We all know about omitted variable bias: when X⬆️Y, Z is ➕ correlated w/ X & Y, & we omit Z, it biases the effect of X on Y upward (too positive).🥱 But what about when the true effect is 𝐧𝐞𝐠𝐚𝐭𝐢𝐯𝐞? Is OVB just the mirror opposite, biasing the effect to be too negative? No.👇

Hoping to coin a new term: parAInoia Definition: The increasing feeling of dread (felt among teachers/professors while grading a paper) that a student's work was written, at least in part, by AI. Example: "It took forever to get my grading done because I was feeling intense parAInoia."

Stata users: if you aren't already, definitely start using bookmarks to create headings and sub-headings in your .do files. Such a big help! **# in a .do file creates a heading; **## = subheading; **### = sub-subheading. Then double-click on them to jump right to that section 👍

Always amazed how, each semester, teaching data analysis inevitably teaches me something new. This week I learned that spatially correlated residuals in FE models (w/ data from Bailey's great book) probably doesn't require clustering SEs. Thx to @nickchk.com's AMAZING website--what a resource! 🙏

I think a lot about how quant & qual methods relate/differ. At its best, qual research feels like shining a huge light inside the black box that quant folks often ignore. 👍 At its worst, qual research feels like a scatterplot for which the researcher gets to choose which dots to show. 😬

Join us tomorrow as we repeat ourselves. Join us as we repeat ourselves tomorrow. Tomorrow, we repeat ourselves. Join us!

Wrote up a little intervention post/explanation for my class about why using LLMs for trying to learn programming (as first time learners!) is bad and detrimental datavizf25.classes.andrewheiss.com/news/2025-11...

ANES Data Release! electionstudies.org/data-center/... The 3-wave ANES panel is now available. It merges data from 3 election studies (2016-2020-2024), the first time the ANES has collected interviews of the same respondents across 3 presidential elections.

Halloween fun fact 🎃: "Skeleton" has its own IMDB page. 💀 In fact, it stars in one of my favorite horror movies as a kid: House on Haunted Hill (1959 version). Not necessarily the best performance in the film, but also not bad given that it lacks a central nervous system.

🚨 I created a new guide on how to make nice dot charts. These can be really informative, yet remain weirdly underutilized. The examples are with @statacorp.bsky.social but the logic applies to all software. Link in thread. 👇 Made it mostly for my students but hope you find it useful, too! 😁

A short-and-sweet guide to developing research questions published in @nature.com. Lots of good advice here 👍

Are you or one of your students considering doing a Ph.D. in a social science? I've spent a lot of time talking about this w/ students & finally wrote something up. IMO, there are only 3 good reasons to do it. One of them needs to be true--otherwise, don't. medium.com/the-quantast...

🚨 Very flattered to have a piece, w/ the terrific @miacosta.bsky.social, in the new @apsa.bsky.social Experiments Newsletter! In it, we offer some thoughts on how to think about how "realistic" experimental results are. Might be useful for students 😁 connect.apsanet.org/s42/newslett...

If you use Stata and you want to make nice coefficient plots, this article by @uptonorwell.bsky.social is really fantastic! medium.com/the-stata-ga...

Just submitted book revisions 🥳 and wanted to write something a little more "fun." I've been wanting to write this one for a long time and just posted it on Medium (no paywall 👍). For the profs out there, it might serve as a useful class reading (?). Hope you enjoy! medium.com/@jvk221/3-st...

Inflation was a big issue--maybe even *the* issue--in 2024. But do citizens understand how to interpret inflation rates? Using some new data from @verasight.bsky.social, the answer seems to be largely: no. On top of that, Republicans show significantly less understanding than Democrats.

Truly wild to see that this just hit 300 citations on GS 🤯 It's a good lesson for PhD students--this started as a short homework assignment for a class on causal inference. At first I just wanted to know if MC placement matters. But over time, it turned into much more & led to several more papers.

My article on incumbency advantage in presidential elections is now out w/ issue and page numbers 🥳 Key finding: Being an incumbent president comes w/ inherent electoral advantages after 1 term. After 2 terms, a party is not inherently *dis*advantaged--just no more incumbency advantage.

A look at partisanship in the U.S. right now--what partisans will tolerate, what they won't, and where the parties differ. (From an ongoing project with @iganson.bsky.social; n = ~1,100; graph made in @statacorp.bsky.social 😊)

Hard to express what a relief it is to have just submitted the first complete draft of my book manuscript to @ucpress.bsky.social 😮‍💨🥳 5+ years in the making. 249 pages. 82,000 words. 0 ChatGPT. 😉 A million thanks to all those who gave me guidance and inspired me along the way! 🙏

Rs forgetting about the deficit is a classic case of what Kane & Anson call "deficit attention disorder." Their classic article: link.springer.com/article/10.1... @uptonorwell.bsky.social @iganson.bsky.social

For those currently writing and/or grading papers, always good to keep this distinction in mind

Do most people understand how inflation works? That drops in the inflation rate don't mean drops in prices? 🤔 No. And misunderstanding is significantly associated with vote choice in 2024. Thanks to @verasight.bsky.social and @peterenns.bsky.social for letting me field this question 🙏

Had a great time presenting in @stecula.bsky.social 's PhD class on null results in experiments. Thank you again, Dom! If anyone who teaches experiments would like me to present for a course in the Fall, just email or DM me--happy to do it! 😁

Had a great time presenting on null / non-significant results in survey experiments to Samara Klar's Ph.D. class at University of Arizona! 😁 🚨If you teach a course on experiments in the Fall and would like me to present, feel free to email/DM me--I'd be happy to do it! polisky

🚨New WP! What is political bias in its most extreme form? What factors are associated with it? And how can it be reduced? At next week's @MPSAnet conf., @iganson.bsky.social & I will present our paper "Rejoining the Cult" that offers answers these questions: go.shr.lc/4j9bg9j 🧵

Happy to share a new working paper with @uptonorwell.bsky.social! The subject of partisan bias is only increasing in its importance, and in this paper we conceptually distinguish "blatant bias:" i.e., nothing can dissuade a partisan's support for a leader. preprints.apsanet.org/engage/apsa/...

🚨New WP! What is political bias in its most extreme form? What factors are associated with it? And how can it be reduced? At next week's @MPSAnet conf., @iganson.bsky.social & I will present our paper "Rejoining the Cult" that offers answers these questions: go.shr.lc/4j9bg9j 🧵

A perfect encapsulation of our current moment: Celebrating the destruction of the Dept. of Education whilst misspelling “Department”

This is a really thoughtful take on my recent @jepsjournal.bsky.social paper. Carlisle looks at my 7 Alternative Explanations from a statistical power perspective—highly recommend!!! Thank you, Carlisle!

Super important and timely new study by some great folks at @stonybrooku.bsky.social 🙌

For anyone who's just becoming interested in politics, navigating it must be harder than ever. Exhibit A: This is the *WH's* own response to the charge that Trump confused transgender mice w/ transGENIC mice. They cite the studies, NONE of which appears to be aimed to "making mice transgender" 🤷🏻‍♂️

Quick note that a fun way to visualize correlations in @statacorp.bsky.social is with Jann's "heatplot" package.👍 Install with -ssc heatplot- , then run -help heatplot- & scroll to the correlation matrix example. *Maybe* will also make it more fun for students to learn correlation analysis 🤔

🚨It's finally got an issue and page numbers 😊 If you ever do survey experiments and maybe, just maybe, sometimes get null results, this article might help you (plus it's Open Access!). www.cambridge.org/core/journal...

🚨It's finally got an issue and page numbers 😊 If you ever do survey experiments and maybe, just maybe, sometimes get null results, this article might help you (plus it's Open Access!). www.cambridge.org/core/journal...

🚨New WP! Congress makes many bills but passes very few laws. 🤔 Nicole Huffman, David Stack, & I show that introducing "symbolic legislation"--i.e., bills with no hope of becoming law--comes with sizable benefits from one's supporters. preprints.apsanet.org/engage/apsa/...

Today I gave perhaps the most intimidating presentation I've ever given: Intro to Data Visualization. I normally don't sweat it at all, but this time the audience was...my son's kindergarten class😬 Fortunately, @statacorp.bsky.social has a lot of tricks that made it a little more fun 👍

The four phases of having your kids be off from school for winter break for 16 straight days...

Absolutely love the New Year’s Eve “Twilight Zone” marathon and, honestly, this graph might be one of the most sinister things to have ever appeared on the show 😵‍💫 @graphcrimes.bsky.social

📊Ever use or teach interactions in regression? I’ve been using/teaching this tricky concept for years & finally decided to write a guide. Link below; no paywall👍 Meant for users of any software but @statacorp.bsky.social code is provided 😁 CC: @asjadnaqvi.bsky.social medium.com/the-stata-ga...

📊Ever use or teach interactions in regression? I’ve been using/teaching this tricky concept for years & finally decided to write a guide. Link below; no paywall👍 Meant for users of any software but @statacorp.bsky.social code is provided 😁 CC: @asjadnaqvi.bsky.social medium.com/the-stata-ga...

🚨Stata users: Stata is now on Blue Sky! 😊 @statacorp.bsky.social Polisky

🚨Stats folks: Remind your stats students: Run your correlation, run your regression, but ALWAYS look at the underlying data. Here we have r and beta basically equal to 0. And yet...it kinda looks like X and Y might be related 🤔 Thanks to @albertocairo.bsky.social for creating the dataset! 😁🙏

Profs who use Stata: If you ever show your students (anonymized!) data on assignment scores, I highly recommend the "graph matrix" command to produce multiple scatterplots b/w different assignments 👍 graph matrix var1 var2 var3, half

🚨new & improved version of @miacosta.bsky.social & my paper on inattentiveness in conjoint experiments! We propose using "conjoint attention checks" (CACs) to assess respondents' attention to the conjoint itself & find big differences in treatment effects 👍 preprints.apsanet.org/engage/apsa/...

We like to think that we make our own political choices. Looking at some @electionstudies.bsky.social data, though, the relationship b/w parents' party and our own is striking. Even allowing for some misreporting, I'm surprised how strong this has been, and over multiple generations.

Graph of margins of victory in presidential elections over the past 100 years (using Stata). One can only wonder how long the trend can stay at near-zero before some kind of major break or realignment. 🤔

Either I’m having end-of-semester hallucinations again, or I just received an early Christmas gift from Stata??!! 🤯 I’m genuinely touched—thank you, @Stata!!! 🙏🥹

For all the folks new to BlueSky, here's the type of stuff I study (thanks to @daveamp.bsky.social for alerting me to Scholar Goggler!)

Methods/stats profs: if it’s that time of the semester where you’re teaching correlation & regression, I made a post that covers a subtle but key difference b/w the two. No paywall and has gotten a surprising amount of engagement, so it might help! 👍 medium.com/the-stata-ga...