Die Adenauer School of Governance wurde letzten Montag offiziell eröffnet. uni-koeln.de/universitaet... Ziel der ASG ist „problemlösungsorientierte Forschung und Lehre im Bereich Public Governance an der Schnittstelle von Wirtschaftswissenschaft, Rechtswissenschaft, Technologie und Staat". 1/

Warning to users of R studio. Don't upgrade to Version 2026.07.0+139 (2026.07.0+139). It has a known bug that adds an irritating error message 'Error in .rs.exprMutatesPackageLibrary(part) : argument "part" is missing, with no default'. Apparently it'll be fixed soon, so worth waiting. #Rstudio

Diese Professorin wurde an ihrer Universität mit dem Tod bedroht. Jetzt hat sie ein geheimes Büro. Mein Report über die zunehmende Wissenschaftsfeindlichkeit in Deutschland – jetzt im @spiegel.de und online (€): www.spiegel.de/wissenschaft...

Ich habe in meiner Statistikveranstaltung aus aktuellen Anlässen diese Grafik vom RKI gezeigt, um zu illustrieren, wie sie Unsicherheit ausweisen über das Prädiktionsintervall. (aus https://www. rki.de/DE/Aktuelles...). 1/

Interesting read for sociologists given how sociologists vary in whether anthropology is to be seen as a role model or cautionary tale.

As a guy whose name comes up when you search "many-analyst" now I've gotten quite a few referee requests now for papers where the "many analysts" are AI bots, or undergrads, or random people, rather than professional researchers.

Feel bad but teaching my last stats/data science in fall 2023 was getting the last chopper out of Vietnam

Preliminary Evidence Linking #OpenScience to Research Integrity scholarlykitchen.sspnet.org/2026/01/28/p... It's worthwhile to look at how common certain features are among legit and papermill articles. Results indicate differences between types of papers, but there are some issues, I think. 1/

They could have made this map any color.

Genuinely interested in your views: What is the future of teaching statistical programming language for data science in the age of AI? Will we even still teach programming R or Python?

The Cover-Up in Gino v Harvard datacolada.org/136

Arguing with economists: the case for preregistration #MetaScience kdoroc.substack.com/p/arguing-wi... Assuming this represents a broader picture in econ: I would have assumed that econ was more open to preregistration than other disciplines. This post suggests differently. 1/

Wie können wir die Wissenschaftsfreiheit gegen zunehmende Angriffe schützen? Ein neues Positionspapier einer DFG-Senats-AG zeigt auf, wie Resilienz von Wissenschaft & Wissenschaftssystem gestärkt werden kann & wie auch Politik, Wirtschaft & Gesellschaft unterstützen können: sohub.io/9tje

"To what extent are concepts like 'evidence', 'theory' or others instantiated or used differently in different fields and what does this imply?" #PhilSci #MetaSci

“move from preprint to peer-reviewed publication leaves the central claims of most abstracts intact, indicating preprints are a reliable source…papers that were never posted as preprints were retracted at roughly twice the rate of those that were…” www.biorxiv.org/content/10.6...

If you were thinking about the most impressive qualitative publications in Poli Sci, the ones that did a fantastic job of presenting their data or making the case for their chosen methods... ...what are you choosing?

A Comparison of Packages to Generate Codebooks in R | Crystal Lewis cghlewis.com/talks/rladie... This is a short and very useful overview over the alternatives for codebook generation. There are many more than I knew, but only few meet most requirements #rstats

Wenn Verlage sich tatsächlich uneinsichtig zeigen und an langen Zweitveröffentlichungsembargos festhalten, so dass Autor:innen OA-Mandate der Forschungsförderung nur einhalten können, wenn sie horrende Publikationsgebühren zahlen, dann ist das bitter. www.statnews.com/2026/06/11/o...

The emerging disenchantment with retrospective meta-analysis www.nature.com/articles/s41... Retrospective meta analysis lacks prior unobservability of data, hence all problems known from empirical research apply. Multilab preregistered studies are seen as best alternative.

The Q1Q2Q3 Workflow for Statistics and Data Science Collaborations www.tandfonline.com/doi/full/10.... This is a useful discussion of the qualitative parts that complement a quantitative data analysis. If the goal is to introduce students to basic inferential stats, I think there is only room 1/

Welcher Post hätte denn verschickt werden sollen? Irgendwas mit zuviel Work-Life Balance im Fußball? Kann man das mit IFG herausfinden?

📢 Open call for two additional contributions to our Special Issue "Methodological Advances for the Study of German Politics" @pvs-journal.bsky.social. 📄 Submit by July 12: www.sosci.uni-hamburg.de/pvs/ ℹ️ www.dvpw.de/gliederung/s... @gessler.bsky.social @lukrudolph.bsky.social @dvpw.bsky.social

Heute bin ich in @jmwiarda.bsky.social's Blog zu Gast, mit einem Beitrag zur Hitze im Hörsaal und wie sie sich ganz einfach vermeiden liesse: www.jmwiarda.de/blog/2026/06...

Hitzekollaps im Hörsaal Wer Studierende und Lehrende vor den immer häufigeren Hitzetagen schützen will, sollte die Semesterzeiten neu denken. Ein Gastbeitrag von Nils Weidmann. Im Wiarda-Blog: www.jmwiarda.de/blog/2026/06...

The debate about Borjas/Breznau's paper on ideological bias in migration research is out now @i4replication.bsky.social. With a comment by Katrin Auspurg & Josef Brüderl: hdl.handle.net/10419/341575 And a reply by the original authors: hdl.handle.net/10419/341576

Most of the statistical tests you learned in stats class — t-test, ANOVA, correlation — are actually special cases of a single thing: linear regression. Ch 7 of Experimentology argues that thinking in models, not tests, is more flexible and a better foundation for theory. 🧵 experimentology.io

"Whether a paper that fails a robustness check or skips preregistration is worse science depends on assumptions about what those signals are proxies for — assumptions reformers have been debating for years without resolution." jessicahullman.substack.com/p/ai-review-...

NIH grants have been under political scrutiny since early 2025. But what does the screening process actually look like? What is being flagged? I spent a month digging into leaked docs and emails. We've published the 235-word list that are being flagged by an internal AI tool. And so much more. 🧵

Im März wurde gefeiert: die UHH behält ihren Exzellenzstatus. Seit dieser Bekanntgabe, dass die UHH Exzellenzuniversität ist, fehlen ‚plötzlich‘ überall Gelder. 🤷🏻‍♀️ exzellent und pleite. Läuft.

There’s been a lot of buzz around the Borjas/Breznau (2026) paper (www.science.org/doi/10.1126/...) and potential ideological bias in research. We took a closer look at the data. Could there be a "bias in bias research"? Here is what we found: www.econstor.eu/bitstream/10...

An important read. Naive researchers are using TriNetX easy-to-use tools without realizing that their analyses of routinely collected data are meaningless, even claiming that nonexistent TriNetX tools fixed biases.

HBS' latest filing in the Gino lawsuit contains a few new details: 1. A committee of 26 professors, admins, and fellows reviewed the allegations, and *unanimously concluded* that there was *clear and convincing evidence* of research misconduct.

Open a paper to read Open Supplement Open OSF Repo Open Prereg Open Github Open Data (12 tabs) Open Methods reference Open Methods reference supplement Open methods reference github Open relevant code (3 tabs) Open relevant python documentation "Your computer is low on memory"

A minor issue that keeps anonoying me: For many articles (qualitative and quantitative), I noticed that some material is posted in a repo and some material as a supplement on the article website. Can't we just put everything into the repo? All would be in one place with a stable URL.

People often ask me whether it's risky to use clear causal inference theory and language, in case they get ignorant peer reviewers. My advice is that high quality and transparent research methods shine though. I fear my advice might be wrong. Too many good people are getting feedback like this.

The Department of Political Science at Aarhus University is hiring both Assistant Professors and Associate Professors. Assistant Professor: international.au.dk/about/profil... Associate Professor: international.au.dk/about/profil... Please share these opportunities with your networks!

This paper finds that economists whose job market papers report (marginally) statistically significant results are more likely to secure academic jobs. Hiring committees prefer statistical significance; this creates incentives for researchers to p-hack.

I am reading the report that other reviewers made for a manuscript I also reviewed. One reviewer (R2, actually) simply says the paper has been AI-generated, therefore he/she is not reviewing it. There is no explanation where this impression comes from. This is not a helpful report.

This looks absolutely incredible. @lars.vilhuber.com I think you're going to love this

*NEW* @rorinstitute.bsky.social Substack post: @lukaswallrich.bsky.social, @aufdroeseler.bsky.social & @josefinaw.bsky.social explain why they created a database to facilitate discovery of social & behavioural sciences replication studies. researchonresearchinstitute.substack.com/p/did-you-kn...

I sometimes log back into X to see who's still active. I am always disappointed. But it's the journals, employers, & other health organisations that shock me most. The message this sends about culture & values is so negative. From a quick scroll, I saw JAMA, Lancet, BMJ, APHA, Wellcome, NCRM, etc

Reading this has subtly reframed my thinking on the replication crisis as being less a crisis and more just a special case of Sturgeon's Law.

The ultimate get-out-of-jail-free card

What's inhibiting science reform? "Cognitive conservatism, self-reinforcing feedback loops, epistemic dominance, and the selection bias of reform agents." Preprint: philsci-archive.pitt.edu/29963/ #MetaSci #PhilSci #STS

Über 10 Jahre wurden die Universitäten durch tausende von Cancel Culture Leitartikeln sturmreif geschossen, jetzt werden sie in den USA durch eine rechte Regierung systematisch zerstört. Dieses extrem deprimierende Video illustriert das an einem Fall. www.youtube.com/watch?v=jFMc...

Now that's what I call a spicy opeing: Understanding mechanisms is not a research question.

Next time I am asked for evidence of anything, I also just say "All I need to do is look." (at 0:30)

I like this @thejop.bsky.social : "We are now implementing a system in which the editorial team has the right to not consider submissions from authors who have failed to review for the journal three or more times, when invited to do so by the editorial team."

The numbers are not good. Only 5% of the selected articles (32 of 728 from AJS, ASR, SF, 2019-2024) share code and data such that verification attempts are possible. Given that material is available, all articles are reproducible to a lesser or larger degree (sometimes after code adjustment).