Seth Warner

2.8K posts

Seth Warner banner
Seth Warner

Seth Warner

@sethbwarner

Asst Prof @UConn, PhD @psupolisci. Research on affective polarization, intergroup relations, state/local politics. Nutmeg State native. 🗽 🗳️ ⚾

Storrs, CT Katılım Eylül 2018
355 Takip Edilen533 Takipçiler
Sabitlenmiş Tweet
Seth Warner
Seth Warner@sethbwarner·
🚨 New working paper! 🚨 Happy to share my new data on affective polarization by party across states, CDs, counties, and towns from 2009-23. Key point: polarization is not just an individual trait... contexts and electorates can be "polarized" too! 1/ papers.ssrn.com/sol3/papers.cf…
Seth Warner tweet mediaSeth Warner tweet media
English
1
15
61
6.9K
Seth Warner
Seth Warner@sethbwarner·
@ML_Burn Given the exponential rate of change, probably a few months?
English
0
0
1
42
Mike Burnham
Mike Burnham@ML_Burn·
Hey everyone, my 1 year old is predicting about 13 tokens now. Wondering how much more training is required before she can start doing my lit reviews?
English
1
0
9
555
Seth Warner
Seth Warner@sethbwarner·
Twice, I've prepped my students well for a test, saw a 72% average and wondered what happened. "Huh... The toughest Qs were fair! The discrimination scores are high?" Slowly, I realize it's actually a great test and I add a curve with a smile. This is what the SAT needs to do.
Brad Carson@bradrcarson

As a former uni prez...Best higher ed reform no one talks about: raise the ceiling on standardized tests - a lot! Too many top students max out, then compete on expensive pseudo-extracurriculars instead of actual merit. Wider score distribution fixes this.

English
0
0
0
135
Zachary Lorico Hertz
Zachary Lorico Hertz@zacharylhertz·
@jdbk Does that approval still hold despite the banana syrup because if so I'll mark you down as one of the few supporters
English
1
0
0
37
Seth Warner
Seth Warner@sethbwarner·
@bwitkovsky That's awesome, you might like to meet my colleague Michael Morrell sometime, he is very interested in that line of work. Thanks also for the older cites. You're right, there's some very good stuff here! Also a joy to be able to cite Susan when I can. 😊
English
0
0
1
19
Seth Warner
Seth Warner@sethbwarner·
Data request: does anyone know where I can find out which US jurisdictions use partisan vs nonpartisan elections? Primarily focused on local elections...
English
2
0
1
167
Seth Warner
Seth Warner@sethbwarner·
@mungowitz Congrats to you, Kevin, and the whole Munger family!!!
English
0
0
0
49
Seth Warner
Seth Warner@sethbwarner·
@MichaelRKistner Yeah, I saw that. It seemed kinda... dinky? 😅 Honestly, the best option might be to identify the offices we're interested in and ask Claude Code to grab the information. 😁
English
0
0
1
33
Michael Kistner
Michael Kistner@MichaelRKistner·
@sethbwarner That's the only error I've noticed so far, but it makes me worried there may be others. I've double-checked its listings for my uses. If you find a more reliable source, please let me know, working on a project where this matters now.
English
1
0
1
31
John V. Kane
John V. Kane@UptonOrwell·
@EfrenPoliPsy The five manuscripts under review at once probably says a lot more about long journal wait times than my (and my coauthors’) industriousness, but still, nice to see it happen 👍
English
1
0
3
125
John V. Kane
John V. Kane@UptonOrwell·
For 63 straight minutes, I had 5 manuscripts under review--a personal best. And then I received a desk rejection. But it really was exciting while it lasted ☺️
English
3
1
24
2.5K
Seth Warner retweetledi
Sarah Anzia
Sarah Anzia@SarahAnzia·
New essay in @TheForumJournal: Policies and institutions created decades ago solved problems and did some good. The challenge, though, is that after half a century of this – of these groups pushing in one direction, helped by institutions – society has lost its balance. It has led to bad outcomes that now need to be corrected, and new problems that need to be solved. And today’s policymakers are not starting with a blank slate. They must confront the interest groups and institutions that are in place, and those can make for some formidable political obstacles to problem solving. degruyterbrill.com/document/doi/1….
English
0
10
27
3.9K
John V. Kane
John V. Kane@UptonOrwell·
🚨 Interested in survey experiments? I'll be giving a virtual talk on (false) null results in experiments--and how to protect against them--at SWERP on 3/13 (3/12 8pm EST). Hope to see you there, and thanks to @sysilviakim & @_Brandon_Ives_ for inviting me!
John V. Kane tweet media
English
1
8
13
3.4K
Carlisle Rainey
Carlisle Rainey@carlislerainey·
Do you use logistic regression? If so, you’ll want to read the thread below. ⚠️ Warning: Memes, charts, #rstats, and practical advice ahead.
Carlisle Rainey tweet media
English
14
131
696
210.6K
Seth Warner
Seth Warner@sethbwarner·
@kvallier @AlexTolkin What do you mean by multiple instances? Alex, I've gotten some mileage over picking out a few studies first, then having Claude: - follow up on what it cites - brainstorm search terms - summarize what it can find full texts of - ask me for any PDFs it can't locate
English
1
0
0
42
Kevin Vallier
Kevin Vallier@kvallier·
The rank AI prejudice in the comments is striking in both hostility and ignorance. I *never* do literature reviews without asking ChatGPT Deep Research, now that it is based on GPT 5.2. It is *phenomenal*. If you add Gemini and Claude, the coverage is unparalleled.
Sean Westwood@seanjwestwood

The academic paper is a dead format walking. AI does lit reviews better. AI will do (is doing)peer review. Users will skim AI summaries. The real science is the question, the pre-analysis plan, and the analysis. The 30-page paper is just vestigial wrapping paper.

English
14
7
61
20.2K
Seth Warner
Seth Warner@sethbwarner·
That said... it will massively change how we research, and to some degree, how cutting-edge scholars view one another's research.
English
0
0
0
30
Seth Warner
Seth Warner@sethbwarner·
Contra the discourse, I think disciplinary change will be gradual amid AI... - Many faculty are still out of the loop on causal inference, "big data" - Qual/theory may not see disruption - Being front line, journals will probably adapt first, hiring/tenure committees years later
English
1
0
1
132
Seth Warner
Seth Warner@sethbwarner·
@matt_blackwell Can you say more about what that could look like? I'm imagining... - Someone used AI code, the analysis doesn't do what they claim, and they have to write an erratum - The pendulum swings against AI in the future, and low-quality outputs from 2026/27 will carry stigma
English
0
0
0
146
Matt Blackwell
Matt Blackwell@matt_blackwell·
I know the AI hype can drive anxiety about academic productivity, but please remember that an additional paper can have negative returns to both society and to your reputation.
English
1
2
45
8.1K
Seth Warner
Seth Warner@sethbwarner·
@ML_Burn Thanks, so good to hear someone put it clearly! Just between Claude Code and the regular ChatGPT, the former is a waaaayyy better planner, and I've found that's where most of my time-savings come from. (I'm not doing anything too complicated in R - ha!)
English
0
0
0
43
Mike Burnham
Mike Burnham@ML_Burn·
@sethbwarner My experience has been that both trivialize scientific computing in Python so long as you’re managing context well and keeping repo documentation current. You’ll get more tokens with OpenAI, but Claude is a better planner and I’d rather give my money to Anthropic.
English
1
0
5
176
Seth Warner
Seth Warner@sethbwarner·
Has anyone been using Codex and the most up-to-date OpenAI products for help with academic research? I hear it's cheaper than Claude but the main advantage is with coding.
English
1
0
1
364

Keşfet

@ML_Burn @zacharylhertz @jdbk @bwitkovsky @mungowitz @MichaelRKistner @UptonOrwell @EfrenPoliPsy