Rense Corten 🟥

Kan wel in Signal

As a fellow native speaker of a language containing bizarre sounds (Dutch) I feel I have to sympathize with the Danes here.

Want to learn about computational social science *for free* and identify new research partners across academic fields? Apply to one of the 2026 Summer Institutes in Computational Social Science (described in yellow in the attached map) here: sicss.io/locations

In veel buitelanden is het veel gebruikelijker dat promovendi zakken bij hun verdediging, wat zonder de poeha dan ook minder pijnlijk is.

Toch is het hele circus eromheen niet alleen voor de sier. Het geeft de hele verdediging ook meer gewicht en zichtbaarheid, waarbij vooral de promotores (maar ook de promovendus zelf) reputatieschade oplopen als er toch iets mis blijkt met het proefschrift.

There was still no answer.

For those following along, some thoughts on Thurner et al.'s reply to our critique to their "connectivity causes polarization" claim. At first sight, this looks like a solid rejoinder, perhaps even justifying the mildly sarcastic tone. Upon closer inspection, not so much. So, another thread... 1/

We mentioned this in our critique. If I where THK, I would get to the bottom of this and make crystal clear that these data exist, and where they can be found. I find the fact that they didn't do so increasingly puzzling. 9/9

Most alarmingly, they once more fail to clarify the source for precisely this one crucial post-2010 US/GSS data point. To the best of our knowledge, GSS 2020 did not measure core discussion networks. The original paper does not provide a plausible reference for it either. 8/

Nearly *all* their "post-transition" measure are based on different instruments than the "before" measures; with the "post" measures likely to produce higher estimates of network sizes. In fact, the *only* consistent time series in their figure are the 3 red squares for US/GSS (2004, 2008, 2020). 7/

THK furthermore fail to acknowledge that establishing a trend requires measures that are consistent over time (Stats 101), while theirs differ *systematically* over time. 6/

By doing so, they seem to treat polarization somehow as a cross-country phenomenon, even though the paper's title says "polarization in societies" (in fact they measure it in only one society, the US). 5/

They also do not address the issue that when looking *only* at the US, the alleged increase in connectivity cannot be established as preceding the increase in polarization. Without justification, they pool the data of *several countries* to establish the causal order. 4/

Likewise, THK invoke inclusion criteria (for measures) that turn out to be rather broad ("close friendships, core discussion networks, and strong ties") and do not in fact exclude our alternative measures 3/

First, not that THK bring up post-hoc exclusion criteria for our counterexamples, which are not even consistently applied. Eg., post-communist societies are now excluded (but not Germany, which is in their Fig), kin ties don't count (although also included in their GSS-based measures), etc. 2/

For those following along, some thoughts on Thurner et al.'s reply to our critique to their "connectivity causes polarization" claim. At first sight, this looks like a solid rejoinder, perhaps even justifying the mildly sarcastic tone. Upon closer inspection, not so much. So, another thread... 1/

I'd be happy to do so but I'm not sure when an LLM would consider its resources as "wasted"... (or even what it would consider as "its" resources - can an LMM claim any ownership over resources?)

Claude.ai just apologized for wasting my time

Today at 12:00, join us for an Online Introduction to High-Resolution Geospatial LISS Data in SANE 🧑‍💻

🔗 zoom.us/j/93104040907

Grafiek van sterfte in NL met primaire doodsoorzaak drugs. Data Jaar Drugs excl alcohol en tabak Alcohol Tabak 2014 123 882 19017 2015 198 957 19244 2016 235 989 19587 2017 262 974 19420 2018 224 1034 19275 2019 252 939 19272 2020 295 1019 20481 2021 298 991 20955 2022 332 1092 20221 2023 338 1125 19880 2024 378 1056 19220

Sterfte NL met primaire doodsoorzaak drugs, absolute aantallen.

“Don’t worry,” said Frog. “We will go back to all the places where we walked. We will soon find your button.”

Out of curiosity, in which industry did you witness this? Also, I wonder what sort of thinking they're still expected to do by themselves (and how we are supposed to teach and evaluate that)....

Please enlighten us!

Zeker, dat kan haast niet missen dan

The “Epstein Files” and the anatomy of hidden social networks How secrecy reshapes structure and why its analysis is conditional on the data

A lot of “network analysis” about the Epstein Files.

Nearly all of them missing the point: when a system operates under strong incentives for discretion, what gets documented is NOT a neutral sample of reality.
Results are conditional on what is observed!

open.substack.com/pub/manlius/...

Maar was die koffie voor of na zijn Nobelprijswinnende werk?

I’m looking for three PhD students for my new ERC project, starting 1 September. The goal is to understand how firms shape inequality in workers’ careers—using population registers.

Please spread the word! Deadline is March 8, more info here (see projects 4-6):

ics-graduateschool.nl/vacancies/

Deze collega's schrijven:
"Zo berichte NRC over een studie die liet zien dat migratieonderzoekers met uiteenlopende ideologische opvattingen tot tegenstelde conclusies kwamen op basis van dezelfde dataset"
Deze grafiek uit de studie laat zien dat dat n volstrekt onjuiste interpretatie is.

I know, of course. I was (jokingly) comparing to the Netherlands in the past

Just their livers though. Also, it's no longer as common here.