Lightnews — Scholar-powered news

Marek Vranka @mvranka.bsky.social · 12d

yeah, but also "... —or if you need to do it to pay the bills." 🙃

Julia M. Rohrer @dingdingpeng.the100.ci · 12d

“Doing research is a choice, and unless you’re involved in some urgent project—curing a disease or winning a war or righting some injustice or raising living standards or whatever—or some interesting project—baseball statistics or the theory of random walks or whatever—you shouldn’t do it.”

Andrew Gelman et al. @statmodeling.bsky.social · 14d

More on the decline and fall of Steven Levitt
statmodeling.stat.columbia.edu/2025/09/24/m...

1 1

Reposted by Marek Vranka

Ekaterina Pronizius @epronizius.bsky.social · 13d

This is the largest cross-linguistic study on SP to date that used computationally selected stimuli and a continuous lexical decision task. The results reveal consistent & robust priming effects across both Latin & non-Latin languages, pointing to a generalizable cognitive mechanism of SP.

a woman wearing a plaid shirt and a hat says we speak the same language

ALT: a woman wearing a plaid shirt and a hat says we speak the same language

media.tenor.com

1 1 2

Reposted by Marek Vranka

Andrew Gelman et al. @statmodeling.bsky.social · 13d

“On the poor statistical properties of the P-curve meta-analytic procedure”
statmodeling.stat.columbia.edu/2025/09/25/o...

“On the poor statistical properties of the P-curve meta-analytic procedure” | Statistical Modeling, Causal Inference, and Social Science

statmodeling.stat.columbia.edu

2 9 27

Marek Vranka @mvranka.bsky.social · 14d

we can even play bingo with this thread bsky.app/profile/ianh... by @ianhussey.mmmdata.io !

Brendan Nyhan @brendannyhan.bsky.social · 15d

Depolarization is not "a scalable solution for reducing societal-level conflict.... achieving lasting depolarization will likely require....moving beyond individual-level treatments to address the elite behaviors and structural incentives that fuel partisan conflict" www.pnas.org/doi/10.1073/...

Reposted by Marek Vranka

Daniel Lakens @lakens.bsky.social · 16d

New blog post from Data Colada, responding to the recent criticisms on p-curve analysis. It is a *very* good response. As in, it addresses exactly the points I would have expected in a reply, and it explains why I will still teach p-curve analysis. datacolada.org/129

8 22

Reposted by Marek Vranka

Kinlika @kinlika.bsky.social · 17d

'Tylenol' is known in the civilised world as Paracetamol

2.4 million children born in Sweden between 1995 and 2019. Studied 186,000 children whose mothers were treated with paracetamol during pregnancy.

news.ki.se/no-link-betw...

No link between paracetamol use during pregnancy and autism or ADHD in children

In the largest epidemiologic study to date of the risk of giving birth to a child with autism, ADHD or intellectual disability following acetaminophen use during pregnancy, researchers found no associ...

news.ki.se

42 710 1.7K

Reposted by Marek Vranka

samuel mehr @mehr.nz · 17d

this is a very sharp piece on why it makes no sense to run universities as if they are businesses. They're not businesses.

www.afr.com/work-and-car...

The net result is the worst of both worlds. Universities invoke the rhetoric of business discipline, but they lack the governance structures that give that discipline bite. They operate without the checks that private ownership provides, yet subject staff and students to the cost-cutting and efficiency drives that profit-maximising firms pursue. The result is waste at the top and insecurity at the bottom.

15 130 450

Marek Vranka @mvranka.bsky.social · 19d

😢

Marek Vranka @mvranka.bsky.social · 19d

Villeneuve should make finishing this www.imdb.com/title/tt3188... the top priority, especially in light of developments like these: thebulletin.org/2025/09/the-... ...

Marek Vranka @mvranka.bsky.social · 20d

since I first heard about LLM "silicon samples," I've been baffled by the idea. even if a group of humans were able to predict the behavior or answers of others, no one would use it instead of real data, cause there would clearly be unknown biases and situations in which the predictions would fail.

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

1 4

Marek Vranka @mvranka.bsky.social · 21d

just found out that a study about chatGPT predictions of experimental condition effects that I was a tiny bit involved in was published - already a year ago 😅 ‍royalsocietypublishing.org/doi/full/10.... - even the now archaic gpt-4 did as well as humans, would be interesting to see current models

1

Reposted by Marek Vranka

Peter Luca Versteegen @versteegenluca.bsky.social · 21d

What does that mean?
Bad news: democratic views are even more fluid than recently shown. In times when we feel poorly, we evaluate opponents + institutions more harshly.
Good news: there are limits to this. Short-term changes in affect (eg traffic jam) won’t shift views or prompt intolerance
7/8🧵

1 1 3

Marek Vranka @mvranka.bsky.social · 25d

thanks to all attendees and panelists for a great discussion, and a special shout-out to Jeff for bringing up Imre Lakatos ❤️

1

Marek Vranka @mvranka.bsky.social · 28d

if you’re at #APSA2025 in Vancouver, check out our roundtable on the Nuclear Taboo (Sat, Sep 13, 8 AM), with a great lineup of nuclear non-use scholars discussing norm contestation, elite–public attitudes, and the alleged erosion of the nuclear taboo

APSA @apsa.bsky.social · Sep 4

Theme Panel: Russia’s Invasion of Ukraine and the “Nuclear Taboo” #APSA2025

Theme Panel: Russia’s Invasion of Ukraine and the “Nuclear Taboo”

In-Person Roundtable Participants: (Chair) Michal Smetana, Charles University (Presenter) Lauren Sukin, University of Oxford (Presenter) R. Charli Carpenter, University of Massachusetts, Amherst (Presenter) Stephen Herzog, Middlebury Institute of International Studies at Monterey (Presenter) Naomi Egel, University of Georgia (Presenter) Jeffrey W. Knopf, Middlebury Institute of International Studies at Monterey (Presenter) Marek Vranka, Charles University (Presenter) Ondrej Rosendorf, Institute of Political Science, Faculty of Social Sciences, Charles University…

politicalsciencenow.com

1 5

Marek Vranka @mvranka.bsky.social · Sep 4

funny how banning smartphones in classrooms has become one of the few things both sides agree on 🙃 however, the research was kind of flimsy... now a new big study out of India shows some benefits; especially for struggling students – although the overall boost looks small (about d 0.08)

The Economist @economist.com · Sep 4

The first large-scale randomised control study yields a ringing endorsement

Banning smartphones in classrooms helps students

The first large-scale randomised control study yields a ringing endorsement

econ.st

Marek Vranka @mvranka.bsky.social · Sep 4

it simply comes with the mid-century design ❤️

1

Marek Vranka @mvranka.bsky.social · Sep 4

pinging @fsv.charlesuni.cuni.cz for some inspiration re: office furniture 😉

Daniel Lakens @lakens.bsky.social · Sep 3

New office furniture. :)

2

Reposted by Marek Vranka

Shashank Joshi @shashj.bsky.social · Sep 4

Experts of BlueSky: what are the best recent papers, studies & sources on mercenaries / private security companies, particularly how the field has evolved as business over the last few years? I'm interested in global examples, rather than just Western & Russian ones.

6 22 65

Reposted by Marek Vranka

Aaron Charlton @aaroncharlton.bsky.social · Jan 1

I've updated the marketing replication tracker through 2024. So far, 5 out of 45 (11%) of all direct replications of marketing studies (studies published in scientific marketing journals such as Journal of Consumer Research and Journal of Marketing Research). openmkt.org/research/rep...

3 4 6

Reposted by Marek Vranka

Daniel Kelly @dryan149.bsky.social · Sep 1

The Moral Psychology Research Group will host an Online Gathering this fall November -- all are welcome! Speakers include Joshua Greene, Meltem Yucel, and Paul Bloom as inaugural recipient of the The Stephen P. Stich Award for Career Achievement in Moral Psychology

sites.google.com/view/mprg/on...

MPRG - Online Events

An Online Gathering for Moral Psychology -- Fall 2025* Friday, November 7th: 4:00pm to 7:00pm (EST) Saturday, November 8th: 11:00am to 3:15pm (EST)

sites.google.com

1 4 19

Reposted by Marek Vranka

Tomas Weiss @tomasweiss.bsky.social · Aug 27

Interesting read, even if worrying. Much more reliance on paid WoS in Czechia. But some promotion models started going the Scholar way. Maybe a mistake…

Hannah Shelley, MLIS (Metadata, Lattes & Impostor Syndrome) @hannahshelley.bsky.social · Aug 13

Since search is dead, how soon do you think Google Scholar is headed for the Google Graveyard? I'm betting it's soon, and academia is NOT prepared

Google Scholar Is Doomed

Academia built entire careers on a free Google service with zero guarantees. What could go wrong?

hannahshelley.neocities.org

1 2

Reposted by Marek Vranka

Marcus Crede @marcuscrede.bsky.social · Aug 18

This thread. Sheesh.

Martin Plöderl @ploederl.bsky.social · Aug 17

1. I need to recover this thread from X about a review on lithium in tap-water and suicide. It shows that papers aren't retracted despite being total nonsense

Martin Plöderl @ploederl.bsky.social · Feb 10

1. There seem to be several issues with this new paper about lithium in tap water and suicide. Thanks for sending me a pdf >

1 4

Marek Vranka @mvranka.bsky.social · Aug 17

although my priors regarding claims by @jburnmurdoch.ft.com are quite low, there seems to be a real d=0.4 difference (cf grimoiremanor.substack.com/p/no-conscie...) However, it's based on a non-representative convenience sample of online respondents, so no way to tell what is the cause...

Christian Odendahl @codendahl.bsky.social · Aug 17

Will only post one of @jburnmurdoch.ft.com's charts, you need to read the whole thing and look at the others in the FT.

1

Reposted by Marek Vranka

Society for the Improvement of Psychological Science (SIPS) @improvingpsych.org · Aug 15

PsyArXiv is seeking new moderators to help combat an increase in AI submissions! If you've ever posted a preprint to PsyArXiv, please consider joining. Minimum commitment 1h/month, there's a training session this Monday @ 1pm ET. More info here: forms.gle/9LB1rEtxHAeZ... #PsychSciSky

Expression of Interest in Serving as a PsyArXiv Moderator

As you might have heard, PsyArXiv is having some issues with an increase in low-quality submissions, ranging from AI generated manuscripts to inflate citation metrics, incoherent or nonsensical docume...

forms.gle

91 75

Marek Vranka @mvranka.bsky.social · Aug 17

reading about left-wing authoritarianism (doi.org/10.1111/pops...) reinforced my pet peeve with political psych: treating ideology as a personality trait is doomed to fail as complex, context-bound worldviews can’t be meaningfully flattened into neutral questionnaire items...

A specter is haunting political psychology—a specter of left‐wing authoritarianism: Development and validation of left‐wing authoritarianism scale in a post‐communist society

The position of left-wing authoritarianism (LWA), contrary to its right-wing counterpart, has long been contested within the psychological literature. Efforts to examine the nature of LWA have recent...

doi.org