Stephen Vaisey

@vaiseys.bsky.social

Posts

Professor at Duke University. Cultural evolution, public opinion, social change, nerdy quant stuff in R. https://vaiseys.github.io/

Posts Replies Media

Reposted by

Stephen Vaisey

our methods for studying political change and stability typically provide group-level estimates, while our theoretical goal is often to understand individual-level processes. Steve Vaisey (@vaiseys.bsky.social) & I ask whether we can go from the former to the latter in panel data: osf.io/rhf4q

Reposted by

Stephen Vaisey

taxes

Reposted by

Stephen Vaisey

Using the choice to call 311 vs 911, phantom rules—frequently broken and rarely enforced codified rules—are more often enforced for Black people than White people (in US) journals.sagepub.com/doi/abs/10.1... #psychology #socialpsyc #PsychSciSky #BehSci Polisky

Reposted by

Stephen Vaisey

Lots of confusion between “every great new technology was initially resisted” and “every technology that is initially resisted will be great.”

Reposted by

Stephen Vaisey

"Read The Syllabus" t-shirt to support Fast Capitalism (no logo on it) bonfire.com/read-the-syllabus

Read the Syllabus | Bonfirebonfire.com A fundraiser for Fast Capitalism. A fundraiser for Fast Capitalism http://fastcapitalism.com

Reposted by

Stephen Vaisey

I got to explain the argument of @kjhealy.bsky.social's instant classic of an article "Fuck nuance" today not once but twice (the first time in a large first-year revision class, the second time in a final-year revision supervision), so today was a good day. #seriously #fuckit

Reposted by

Stephen Vaisey

This is a really smart piece/experiment about the way questionnaire design can affect electoral preference qs. (Also piggybacking to reshare my story from a little while ago about why third party candidates are so difficult to poll: www.cnn.com/2024/04/22/p...)

The order of the questions in a poll can matter. www.nytimes.com/interactive/...

A Simple Experiment Reveals Why It’s So Hard to Measure R.F.K. Jr.’s Supportwww.nytimes.com Which polling question is asked first can matter a great deal.

Reposted by

Stephen Vaisey

New paper out! @vaiseys.bsky.social and I examine mechanisms of change across many variables and countries. We find that issues that are more difficult to discuss change more via cohort replacement. 1st step twrd studying what issues change through which mechanisms! www.cambridge.org/core/journal...

Opinions on hard-to-discuss topics change more via cohort replacement | Evolutionary Human Sciences | Cambridge Corewww.cambridge.org Opinions on hard-to-discuss topics change more via cohort replacement - Volume 6

Reposted by

Stephen Vaisey

❗New methods paper out on OSF ❗ A new R-squared for multilevel models 👀👀 Resolves a long-standing R-squared oddity identified by Snijders/Bosker 30 years ago! with the amazing Anders Holm #sociology #erc Paper: osf.io/preprints/so...

Reposted by

Stephen Vaisey

{gssr} is now two packages. kieranhealy.org/blog/archive...

gssr is now two packages: gssr and gssrdockieranhealy.org Summary My gssr package is now two packages: gssr and gssrdoc. They’re also available as binary packages via R-Universe which means they will install much faster. The GSS is a big survey with a big co...

Reposted by

Stephen Vaisey

It’s a struggle to keep up with the pace of change, but imagine how impossible it would be without blog tutorials, open source, github, Jupyter notebooks, &c. The open web grew an amazing knowledge infrastructure in the last couple of decades. AI is in a way a continuation of that project.

Reposted by

Stephen Vaisey

do life-course transitions---ranging from first-parenthood to unemployment---change political orientations? using difference in differences models over 6 life events & 40 survey items, I show that the answer is *no.* a new working paper: osf.io/rk47w

Reposted by

Stephen Vaisey

The econ paper that found a 42-64% increase in rapes after prostitution banned (in Sweden)? Turns out result from coding error + Stata prioritization rues + seasonality. * always plot raw data * always sniff test your estimates * open science works! On X or here: drive.google.com/file/d/1oGP3...

Re-analysis of Ciacci (2024).pdfdrive.google.com

Reposted by

Stephen Vaisey

Science runs on D&D rules because if you roll a natural 20 you get to publish.

The General Social Survey now has post-stratification weights for all waves (1972-2022)! You can get them in the new release of the data. Check out the codebook (~p. 36) for technical details: gss.norc.org/Documents/co...

Reposted by

Stephen Vaisey

Good statement re bogus Rufo plagiarism charge from PAA President Jennifer Glass.

Reposted by

Stephen Vaisey

It's not that I care at all about Musk's personal opinions. It's that I'm committed to an open internet, X is digging in on "walled garden," and it feels dumb to keep networking inside a trap we can walk out of.

Reposted by

Stephen Vaisey

Lovely review of my book from Mirta Galesic/ @sfiscience.bsky.social www.santafe.edu/news-center/...

New book offers thoughtful approach to modeling complex social systemswww.santafe.edu <p>In his new book <em>Modeling Social Behavior: Mathematical and Agent-Based Models of Social Dynamics and Cultural Evolution</em>, SFI External Professor Paul Smaldino provides an excellent introduc...

Reposted by

Stephen Vaisey

seeing here and on the other place evidence of AI garbage in Google Scholar I feel moved to say that I and other humans moderate SocArXiv osf.io/preprints/so... and do our best to block AI garbage and inflated self-citations, and it is a growing burden to do so. #sociology #polisky

OSFosf.io

Reposted by

Stephen Vaisey

Kind of remarkable to see how this fairly tightly-knit network of frauds constructed a profitable subfield for themselves. It’d be a good case-study style dissertation.

Reposted by

Stephen Vaisey

🚨 SCOOP: Here’s the unsealed, 1,300-page report showing how Harvard Business School concluded that dishonesty expert Francesca Gino committed research misconduct. www.chronicle.com/article/here...

Here’s the Unsealed Report Showing How Harvard Concluded That a Dishonesty Expert Committed Misconductwww.chronicle.com Francesca Gino said that another professor could have meddled with her data — but Harvard said she didn’t offer any proof.

Reposted by

Stephen Vaisey

Not great to see "rigging" language spreading to a Democrat who lost in a free and fair election. Not acceptable for anyone of either party to undermine trust in the system this way.

This election rigging language pisses me off to no end. Sometimes you just get outspent. Sometimes that results in a loss. Grow up, move on, and support the Democrats moving forward.

Less concerned about AI than I was a few minutes ago

Reposted by

Stephen Vaisey

Man even the Fremen were able to organize high-speed mass transit

Reposted by

Stephen Vaisey

This is very important because it's become very common for polling to effectively be 'opt-in'. There are real biases, and to be clear, you can't fix this with very large samples!! We can come to dangerously erroneous conclusions because of it.

That very viral "Holocaust denial among young people" finding doesn't replicate. t.co/9wT82XfBzW

Reposted by

Stephen Vaisey

There’s a moral asymmetry in free will attributions. People think you have more free will when you do bad things than when you do good things. But why? Previously, there were two rival explanations for this effect. Then today, a new one just came out

Reposted by

Stephen Vaisey

Do you think that learning more about causal inference is not worth it because you're running experiments anyway, or because you're interested in predictive questions? In that case, I've written a paper just for you, out now in SPPC: compass.onlinelibrary.wiley.com/doi/10.1111/...

Reposted by

Stephen Vaisey

Got useful comments, collected a bunch of new data, and our paper with @vaiseys.bsky.social about mechanisms of cultural change across different issues and contexts is much better: osf.io/preprints/so... Some generative modeling about social change and some cool stats are my favorite parts!