Lightnews — Scholar-powered news

Paolo Crosetto

@paolocrosetto.bsky.social

2.7K followers 1.2K following 510 posts

Experimental & Behavioural economist INRAE Grenoble • President of the French Association of Experimental Economists • Scientific publishing measurement & reform • Experiments on food labeling - risk - choices • Rstats • Italian Food Police honorary member

Posts Media Videos Starter Packs

Pinned

Paolo Crosetto @paolocrosetto.bsky.social · Jun 15

Springer-Nature launched a series of "Discover" journals that closely mimic MDPI titles -- sharing *identical* journal names, and likely similar business model.

What is going on, and why researcher will - as always - fall for it?

A 🧵

the-strain-on-scientific-publishing.github.io/website/post...

Springer Nature Discovers MDPI – The Strain on Scientific Publishing

Home page for the paper ‘The Strain on Scientific Publishing’ by Mark A Hanson, Dan Brockington, Paolo Crosetto and Pablo Gomez Barreiro

the-strain-on-scientific-publishing.github.io

9 190 250

Reposted by Paolo Crosetto

Marin Dacos @marind.bsky.social · 2d

Finland Now Has a National Roadmap for Open Science and Research Funding avointiede.fi/en/news/finl...

Finland Now Has a National Roadmap for Open Science and Research Funding

avointiede.fi

8 16

Reposted by Paolo Crosetto

Ilya Kashnitsky @ikashnitsky.phd · 6d

bsky.app/profile/ikas...

The very special paper on special issues by @paolocrosetto.bsky.social and @hansonmark.bsky.social 👀

Ilya Kashnitsky @ikashnitsky.phd · Dec 22

I'm so happy to see this paper published in QSS!

When it comes to MDPI, many people in academia are still playing the lost game of allowing them the benefit of doubt.

This figure had been faithfully killing those ridiculous demagogic attempts for over a year now 🙌

#ScientificPublishing

1 2

Reposted by Paolo Crosetto

Ilya Kashnitsky @ikashnitsky.phd · 6d

Guardian's @scienceweekly.bsky.social podcast produced probably the best succinct discussion and overview of what's happening in #ScientificPublishing — a fully blown crisis, featuring @hansonmark.bsky.social co-author of a seminal paper on special issues 👇
🎙️ www.theguardian.com/science/audi...

Fraud, AI slop and huge profits: is science publishing broken? – podcast

Scientists warn academic publishing needs reform in order to retain trust in research system. Ian Sample talks to Madeleine Finlay and Dr Mark Hanson proposes potential solutions

www.theguardian.com

1 2 12

Reposted by Paolo Crosetto

Mark A. Hanson @hansonmark.bsky.social · 6d

See two really key papers

Butler et al.: publishers are profiteering immensely off scientific journals, and it has to stop. direct.mit.edu/qss/article/...

Hanson et al.: the strain scientists feel is real, & it's because too many papers coming from #PublishOrPerish pressures
#ResearchIntegrity

The oligopoly’s shift to open access: How the big five academic publishers profit from article processing charges

Abstract. We aim to estimate the total amount of article processing charges (APCs) paid to publish open access (OA) in journals controlled by the five large commercial publishers (Elsevier, Sage, Spri...

direct.mit.edu

3 6

Paolo Crosetto @paolocrosetto.bsky.social · 6d

timeo danaos et dona ferentes

Reposted by Paolo Crosetto

Mark A. Hanson @hansonmark.bsky.social · 6d

Thanks @iansample.bsky.social at @theguardian.com podcast for chatting #ScientificPublishing.

This pod follows from The Strain on Scientific Publishing & reports of publisher profit margins rivalling Google etc...

Paper: direct.mit.edu/qss/article/...

Pod: podcasts.apple.com/gb/podcast/s...

Fraud, AI slop and huge profits: is science publishing broken?

Podcast Episode · Science Weekly · 02/10/2025 · 18m

podcasts.apple.com

1 11 23

Paolo Crosetto @paolocrosetto.bsky.social · 16d

In the Alps we have Chartreuse, Dahu, snow, and the Fourth Winter Workshop in Behavioral and Experimental Economics!

sites.google.com/view/winterw...

Keynotes by Jutta Roosen (Munich) & Liam Delaney (LSE), a cozy relaxed atmosphere -- Jan 28th-30th, 2026, Villard de Lans [by Grenoble].

Join us!

Paolo Crosetto @paolocrosetto.bsky.social · 16d

Oral exams.

That's it, that's the solution.

Roxane Gay @roxanegay.bsky.social · 16d

I cannot tell you how grim the AI in higher ed situation is. Many of the students have completely surrendered to letting AI do their homework, badly, I might add. How do you fix this? Truly, what the hell do we do, beyond what grading can address, which isn't a solution?

Paolo Crosetto @paolocrosetto.bsky.social · 20d

The "worse neighbour" part might not age that well given what is going on south of Ontario

1 1

Reposted by Paolo Crosetto

Jamie Cummins @jamiecummins.bsky.social · 21d

Can large language models stand in for human participants?
Many social scientists seem to think so, and are already using "silicon samples" in research.

One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want.

THREAD 🧵

The threat of analytic flexibility in using large language models to simulate human data: A call to attention

Social scientists are now using large language models to create "silicon samples" - synthetic datasets intended to stand in for human respondents, aimed at revolutionising human subjects research. How...

arxiv.org

12 150 330

Paolo Crosetto @paolocrosetto.bsky.social · 22d

yes! and IF is *also* easy to game.

An argument I often heard, mostly from economists, when presenting the rent-extraction practices of MDPI & co: "but the IF will go down!".

Prices are sticky, the IF is way stickier. People who think that the IF helps to self-regulate the market are delusional.

1 3

Reposted by Paolo Crosetto

Professor Dr Richard S.J. Tol FREcon MAE @richardtol.bsky.social · 24d

All is not well at Elsevier richardtol.substack.com/p/fraud-and-...

Fraud and cover-up

richardtol.substack.com

7 46 78

Reposted by Paolo Crosetto

Joel Wood 🇨🇦 @joelwwood.bsky.social · 23d

Whoa! Elsevier fired @richardtol.bsky.social, longstanding Editor-in-Chief of the journal Energy Economics. Richard's side of the story should be ringing alarm bells. #EconSky

Fraud and cover-up
richardtol.substack.com/p/fraud-and-...

Fraud and cover-up

richardtol.substack.com

4 44 91

Reposted by Paolo Crosetto

Professor Dr Richard S.J. Tol FREcon MAE @richardtol.bsky.social · 23d

I once got an email from an irate author. Why can't I pay you to publish my paper? Editor X of Journal Y accepts my money! Why don't you?

I forwarded that email to the publisher.

The editor is still there.

4 15

Paolo Crosetto @paolocrosetto.bsky.social · Sep 7

Excellent thread on what is going on at academic journals, from the perspective of an editor.

Clearly highlights the strain (direct.mit.edu/qss/article/...) caused by quantitative incentives, and that publishers cannot be allies in solving the mess -- they are part of the problem & profit from it.

5 9

Reposted by Paolo Crosetto

Paolo Crosetto @paolocrosetto.bsky.social · Jun 15

Springer Nature Discovers MDPI – The Strain on Scientific Publishing

Home page for the paper ‘The Strain on Scientific Publishing’ by Mark A Hanson, Dan Brockington, Paolo Crosetto and Pablo Gomez Barreiro

the-strain-on-scientific-publishing.github.io

9 190 250

Paolo Crosetto @paolocrosetto.bsky.social · Sep 4

Sometimes you have to ease up. For me, this means exploiting the (still) long days and the fresh weather to head to the mountains.
In Grenoble, this is what 30min by car and 1h on foot gets you -- sunset over the Western Alps.
It's a real privilege to live in such a beautiful place.

The city of Grenoble is already in the shade and lights turned on after sunset, as we still enjoy the last bits of light on the Grand Rocher Belledonne peak in the French alps

Looking northwest from the Belledonne range, France, at sunset. The Lac du Bourget in the distance.

the peak of the "Grand Rocher" in the Belledonne range, France, at sunset

the Belledonne range at sunset, the moon rising behind it, in the foreground a high-altitude marsh

1 9

Reposted by Paolo Crosetto

Julia M. Rohrer @dingdingpeng.the100.ci · Aug 25

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...

Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities

Abstract
Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve.

A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals).

Illustrated are
1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals
2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and
3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.

49 280 940

Reposted by Paolo Crosetto

Robert Finger @robertfinger.bsky.social · Aug 31

Fellow of the European Association of Agricultural Economists🥳

I am beyond words and incredibly honored to receive the European Association of Agricultural Economists' (EAAE) Fellowship Award, the highest recognition in European agricultural economics.

eaae.org/ItemText.asp...

7 6 34

Paolo Crosetto @paolocrosetto.bsky.social · Aug 28

let's hope you are right. I have seen enough propaganda spin to fear the worst, but here's to hope :-)

Paolo Crosetto @paolocrosetto.bsky.social · Aug 27

I agree but at the same time it is easy to spin empty hands. "Look, your hands are empty because the evil foreign countries stopped shipping! We must hit them harder".

1 1

Paolo Crosetto @paolocrosetto.bsky.social · Aug 27

I guess they'll use this too to blame the foreign countries and to justify yet another round of nonsensical "reciprocal" tariffs. And the angry people will swallow it whole and be even more in support of said nonsense.

Paolo Crosetto @paolocrosetto.bsky.social · Aug 18

I teach R (wrangling & plotting), little to none specific econ content (bar the data used). The reality of coding is that AI is ubiquitous. So they need to be exposed to it and use it, but as an assistant not a substitute. the in-person no-AI exam part looms over them, discouraging the latter.

2 5

Reposted by Paolo Crosetto

mark brandt @mjbsp.bsky.social · Aug 18

Large Language Models Do Not Simulate Human Psychology

arxiv.org/pdf/2508.06950

9 69 190

Paolo Crosetto @paolocrosetto.bsky.social · Aug 18

Same here

2 2