iiinova — where your digital self keeps evolving

I just collected more data on first- and second-order misperceptions and will post here at a later date when I post the results 5/5

because estimating (often obscure) quantities is challenging. So telling someone the correct answer will likely update an answer they subsequently give, but it likely won't affect existing attitudes (especially since the attitudes are likely not caused by the estimate to begin with). 4/n

Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...

Our 2025 PNAS paper (www.pnas.org/doi/10.1073/...) offers one explanation for this: anytime people estimate quantities (% of population that is foreign-born, % of Americans who own an Apple product, % of dots on a page that are blue), they make similar errors ... 3/n

pnas.org

PNAS

Barnfield et al.'s paper looks similar: they find that providing correct information leads to more accurate estimates of thing people are estimating (here, second order beliefs), but not attitude-related outcomes ("policy feasibility perceptions, attitudes, and behavioural intentions"). 2/n

Exactly. This is also in line with a large body of work that shows correcting misperceptions (including numeric ones, see Hopkins et al. and Thorson's work on estimates related to immigration in the JOP and APSR, respectively) often results in updated estimates, but not attitudes. 1/n

Brian Guay @brianguay.bsky.social · 04/11/2025

Published a political psychology book this year? @erincassese.bsky.social, @lchristensen.bsky.social, and I are on the Robert E. Lane Award committee for the best book in political psych published in the past year Deadline for nominations: March 1, 2026. Details below apsanet.org/membership/o...

apsanet.org

Section28 - American Political Science Association (APSA)

Political Psychology (Section 28) The purpose of this section to facilitate communication across subfields and disciplinary boundaries among individuals

072

Brian Guay @brianguay.bsky.social · 24/09/2025

UNC Political Science is hiring in methods! UNC has an amazing department and the triangle is a great place to live. Tenure Track Assistant Professor in Methods (Deadline Oct 24) link to the posting👇 unc.peopleadmin.com/postings/307... #polisky #psjobs #poliscijobs

unc.peopleadmin.com

11513

Reposted by Brian Guay

Will Lowe @conjugateprior.org · 18/09/2025

Every time this worry comes up (www.ft.com/content/d419...) I post some Landy et al. (2018). People just answer questions about proportions (of anything) in a rather particular way. So I think it's unlikely that what they are being asked about is as important as you might expect it should be.

issues, and the results on immigration are
sobering. UK respondents dramatically overestimate how many immigrants there are and how many of them are Muslim. They overestimate what proportion of immigrants are from north Africa by a factor of 10, and from the Middle East by a factor of two, and underestimate how many are from North America. Respondents underestimate how many are Christian, and also underestimate the education levels and employment status of immigrants relative to the UK-born population. These misperceptions are not unique to the UK - they are common in rich countries.

People seem to answer proportion questions in log odds space, not probabilities. And when you know that, there seems a lot less substantive interpretation worth doing.

42211

Brian Guay @brianguay.bsky.social · 18/09/2025

@fguelzau.bsky.social Here's the ungated draft of the PNAS paper: www.brianguay.com/files/guay_2...

brianguay.com

010

Reposted by Brian Guay

Brendan Nyhan @brendannyhan.bsky.social · 27/06/2025

Zaller remains undefeated

14410

Brian Guay @brianguay.bsky.social · 09/04/2025

thanks so much Conrad, I'll fix this!

010

Thanks to my fantastic co-authors @tylermarghetis.bsky.social , @david-landy.bsky.social , Cara Wong and everyone who gave us feedback over many years Ungated earlier version of the paper here: www.brianguay.com/files/guay_2...

brianguay.com

0130

Our findings suggest that the public knows more about politics than we give them credit for: People make errors when estimating politically-relevant percentages, but this is due to the format of the question not underlying misinformation about what they are estimating

2130

Of course, characteristics of specific groups may matter, but only at the margins. We should first account for the domain-general errors people make *anytime* they estimate a percentage, then examine group-specific explanations

1111

The same is true of theories that people overestimate the size of groups that they have a lot of social contact with. Very little evidence of this!

1110

We also test popular theories that people overestimate the size of groups they fear. Not the case. Again, misestimates result mainly from the psychological errors we make anytime we estimate %s, not from anything specific to the group being estimated

1153

We argue that this pattern of over-under estimation arises from 🧠Bayesian reasoning under uncertainty🧠: people often have uncertain ideas in their minds about the size of these groups, but when they convert these ideas to percentages they ‘hedge’ their estimates toward a prior

2161

And this is the same pattern of errors people make when estimating things like the percentage of dots on a page that are red 👇

1181

Here’s the key figure: people make the same estimation errors regardless of what they are estimating---political and *entirely non-political* quantities. These are 100k estimates of the size of racial and non-racial groups made by 37k people in 22 countries

1213

Instead, people are just really bad at estimating percentages They systematically overestimate smaller %s and underestimate larger %s, including ENTIRELY NON-POLITICAL %s, such as the % of the population that owns an Apple product, has a passport, or has indoor plumbing

2317

We argue that journalists and academics are *wrong* when they interpret these misperceptions as evidence that the public is ignorant and misinformed 👇

2203

Mike Wagner @prowag.bsky.social · 03/02/2025

New paper on misperceptions out in PNAS @pnas.org www.pnas.org/doi/10.1073/... Why do people overestimate the size of politically relevant groups (immigrant, LGBTQ, Jewish) and quantities (% of budget spent on foreign aid, % of refugees that are criminals)?🧵👇

1226896

Reposted by Brian Guay

This is not a time for passive citizenship. Silence means approval. So what to do?

4251872

Brian Guay @brianguay.bsky.social · 21/01/2025

I'm very happy to share that I'll be joining the Department of Political Science at UNC Chapel Hill as an Assistant Professor this fall. I'm excited for this next chapter and will always be incredibly grateful for my amazing experience at Stony Brook.

2250

Brian Guay @brianguay.bsky.social · 13/12/2024

I haven't been on bsky for almost a year. It's nice over here!

But the MOST important thing is that researchers *justify* their research design & analysis approach on normative/theoretical grounds and *pre-register* it Doing so will help prevent researchers from talking past each other and move toward tackling problem of misinfo

Key takeaway: choose the design that aligns with your normative claim about how people should interact with information e.g., the normative claim that aligns with discernment is that people should maximize accuracy of the content that they believe and share

100

We demonstrate these differences empirically by re-analyzing data from recent misinformation studies Different research designs and outcomes = different conclusions about whether misinformation interventions work

111

We recommend designs that measure discernment over those that measure only belief/sharing of false content Importantly, these two designs can lead to different conclusions about whether misinformation interventions work👇

111

A design that addresses these issues exposes people to a mix of true and false content, and measures the difference between ratings of each This difference is called *discernment*, and measures how well people can discern between true and false content

101

and 2. These studies conflate believing/sharing false content with believing/sharing *all* content. Without exposing people to true content as well, we don’t know whether an intervention makes people skeptical of false content or all content.

101

Some studies measure only if ppl believe/share false content. This is problematic bc: 1. Believing false content (ivermectin treats covid) is bad, *but so is* disbelieving true content (vaccines work). Disbelieving true content is a big problem, esp. since most news is true

122

Brief summary: Who believes and shares misinfo? Why? What can we do about it? Answering these questions requires measuring whether ppl believe and share misinfo But studies purporting to answer the same question often use different research designs, inhibiting progress on combating misinfo

121

How should researchers determine whether misinformation interventions work? In a recent paper in Nature Human Behavior (t.co/m3YaS7VaJl) we argue that researchers should 1) measure whether people believe/share both false AND true content and 2) assess efficacy using a measure of discernment.

121

Brian Guay @brianguay.bsky.social · 07/02/2024

*Less than half* of misinformation studies expose people to both false and true info, and ONLY 7% measure discernment (the difference between believing/sharing true and false info)!! misinforeview.hks.harvard.edu/article/what... Short 🧵 on why you should be measuring discernment👇

33924

Headed to SPSP2024 in San Diego to talk about how we should measure whether misinformation interventions work! Let me know if you're gonna be there (DM/email/text).

Lisa Fazio @lkfazio.bsky.social · 02/02/2024

Reposted by Brian Guay

Less then a week until the 2nd Misinformation and Belief Science preconference at SPSP!! Really excited for the lineup Rakoen Maertens and I have put together. We've got ...

151

Reposted by Brian Guay

APSA @apsa.bsky.social · 27/11/2023

Please share and apply to the APSA Ralph Bunche Summer Institute, an annual, intensive 5-wk program designed to introduce the world of doctoral study in political science to undergrad students from under-represented racial and ethnic groups. Deadline: Jan. 22 politicalsciencenow.com/call-for-app...

Apply to the APSA Ralph Bunche Summer Institute (RBSI), an annual, intensive 5-wk program designed to introduce the world of doctoral study in political science to undergrad students from under-represented racial and ethnic groups.

03355

Reposted by Brian Guay

Rafael M Batista @rafmbatista.bsky.social · 31/10/2023

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/...

107738

Brian Guay @brianguay.bsky.social · 03/10/2023

The last time there was a "presidental alert" Jesse Lopez & I wrote a POQ paper (doi.org/10.1093/poq/...) about it. Back then I thought it was crazy that this was being framed as such a polarizing issue. Enter 2023...

120

Reposted by Brian Guay

Reed Orchinik @rorchinik.bsky.social · 26/09/2023

Very happy to share a new working paper with some great coauthors! 97% of climate scientists agree that human-caused climate change is occurring. But how do people interpret this consensus? Is it a testament to expertise or a signal of bias? psyarxiv.com/ezua5/

13720

Brian Guay @brianguay.bsky.social · 25/09/2023

In addition to the junior line in American Politics, we have a senior line in political psychology at Stony Brook! apply.interfolio.com/131893 polisky psjminfo

apply.interfolio.com

Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio

010

Brian Guay @brianguay.bsky.social · 25/09/2023

Job alert! We're hiring an assistant professor (American Politics) in the Department of Political Science at Stony Brook! Drew Englehardt and I started this semester and it's been great! apply.interfolio.com/130609 10/13 deadline polisky psjminfo poliSciJobs

apply.interfolio.com

Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio

12013

Reposted by Brian Guay

Brendan Nyhan @brendannyhan.bsky.social · 24/09/2023

Kids: Study statistics

1192

Brian Guay @brianguay.bsky.social · 24/09/2023

Just joined! Seems nice over here!