People feeling hopeless about AI/LLMs spoiling all online polling and survey research should take a look at what @yamilrvelez.bsky.social is doing with “Pulse” — a tool to detect proof of life that’s compatible with Quatrics

An AI Voter bot improves knowledge about politics But, the AI bot has weak effects on downstream outcomes like vote preferences and party evaluations among respondents whose primary issue position aligns closely with one of the parties. Partisan action is hard to change. www.pnas.org/doi/10.1073/...

This paper was a blast to work on. The challenge: present party positions across many issues, in real time, using language voters actually use. 🧵 on why we went with a more involved retrieval-based approach and where I think these tools are headed.

🚨Excited to share our new paper published in PNAS (joint with @yamilrvelez.bsky.social and Don Green)! AI can enhance political knowledge and provide balanced information about politics with proper guardrails and vetted sources (e.g., party platforms). www.pnas.org/doi/full/10....

Looking for a tl;dr for these two excellent papers? I've got you covered: www.science.org/doi/10.1126/...

🚨 New in Nature+Science!🚨 AI chatbots can shift voter attitudes on candidates & policies, often by 10+pp 🔹Exps in US Canada Poland & UK 🔹More “facts”→more persuasion (not psych tricks) 🔹Increasing persuasiveness reduces "fact" accuracy 🔹Right-leaning bots=more inaccurate

As @seanjwestwood.bsky.social's terrifying new PNAS article demonstrates, LLMs can now pass almost every attention check, mirror personas, stay consistent across pages, and systematically bias responses in the aggregate. So here’s a different angle: verify physical presence, not text.

#OpenAccess from @polanalysis.bsky.social - Crowdsourced Adaptive Surveys - https://cup.org/4pNclb0 - @yamilrvelez.bsky.social#OpenAccess

The latest issue of PA is out now. We have a great collection of papers by @yamilrvelez.bsky.social, @sysilviakim.bsky.social, @mattblackwell.bsky.social, @sophieehill.bsky.social, @dwlee.bsky.social, @melissazrogers.bsky.social, @kaipingchen.bsky.social, @samuelbaltz.bsky.social (1/2)

Ever stared at a table of regression coefficients & wondered what you're doing with your life? Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social) Preprint: doi.org/10.31234/osf... Website: j-rohrer.github.io/marginal-psy...

Today, @catiebailard.bsky.social and I take the reins at GW's Institute for Data, Democracy + Politics! We'll work to make IDDP a one-stop shop for research on misinfo, platform manipulation, and threats to democratic ideals. To follow our work + learn about opportunities, visit: bit.ly/45GLggi

I had a blast presenting my work on tailored experiments and adaptive surveys at the Summer Institute of Computational Social Science. Nothing better than a room full of engaged and smart people getting into the weeds about algorithms, causal inference, and the messy realities of working with AI.

From our new issue: "Confronting Core Issues: A Critical Assessment of Attitude Polarization Using Tailored Experiments " by YAMIL RICARDO VELEZ (@YamilRVelez) and PATRICK LIU (@patrickpliu) #APSRNewIssue www.cambridge.org/core/journal...

Currently in FirstView: In “Crowdsourced Adaptive Surveys,” @yamilrvelez.bsky.social introduces a methodology (CSAS) that converts open-ended text from participants into survey items and applies a multi-armed bandit algorithm to determine which questions should be prioritized in the survey.

Another survey methods talk at 𝗡𝗬𝗔𝗔𝗣𝗢𝗥 coming up next week to put on your calendar: Wed April 23 - 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜 𝗳𝗼𝗿 𝗦𝘂𝗿𝘃𝗲𝘆 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 (led by @yamilrvelez.bsky.social , @joshua-lerner.bsky.social , and myself) RSVP below 👇

🧵 Why do facts often change beliefs but not attitudes? In a new WP with @yamilrvelez.bsky.social and @scottclifford.bsky.social, we caution against interpreting this as rigidity or motivated reasoning. Often, the beliefs *relevant* to people’s attitudes are not what researchers expect.

New paper! AI lets us query humanity’s collective knowledge (or a subset of it) – rapidly addressing any given concern about vaccines in detail. So can information-focused LLM conversations shift vaccination intentions? In an RCT of 1,124 HPV vax-hesitant parents: yes!

New working paper with two great coauthors!

While there is a growing body of work suggesting information can change beliefs, effects on attitudes tend to be muted. We sketch out why this might be the case, drawing attention to the role of belief relevance. See 🧵 below!

🧵 Knowing which questions to ask is one of the most critical aspects of survey design, but we often rely on guesswork to devise questionnaires. What if we were to rely directly on participants?

#OpenAccess from @polanalysis.bsky.social - Crowdsourced Adaptive Surveys - cup.org/4aY2OH4 - @yamilrvelez.bsky.social "This paper introduces a crowdsourced adaptive survey methodology (CSAS) that unites advances in natural language processing and adaptive algorithms..." #FirstView

Special Issue Alert! How do digital & social media fuel alternative identities, extreme narratives & online communities in times of crisis? Submit your work exploring their role in propaganda, misinformation, and reshaping society! ispp.org/wp/wp-conten... @polpsyispp.bsky.social #polisky

Good discussion featuring @profsanderlinden.bsky.social, @yamilrvelez.bsky.social & @sumitra.bsky.social re. Ecker et al´s "Why Misinformation Must Not be Ignored" piece in this thread. Summarizing my thoughts.... 1/5

Interesting perspectives. Feels like some disconnect from standard polisci accounts (e.g. effects of misinfo/fact-checks often have limited scope, but that doesn't mean misinfo is overall inconsequential www.pnas.org/doi/abs/10.1...). Was also curious about this part and dug around on the cites...

Important paper clarifying what reproducibility looks like in the age of LMs. The authors argue that local open-weight LMs are preferable to proprietary ones due to the temporal instability of model outputs.

How concerned should we be about backsliding during the second Trump term? Have past concerns come to fruition? Bright Light Watch asked experts to assign probabilities to concrete "bad" events in several past waves and the most recent one. We collaborated on a report summarizing these predictions.

Psychological inoculation against misinformation does not appear to transfer to real posts or synthetic posts using other misinformation techniques osf.io/preprints/ps... #polisky #polpsych #socpsych #misinformation

What's the best way to debrief study participants? At @pnasnexus.org, Katie Clayton, @yamilrvelez.bsky.social, @thomasjwood.bsky.social and I introduce a new debriefing method that leaves participants much better off than standard debriefs. Especially relevant for misinfo research. bit.ly/41i19sU

When depicted this way, 2008 really appears as the exception 21st century US presidential election.

🎉 Thrilled that my new book, *Creating Partisans: The Organizational Roots of New Parties in Latin America*, is finally out! 🎉 @cambridgeup.bsky.social www.cambridge.org/us/universit... A brief summary: 1/n

Can AI-powered survey interviews boost response quality? Our study shows conversational LLMs elicit richer, more detailed responses vs standard formats. Read the full brief from NORC staff on enhancing survey methodology with GenAI: go.norc.org/3V2G81D

Polsby’s comments on “theory” still ring true for much of the social sciences: stylized facts that guide research agendas are often what we have, but that’s a far cry from the strict deductive apparatus implied by the term.

How much should we conclude based on 2 questions? If we measure voters' views on 14 (culture-war!) issues, many Dems overlap with Independents

High-quality discussions in the replies about what we can learn from “imperfect” research designs. Nature is healing.

New in NHB: We systematically replicated 26 of 41 PNAS social science studies that used MTurk. The approach provides a proof-of-concept for using decision markets to select findings to replicate. Replication effect sizes were 45% of original effect sizes. www.nature.com/articles/s41...

A new piece from me @goodauth.bsky.social: How to think about the “racial realignment” in US politics. Two key points: 1 ) GOP gains among Black and Latino voters are not about gender or the “working class.” They are about IDENTITY and IDEOLOGY. goodauthority.org/news/electio...

Super-interesting approach

🚨 How should we design surveys to capture opinion in fast-changing contexts like elections or within hard-to-reach communities? 🚨 My paper, conditionally accepted at Political Analysis, develops a method that leverages LLMs and adaptive algorithms to construct surveys that evolve with user input.

As the dust settles, it’s worth revisiting the long-running debate about when voters prioritize group-focused policies. Many Latinos no longer view restrictions on immigration as a “deal breaker.” Why?

My and @johannarickne.bsky.social 's paper on the ”Class Ceiling in Politics” is now out in the APSR with open access! We describe the key findings below. 1/9

In the future, we will know the relative contribution of Dem turnout versus vote-switching versus compositional changes in the electorate. But not now!

NSF makes you say who you got conflicts (coauthored) with. We (really just Jordan Matelsky) just built you a tool for that. Literally one click: bib.experiments.kordinglab.com/nsf-coa

Here are the real reasons Latinos are moving to the Republican Party. The big challenge for Democrats: It will be hard to win them back. goodauthority.org/news/the-rea...

from @thomasjwood.bsky.social x.com/thomasjwood/...

Interview with the ever-thoughtful @yamilrvelez.bsky.social on ideological sorting

Latino voters are matching their voting preferences with their ideology & issue views; "once voters learn to connect their ideology to their vote choice, it’s hard to reverse" goodauthority.org/news/the-rea...

We’re going to hear lots of stories about which people, policies and rhetoric are to blame for the Democrats’ defeat. Some of those stories may even be true! But an underrated factor is that 2024 was an absolutely horrendous year for incumbents around the world 👇