Since their rollout, I have made heavy use of the custom GPT tools in ChatGPT. Here is a thread of some of the tools that I use regularly (all of them require ChatGPT 4). I think they are most useful for people doing quantitative social science, but some are more general.

What explains the recent wave of military coups in Sub-Saharan Africa? In the latest PRIF Spotlight, Markus Bayer (BICC), @felixbethke.bsky.social, Aurel Croissant & Nikitas Scheeder (Heidelberg University) take a look at political militarization as a risk factor. blog.prif.org/2023/12/28/b...

New working paper with Werner Krause & Samer Massoud on the effects of targeting hospitals on violence dynamics in Northwest Syria hicn.org/working-pape... conflictsky polisky menasky

New data alert! UCDP Conflict Issues Dataset (CID) is here! This global yearly dataset contains 14 832 conflict issues – divided into a four-tier typology with 120 categories at the most disaggregated level – raised by all armed non-state groups active in civil war, 1989-2017. t.co/3hrZxZr6Qo

Please see here for a call for doing more to integrate preregistration into the review process: "Open Letter on the Need for Preregistration Transparency in Peer Review" -- #polisky #econsky You can sign too -- link is in the letter. docs.google.com/document/d/1...

In other words, it is not really possible to have observed higher replicability than we did. The evidence provides proof-of-concept that high replicability is achievable. We can't, however, say the extent to which each practice contributed to replicability.

With a total of 16 new discoveries, 16 self-replications, and 48 independent replications, we found high replicability. Independent replication effect sizes were 97% the size of the confirmatory findings. On the "same direction, p<.05" benchmark, 86% replicated, higher than expected based on power

Our paper "High replicability of newly discovered social-behavioral findings is achievable" appears today in Nature Human Behaviour. Just as the large replications provided an inflection point in identifying replicability challenges, this paper does so for solutions. www.nature.com/articles/s41...

Polisky - anyone have a good contact at Lucid? Someone who responds fast? Thanks.

Start of the special contributed session on Large Language Models and Causality at CDSM '23. Amit Shama presents some novel results on this exciting topic coming out of Microsoft Research. ➡️ causalscience.org

I couldn't find a tutorial I liked to get students who know R up and running with Python, so I wrote my own! Part 1 is here: andrewhalterman.com/post/python_... (And if you have a tutorial you like, please let me know!)

Public support for democracy is crucial to prevent backsliding. But existing survey questions focusing on “democracy” may not measure the concept well, as the term “democracy” varies in meaning and may prompt social desirability. polisky 1/3

New draft of paper w/ @davidryanmiller.com ! Quick summary: - Conjoints mimic political choices that people face - Possible problem: If people don’t approve of choices, they may abstain - Findings: This matters!!! - Big update: New ex w/ tied & different attribute prioritisation Link: osf.io/4ws7m

Important stuff. Most conjoints force subjects to choose. This can distort results in unpredictable ways.

Apparently ~40% of crowdsource workers use LLMs when completing surveys Relatively simple fix for text responses: Add Javascript to Qs to prevent pasting Credit to Jiabi Wang for idea + Kaushal Addanki for code below github.com/rafmbatista/...

Israels Selbstverteidigung: Über die humanitäre Lage in Gaza und die Chancen zur Eindämmung des Konflikts diskutieren heute 19 Uhr im Stream &  im Haus am Dom Claudia Baumgart-Ochse & Michel Friedman, Moderation: A. Schwarzkopf, Frankfurter Rundschau: www.hsfk.de/veranstaltun...

Using admin privileges for today's Dictatorsky shoutout. My study on civilian involvement in military coups. Civilian resources (elite vs. mass-based) shape their conduct in coup politics - from instigation to consolidation. More to come :) Polisky, CivMilSky journals.sagepub.com/doi/full/10....

Dear all, just rediscovering my own website that I put online at the beginning of this year. Needs to be updated but pretty much contains most of the #Stata #dataviz I created in the past three years. asjadnaqvi.github.io/stata-portfo... Check it out!

A list of financial data sources, curated for political economy use cases. Official statistics as well as datasets compiled by organizations and academic researchers. Originally assembled for my students, now available on my website. Suggestions welcome! benjaminbraun.org/teaching/dat...

The Uppsala Conflict Data Program (UCDP) will present its new UCDP Conflict Issues Dataset at the Data4Peace 2023 Conference on October 18 at 10:45 Eastern Time #conflictsky Register here: cic.nyu.edu/events/data-...

Religious Minorities at Risk is available for order at OUP! In it, we examine how different forms of discrimination & grievance contribute to varying degrees of peaceful & violent mobilization by religious minorities Use discount code ASFLYQ6 to save 30% at checkout. polisky | polrel | conflictsky

My book on data management is available #OpenAccess at www.cambridge.org/core/books/d... ! Also, check out the companion webpage at dmbook.org. I hope that the book will be useful for your research and teaching.

Great to see this out: blog post in The Loop summarising our recent PLOS ONE paper, w/ Margherita Belgioioso & Kristian Gleditsch. PoliSky ConflictSky theloop.ecpr.eu/community-de...

🍾New paper with Roberto Cerina, Neil Ketchley, and Aaron Zelin now accepted at Political Analysis journal! polisky cssky sociology drive.google.com/file/d/1Xv3m...

Friends: Our interaction models are underpowered intxpower.com?A=0.5&B=0.4&... journals.sagepub.com/doi/full/10....

Gibt eine tolle Tweetdeck-Alternative für Bluesky, die ich seit wenigen Minuten ausprobiere und davon sofort begeistert bin: deck.blue/ Für das Login richtet euch am besten ein App-Passwort in den Bluesky-Settings ein.

A problem in economics research: Null results are considered less relevant and less publishable than statistically significant results, especially when they go against expectations. 📉📈 doi.org/10.1093/ej/u...

Bluesky has @timgill924.bsky.social now.

Introducing Conflictsky, a new feed for political conflict and violence! Like and pin the feed here: bsky.app/profile/did:... To post, tag with 'conflictsky' or 'violencesky' It's a polisky sub-feed, so request to join that list if you haven't already: www.paulmusgrave.info/how-to-use-p...

Groundhog for R is so good for reproducibility. You just load your packages with it and voila, anyone who runs your script will have the same package versions. One less hurdle to reproducible code. Props to the team at Wharton Credibility Lab! groundhogr.com #psychscisky

New take on the replication crisis in the social sciences, proposing "replication games" to address the issue www.nature.com/articles/d41... I like that they highlight coding errors as a problem We write thousand lines of code. There have to be many errors that we did not spot. polisky

Dear @profmusgrave.bsky.social could you please add me to polisky? Thanks!

Hey Stata folks! Want to visually present your regression results? I made a guide. It covers key aesthetics, combining results from multiple models, working w/ logit models, & more. Oh, and I provide all the code. Comments/feedback welcome! drive.google.com/drive/folder...

First post here! Early view of my AJPS article with Leonard Wantchekon on the legacy of African independence movements. We argue that rural insurgencies generated political exclusion & autocratic regimes, while urban protests fostered democratic behavior & institutions. doi.org/10.1111/ajps...