When you collect data online, are the results from humans or AI? In a project led by Booth PhD student Grace Zhang, we estimate the prevalence of AI agents on commonly used survey platforms:
osf.io/preprints/ps...
🧵
Our department is hiring an Assistant Teaching Professor!! This is a joint-appointed position with Computational Social Sciences (css.ucsd.edu). It's 75+ degrees F and sunny today, just thought I'd mention apol-recruit.ucsd.edu/JPF04461
🧵on my new paper "Synthetic personas distort the structure of human belief systems" w Roberto Cerina I'm v excited about...
🚨 Do synthetic samples look like human samples?
We compare 28 LLMs to the 2024 General Social Survey (GSS) to find out + develop host of diagnostics...
So I struggle to read this as a prevalence estimate for the platform.
One last thing is the shared data include original Prolific IDs. Account identifiers shouldn’t be publicly released for privacy concerns (e.g. you can link these IDs with personally identifiable info of participants later) 6/
I wonder whether any Prolific quality control is used in this study (it’s not reported), such as no approval rate filters and bot screening. That would have implications on 2/38 ratio. 5/
I don’t know why authors didn’t use fixed random seed. Bot-likeness can shift every time in each run (that’s more of a reproducibility issue though). 4/
The 3 bot-detection markers don’t agree on who looks “suspicious”. Each actually flags a different participant as worst, which makes the evidence that these ppl are bots is somewhat weaker. 3/
Prolific has 200k + active participants, and this study found at most ~2 participants with a strong anomalous signal out of 38 (the abstract claims 39 though). Can we extrapolate a crisis from that? I am not sure!
With all due respect to the authors (the study raises a “potential” concern), but its framing overstates that concern imo. A few things worth flagging before everyone panics 1/
💸🚨I am hiring 2 Postdocs for my ERC-funded project SOCDEBT on #debt dynamics across countries. One position: #SocialStratification + strong quantitative skills. The other: qualitative research and #EconomicSociology. waitkus.github.io/SOCDEBT/ 🚨💸
How far are we from Soc journals using AI agents to screen replication packages? (Sociologists: replication what)
Seems like low-hanging fruit. Let agents test the code and data and flag reproducibility problems before publication.
My brother wanted a London pub crawl. The result? My new Substack post: "Britain Lost 14,000 Third Places. They were Called Pubs. Is Your Local Next?" How private equity reshaped the local, which pubs are most at risk and most importantly what to do about it.
open.substack.com/pub/laurenle...
A new publication at www.pnas.org/doi/10.1073/...
We find that religious markers (esp. beards) can reduce perceived trustworthiness in Turkey.
(That's why you often see me clean shaven ;-)
This is why I was praising JEPS a few days ago. What they've done is impressive.
In any other context, requiring a high-productivity sector to offshore the production of its highest value-added exports would be seen as the economic illiteracy it so obviously is...
www.ft.com/content/a23c...
Two more weeks to apply for this job! Tenure track or tenured! Living in Paris! Nice colleagues (and me)!
Let me know if you have any questions! Details: www.shorturl.at/E57le
An der Universität Zürich ist auf den 1. August 2027 eine Professur für Soziologie zu besetzen.
The University of Zurich invites applications for a professorship in sociology starting on 1 August 2027.
We are hiring a post-doc to study the impact of AI agents on complex social systems in Duke's Society-Centered AI Initiative and/or the Polarization Lab! Apply here:
academicjobsonline.org/ajo/jobs/314...
Is that 4% Grok agents or libertarians?
And let’s be honest! Manual coding skills for entry-level analyst jobs? Probably obsolete in a year or so. I expect AI won’t save anyone with weak QM reasoning though. It will probably redistribute advantage, not flatten it.
If your QM module still runs on SPSS, the students may actually have a f***ing point here 😆 journals.sagepub.com/doi/10.1177/...
One from within the discipline, academic.oup.com/esr/article/... by @amaliaab.bsky.social and one from outside www.science.org/doi/10.1126/...
Job: Open-rank position in Environmental Health and Implementation Science in the Department of Health Promotion, Education, and Behavior, Arnold School of Public Health, University of South Carolina. uscjobs.sc.edu/postings/196...
📣 Come work with us! The @wzb.bsky.social Center for Civil Society Research is looking for a research associate/doctoral candidate (75%) for a 3 year period, starting on Apr 1, 2026. The researcher will work in the Manifesto Project. More:
wzb.hr4you.org/job/view/428...
Apply by Jan 25, 2026! 🗓️
🌍🌱 New #CLEAN 2026 Seminar Series ♻️🔥💡
#SocialScience seminar series with leading researchers to discuss topics ranging from #ClimateJustice and #Adaptation policies to #Energy transitions and #Health impacts of extreme heat 🏜️
📆 #hybrid Get in touch for zoom link
www.ucl.ac.uk/ioe/departme...
A new paper by George Borjas—who served this past year in the Trump White House designing some of its anti-immigration policies—claims to display evidence of ideological bias among researchers who study immigration.
doi.org/10.1126/scia...
🧵 Thread—>
I’ll be visiting the WZB Berlin Social Science Centre by April. If you’re around and would like to connect, let me know 👋 Feels like perfect hot-chocolate-chat weather ☃️
Looks like you already have an abstract
📢 Analytical sociology is coming home!
Call 4 INAS26 is open
🌍 1-3Jul26 @nuffieldcollege.bsky.social and @sociologyoxford.bsky.social (☔)
☢️ Deadline: 1Feb26
🦹 Organisers: @aksoyundan.bsky.social , @kasimirdederichs.bsky.social, D Kretschmer, @awaldendorf.bsky.social
Info: tinyurl.com/yc2tusjx
Will text you the details :)
