecornu 3
Lecornu 4Ever
Lecornu V Infinite
Lecornu Origins
The Lecornu Tales
Legend of Lecornu
October 10, 2025 at 8:11 PM
Lecornu II : Matignon Drift
Lecornu 3
Lecornu 4Ever
Lecornu V Infinite
Lecornu Origins
The Lecornu Tales
Legend of Lecornu
Lecornu 3
Lecornu 4Ever
Lecornu V Infinite
Lecornu Origins
The Lecornu Tales
Legend of Lecornu
Poisson goes brrr.
But good old OLS is still my favourite.
But good old OLS is still my favourite.
For instance, yesterday I read a paper with a table describing participants' sickness absence days with a mean of 71 and SD = 88. Generating a random (gaussian) sample using these values produces ~20% participants with less than zero sick days.
September 22, 2025 at 1:31 PM
Poisson goes brrr.
But good old OLS is still my favourite.
But good old OLS is still my favourite.
Reposted by Charly Marie, PhD
This is when I realized: we’ve been here before! The garden of forking paths in well-trodden by us psychologists. And we already know that analytic flexibility can allow us to present basically any finding we want as significant.
So I wondered: how much flexibility can we get with silicon samples?
So I wondered: how much flexibility can we get with silicon samples?
September 18, 2025 at 7:56 AM
This is when I realized: we’ve been here before! The garden of forking paths in well-trodden by us psychologists. And we already know that analytic flexibility can allow us to present basically any finding we want as significant.
So I wondered: how much flexibility can we get with silicon samples?
So I wondered: how much flexibility can we get with silicon samples?
LLMs are incredibly powerful tools.
The only limitation is that, in order to unlock their full potential, it is necessary to verify the answer.
In other words, you have to be able to do what the LLM is doing.
The only limitation is that, in order to unlock their full potential, it is necessary to verify the answer.
In other words, you have to be able to do what the LLM is doing.
The thing that struck me most is that even if you take away the code-writing necessity, knowing what to ask the bot for requires so much expertise.
Steps like "These points are overplotted, can we fix that" or "Let's look at only the top 50 restaurants" are incredible non-obvious to beginners...
Steps like "These points are overplotted, can we fix that" or "Let's look at only the top 50 restaurants" are incredible non-obvious to beginners...
Hadley is demonstrating ggbot2 - an llm to help to transform a plot. As an analyst who has done endless iterations on a plot (can you make this gteen?), this is like a dream. #PositCONF2025
September 17, 2025 at 8:58 PM
LLMs are incredibly powerful tools.
The only limitation is that, in order to unlock their full potential, it is necessary to verify the answer.
In other words, you have to be able to do what the LLM is doing.
The only limitation is that, in order to unlock their full potential, it is necessary to verify the answer.
In other words, you have to be able to do what the LLM is doing.
Reposted by Charly Marie, PhD
This would be the right time to push for the creation of an ‘old-school code school,’ proudly advertising that it doesn’t use AI — betting on the idea that programmers who actually know how to program will soon be rare and worth their weight in gold.
September 13, 2025 at 10:23 AM
This would be the right time to push for the creation of an ‘old-school code school,’ proudly advertising that it doesn’t use AI — betting on the idea that programmers who actually know how to program will soon be rare and worth their weight in gold.
Reposted by Charly Marie, PhD
The new {marginaleffects} release for #RStats (0.30.0) comes with two new vignettes:
1. Speed up computation with automatic differentiation (often 10x gains) marginaleffects.com/bonus/perfor...
2. Power analyses with {marginaleffects} and {DeclareDesign}. marginaleffects.com/bonus/power....
1. Speed up computation with automatic differentiation (often 10x gains) marginaleffects.com/bonus/perfor...
2. Power analyses with {marginaleffects} and {DeclareDesign}. marginaleffects.com/bonus/power....
37 Performance – Model to Meaning
marginaleffects.com
September 13, 2025 at 6:37 PM
The new {marginaleffects} release for #RStats (0.30.0) comes with two new vignettes:
1. Speed up computation with automatic differentiation (often 10x gains) marginaleffects.com/bonus/perfor...
2. Power analyses with {marginaleffects} and {DeclareDesign}. marginaleffects.com/bonus/power....
1. Speed up computation with automatic differentiation (often 10x gains) marginaleffects.com/bonus/perfor...
2. Power analyses with {marginaleffects} and {DeclareDesign}. marginaleffects.com/bonus/power....
Reposted by Charly Marie, PhD
I am beyond excited to announce that ggplot2 4.0.0 has just landed on CRAN.
It's not every day we have a new major #ggplot2 release but it is a fitting 18 year birthday present for the package.
Get an overview of the release in this blog post and be on the lookout for more in-depth posts #rstats
It's not every day we have a new major #ggplot2 release but it is a fitting 18 year birthday present for the package.
Get an overview of the release in this blog post and be on the lookout for more in-depth posts #rstats
ggplot2 4.0.0
A new major version of ggplot2 has been released on CRAN. Find out what is new here.
www.tidyverse.org
September 11, 2025 at 11:20 AM
J’ai eu le plaisir d’intervenir à l'@unedic.bsky.social autour de la stigmatisation des "chômeurs" et des manières de composer (ou non) avec l’assignation sociale du chômage. Merci !
Un entretien prolonge la discussion : shorturl.at/5AFS4
(Avec notamment des bouts de @michaelzemmour.cpesr.fr)
Un entretien prolonge la discussion : shorturl.at/5AFS4
(Avec notamment des bouts de @michaelzemmour.cpesr.fr)
Charly Marie : « Le chômage est une expérience de vie majoritaire et normale en France »
Charly Marie, docteur en psychologie, a travaillé sur les ressorts psychologiques des perceptions des personnes au chômage. Ses travaux éclairent la manière dont se construisent les stéréotypes au cœur de la stigmatisation des personnes privées d’emploi et ses effets sur les chômeurs et chômeuses.
www.unedic.org
September 11, 2025 at 10:16 AM
J’ai eu le plaisir d’intervenir à l'@unedic.bsky.social autour de la stigmatisation des "chômeurs" et des manières de composer (ou non) avec l’assignation sociale du chômage. Merci !
Un entretien prolonge la discussion : shorturl.at/5AFS4
(Avec notamment des bouts de @michaelzemmour.cpesr.fr)
Un entretien prolonge la discussion : shorturl.at/5AFS4
(Avec notamment des bouts de @michaelzemmour.cpesr.fr)
I'm In This Photo and I Don't Like It
La TV japonaise qui nous apprend que les autruches aiment bien les chauves car la forme leur rappelle leurs œufs et elles les protègent instinctivement 😄
September 10, 2025 at 12:23 PM
I'm In This Photo and I Don't Like It
Here's a version that's a bit more Bluesky-and-Mobiles compatible.
September 9, 2025 at 10:02 AM
Here's a version that's a bit more Bluesky-and-Mobiles compatible.
@dingdingpeng.the100.ci and @vincentab.bsky.social wrote an amazing paper about marginal effects.
While preparing my advanced statistics and methods class, I (surprisingly) had fun reproducing their Figure 1, adapting it to their Example 1, and ploting the text and arrows directly using ggplot().
While preparing my advanced statistics and methods class, I (surprisingly) had fun reproducing their Figure 1, adapting it to their Example 1, and ploting the text and arrows directly using ggplot().
September 8, 2025 at 1:27 PM
@dingdingpeng.the100.ci and @vincentab.bsky.social wrote an amazing paper about marginal effects.
While preparing my advanced statistics and methods class, I (surprisingly) had fun reproducing their Figure 1, adapting it to their Example 1, and ploting the text and arrows directly using ggplot().
While preparing my advanced statistics and methods class, I (surprisingly) had fun reproducing their Figure 1, adapting it to their Example 1, and ploting the text and arrows directly using ggplot().
Reposted by Charly Marie, PhD
Okay, so you've crunched your numbers and got some awesome statistical models? Sometimes, just knowing "X predicts Y" isn't enough to really get to the juicy bits. That's where the cool post-hoc stuff comes in – think estimated marginal means, contrasts, pairwise comparisons, or #marginaleffects.
August 31, 2025 at 8:27 AM
Okay, so you've crunched your numbers and got some awesome statistical models? Sometimes, just knowing "X predicts Y" isn't enough to really get to the juicy bits. That's where the cool post-hoc stuff comes in – think estimated marginal means, contrasts, pairwise comparisons, or #marginaleffects.
@dingdingpeng.the100.ci and @vincentab.bsky.social are the coolest kids in town — again!
Ever stared at a table of regression coefficients & wondered what you're doing with your life?
Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
August 25, 2025 at 1:58 PM
@dingdingpeng.the100.ci and @vincentab.bsky.social are the coolest kids in town — again!
Open science: Pre-publish your work!
Journal editor: Anonymize your pre-printed manuscript for peer review!
psyarXiv: Incomplete submission. Authors not listed on paper. Pre-print has been withdrawn.
Journal editor: Anonymize your pre-printed manuscript for peer review!
psyarXiv: Incomplete submission. Authors not listed on paper. Pre-print has been withdrawn.
a man with his eyes closed looks at the camera
ALT: a man with his eyes closed looks at the camera
media.tenor.com
August 22, 2025 at 5:33 PM
Open science: Pre-publish your work!
Journal editor: Anonymize your pre-printed manuscript for peer review!
psyarXiv: Incomplete submission. Authors not listed on paper. Pre-print has been withdrawn.
Journal editor: Anonymize your pre-printed manuscript for peer review!
psyarXiv: Incomplete submission. Authors not listed on paper. Pre-print has been withdrawn.
Reposted by Charly Marie, PhD
@gralefrit.bsky.social You have to see these! A friend of mine has been collating a series of images entitled "Things my students have written turned into Philomena Cunk quotes" and they're just glorious.
August 11, 2025 at 11:12 AM
@gralefrit.bsky.social You have to see these! A friend of mine has been collating a series of images entitled "Things my students have written turned into Philomena Cunk quotes" and they're just glorious.
Reposted by Charly Marie, PhD
The landmark Handbook of Social Psychology's sixth edition is now available - and open access for the first time! It features 50 chapters from 50+ scholars.
Learn more + download it today: ow.ly/RPjn50VV8bT
Learn more + download it today: ow.ly/RPjn50VV8bT
May 19, 2025 at 6:15 PM
The landmark Handbook of Social Psychology's sixth edition is now available - and open access for the first time! It features 50 chapters from 50+ scholars.
Learn more + download it today: ow.ly/RPjn50VV8bT
Learn more + download it today: ow.ly/RPjn50VV8bT
Reposted by Charly Marie, PhD
This new package for mediation is great! The user interface is *so intuitive* but the unsung hero of this software is the documentation:
May 6, 2025 at 2:53 PM
This new package for mediation is great! The user interface is *so intuitive* but the unsung hero of this software is the documentation:
« Esprit chagrin », c’est beau comme du Mylene Farmer
A propos des CPJ « contrairement à ce que laissent entendre les esprits chagrins et malthusiens, il n’y a pas d’effet d’éviction »
Bah ouais, parce que constater que le nombre de postes ouverts hors CPJ chute malgré la hausse des départs à la retraite, c'est être « esprit chagrin », tu vois.
Bah ouais, parce que constater que le nombre de postes ouverts hors CPJ chute malgré la hausse des départs à la retraite, c'est être « esprit chagrin », tu vois.
L’entrée dans la carrière des enseignants-chercheurs – 2025
Différents indicateurs relatifs à l'entrée dans la carrière des EC.
cpesr.fr
May 6, 2025 at 9:39 PM
« Esprit chagrin », c’est beau comme du Mylene Farmer
Reposted by Charly Marie, PhD
🧵New paper out in Cognition
Why do people moralize harmless carnal sins (e.g. gluttony, masturbation)?
@danielnettle.bsky.social & I find that these behaviors activate reciprocity-based moral judgment—no need for a distinct "purity" module.
50 days free link: authors.elsevier.com/a/1l18D2Hx2-...
Why do people moralize harmless carnal sins (e.g. gluttony, masturbation)?
@danielnettle.bsky.social & I find that these behaviors activate reciprocity-based moral judgment—no need for a distinct "purity" module.
50 days free link: authors.elsevier.com/a/1l18D2Hx2-...
May 6, 2025 at 9:45 AM
🧵New paper out in Cognition
Why do people moralize harmless carnal sins (e.g. gluttony, masturbation)?
@danielnettle.bsky.social & I find that these behaviors activate reciprocity-based moral judgment—no need for a distinct "purity" module.
50 days free link: authors.elsevier.com/a/1l18D2Hx2-...
Why do people moralize harmless carnal sins (e.g. gluttony, masturbation)?
@danielnettle.bsky.social & I find that these behaviors activate reciprocity-based moral judgment—no need for a distinct "purity" module.
50 days free link: authors.elsevier.com/a/1l18D2Hx2-...
Reposted by Charly Marie, PhD
It's bonkers that in academic science we spend hours obsessing over the wording of a paper, but often only one person has seen the code that produced the results! 2/N
April 16, 2025 at 1:23 PM
It's bonkers that in academic science we spend hours obsessing over the wording of a paper, but often only one person has seen the code that produced the results! 2/N
Reposted by Charly Marie, PhD
Pour les petits besoins de sa com' personnelle, M. Macron est en train d'engager la responsabilité et la réputation de la France, mais sans mettre les moyens, et en ciblant l'inutile.
En vouloir à la France va être tout à fait légitime. Et c'est très grave.
En vouloir à la France va être tout à fait légitime. Et c'est très grave.
The End End GIF
ALT: The End End GIF
media.tenor.com
April 20, 2025 at 11:14 AM
Pour les petits besoins de sa com' personnelle, M. Macron est en train d'engager la responsabilité et la réputation de la France, mais sans mettre les moyens, et en ciblant l'inutile.
En vouloir à la France va être tout à fait légitime. Et c'est très grave.
En vouloir à la France va être tout à fait légitime. Et c'est très grave.
Reposted by Charly Marie, PhD
Je suis très en colère de lire "Ça ne dit rien de la souffrance quand on est down"
Non, avoir une phase dépressive ou maniaque ne veut pas forcément dire que l'"inconscient" a quelque chose à "révéler"
Il n'y a pas forcément de trauma caché derrière une bipolarité!
Non, avoir une phase dépressive ou maniaque ne veut pas forcément dire que l'"inconscient" a quelque chose à "révéler"
Il n'y a pas forcément de trauma caché derrière une bipolarité!
April 12, 2025 at 8:31 PM
Je suis très en colère de lire "Ça ne dit rien de la souffrance quand on est down"
Non, avoir une phase dépressive ou maniaque ne veut pas forcément dire que l'"inconscient" a quelque chose à "révéler"
Il n'y a pas forcément de trauma caché derrière une bipolarité!
Non, avoir une phase dépressive ou maniaque ne veut pas forcément dire que l'"inconscient" a quelque chose à "révéler"
Il n'y a pas forcément de trauma caché derrière une bipolarité!
Reposted by Charly Marie, PhD
From Experienced Stigma to Job Access of Unemployed People: A Longitudinal Study of the Mediating Role of Anticipated Stigma, Internalized Stigma, Self-Efficacy, and Job Search Behavior: https://osf.io/nsdka
April 3, 2025 at 1:04 PM
From Experienced Stigma to Job Access of Unemployed People: A Longitudinal Study of the Mediating Role of Anticipated Stigma, Internalized Stigma, Self-Efficacy, and Job Search Behavior: https://osf.io/nsdka