Devin Caughey
LightNews LightNews
Devin Caughey
@devincaughey.bsky.social
1.1K followers 2K following 20 posts

Political science professor at MIT

Videos
Political science 59%
Mathematics 17%
Posts Replies Media Videos

Reposted by Devin Caughey

rpgriffin.bsky.social
Robert Griffin, Phd @rpgriffin.bsky.social · Oct 7
New piece from me @chriswarshaw.bsky.social , @devincaughey.bsky.social , and Bernard Fraga.

Story: though we've entered a high turnout era, turnout gaps between racial groups have reemerged.

Nerdy story: Using publicly available data, we've generated new, reliable turnout estimates over time.
Devin Caughey @devincaughey.bsky.social · Sep 16
Helpful data on our current situation:

www.cato.org/blog/politic...
Politically Motivated Violence Is Rare in the United States
The government can and should vigorously pursue justice for Kirk and all the others murdered by politically motivated terrorists, but it can and should do so without new political witch hunts, expande...
www.cato.org

Devin Caughey @devincaughey.bsky.social · Mar 14
The logical dependence in the outcome would seem to rule out many standard methods for panel causal inference. In other contexts one would fit a survival model, but that sort of thing isn't too common in panel CI. Thoughts? Maybe PanelMatch would work because it matches on lagged Y?
Devin Caughey @devincaughey.bsky.social · Mar 14
(Trying the sort of thing that used to work on Twitter):

Suppose you have panel data with a binary treatment and an outcome that indicates whether something has happened yet (e.g., whether a policy has been passed). How would you estimate treatment effects? 1/2

#polisky

Reposted by Devin Caughey

polanalysis.bsky.social
Political Analysis @polanalysis.bsky.social · Feb 26
Currently in FirstView: “MODGIRT: Multidimensional Dynamic Scaling of Aggregate Survey Data” by Elissa Berwick and @devincaughey.bsky.social. While traditional public opinion models are restricted to unidimensional latent traits, MODGIRT allows for a multidimensional structure of preferences.

Reposted by Jan Eeckhout, Devin Caughey

ericcolumbus.bsky.social
Eric Columbus @ericcolumbus.bsky.social · Dec 20
100 years ago Saturday the NYT ran a story that did not hold up well at all.
www.nytimes.com/1924/12/21/a...
Headline: HITLER TAMED BY PRISON. BERLIN, Dec. 20 — Adolph Hitler, once the demi-god of the reactionary extremists, was released on parole from imprisonment at Fortress Landsberg, Bavaria, today and immediately left in an auto for Munich. He looked a much sadder and wiser man today than last Spring when he, with Ludendorff and other radical extremists, appeared before a Munich court charged with conspiracy to overthrow the Government. His behavior during imprisonment convinced the authorities that, like his political organization, known as the Volkischer, was no longer to be feared. It is believed he will retire to private life and return to Austria, the country of his birth.

Devin Caughey @devincaughey.bsky.social · Dec 13
Re methods research, Stephen Jessee and @jerzakconnor.bsky.social have a working paper on dealing with measurement error in latent variables. It too deprecates MOC and proposes an IV-based approach instead.
Devin Caughey @devincaughey.bsky.social · Dec 11
Put differently: even the best scholars produce bad work, and even more often produce work that is not a good fit for a given outlet. A paper is not a person.
Devin Caughey @devincaughey.bsky.social · Dec 11
Yeah I hear you on me(). MO is faster. The advantage of both over MOC is allowing other variables in the analysis model to inform estimates of the missing values rather than assuming them to be independent as MOC does (see the appendix to Treier and Jackman).
Devin Caughey @devincaughey.bsky.social · Dec 10
It’s true! Though I have come to the conclusion that MOC is more assumption-dependent than is commonly realized, and it is often better to use @mattblackwell.bsky.social et al.’s “multiple overimputation” or to use Bayesian measurement-error model, which brms makes pretty easy.
Devin Caughey @devincaughey.bsky.social · Nov 17
I am Gen X (barely) and have said “on accident” as long as I can remember. Didn’t realize it was nonstandard until I was made fun of for it it as an adult.

Reposted by Devin Caughey

dfreelon.bsky.social
Deen Freelon @dfreelon.bsky.social · Nov 1
Couple more MIT ones... first, a TT line on the "social, economic, and ethical implications of computing and networks, with specific focus on the Future of Work and the evolving interface between Artificial Intelligence (AI) and Human Interaction" apply.interfolio.com/156476
Apply - Interfolio {{$ctrl.$state.data.pageTitle}} - Apply - Interfolio
apply.interfolio.com

Reposted by Devin Caughey

decentfilms.bsky.social
Steven D. Greydanus @decentfilms.bsky.social · Oct 28
What makes a great comic strip work? For the week of Halloween, an appreciative breakdown of a classic monster tale from Gary Larson’s The Far Side, with more in-depth critical analysis than you probably want or would think possible!
Cartoon Critic #1: A Shocktober “The Far Side” commentary
(Gary Larson’s tragic tale of Edgar, a werewolf, and an obnoxious tie)
greydanus.substack.com

Devin Caughey @devincaughey.bsky.social · Sep 22
Theoretically either is sufficient to drive the approximate bias to zero, but given that this is impossible in practice, it’s definitely better to do both. Also, doing it for the outcome has the additional advantage of reducing variance.
Devin Caughey @devincaughey.bsky.social · Sep 22
For a theoretical treatment see Sarndal and Lundstrom’s 2006 textbook, which shows that the approximate bias under calibration is zero only if the outcome or inverse response probabilities are a linear function of the auxiliary vector. In raking the aux vec is the marginal dist; in PS the joint.
Devin Caughey @devincaughey.bsky.social · Sep 16
I mean short “I”
Devin Caughey @devincaughey.bsky.social · Sep 16
On my first listen I was pretty sure that it was just an aborted mispronunciation of “migrants” with a short a (as in “immigration”), but after a second listen it sure does sound a lot like something else.
Devin Caughey @devincaughey.bsky.social · Sep 8
If you’re looking for more inspiration, you can check out my “political science scope and methods” syllabus, which has a few sessions on philosophy of (social) science: devincaughey.github.io/files/syllab....
devincaughey.github.io

Devin Caughey @devincaughey.bsky.social · Sep 8
For general overviews, I like Godfrey-Smith’s textbook and the Cartwright and Montuschi Philosophy of Social Science edited volume.
Devin Caughey @devincaughey.bsky.social · Aug 22
Looks great! And just in time to make my syllabus for the fall.
Devin Caughey @devincaughey.bsky.social · Dec 22
I’m honestly a little surprised bc my impression was that speed improvements from within-chain parallelization were nonmonotonic in the number of cores due to the fixed costs of each additional core, and 64 is way beyond the
point where I thought performance declined. But maybe that’s outdated?
Devin Caughey @devincaughey.bsky.social · Dec 21
64!

Reposted by Jared Rubin, Devin Caughey

sean-gailmard.bsky.social
Sean Gailmard @sean-gailmard.bsky.social · Dec 20
Coming in 2024! Cover art day is a good day.

There are also blurbs and you can even order a copy: tinyurl.com/4e35mjef

Intro chapter: www.ocf.berkeley.edu/~gailmard/em...
Gailmard, Agents of Empire: cover

Reposted by Devin Caughey

polisciatnu.bsky.social
Political Science Department at Northwestern University @polisciatnu.bsky.social · Nov 11
🚨Forthcoming Must Read!!!

The Political Development of American Debt Relief
Emily Zackin and Chloe N. Thurston (@chloethurston.bsky.social)

Available now for pre-order, May 2024 UChicago Press Polisky press.uchicago.edu/ucp/books/bo...
The Political Development of American Debt Relief Emily Zackin and Chloe N. Thurston A political history of the rise and fall of American debt relief. Americans have a long history with debt. They also have a long history of mobilizing for debt relief. Throughout the nineteenth century, indebted citizens demanded government protection from their financial burdens, challenging readings of the Constitution that exalted property rights at the expense of the vulnerable. Their appeals shaped the country’s periodic experiments with state debt relief and federal bankruptcy law, constituting a pre-industrial safety net. Yet, the twentieth century saw the erosion of debtor politics and the eventual retrenchment of bankruptcy protections. The Political Development of American Debt Relief traces how geographic, sectoral, and racial politics shaped debtor activism over time, enhancing our understanding of state-building, constitutionalism, and social policy.

Devin Caughey @devincaughey.bsky.social · Oct 18
And either way, it’s bad for the country!
Devin Caughey @devincaughey.bsky.social · Oct 18
Two important differences I think. 1) Biden may not have enough control over the party and affiliated groups to prevent a candidate from running; 2) unlike Netanyahu, Biden’s tenure in office does not depend on a congressional majority—in fact - GOP House might increase his reelection prospects!
Devin Caughey @devincaughey.bsky.social · Sep 30
I like the description of Ariely as “enigmatic swami of the but-actually circuit”

Reposted by Andrew Reeves, Devin Caughey

brendannyhan.bsky.social
Brendan Nyhan @brendannyhan.bsky.social · Sep 23
How has legislative attention to race changed over time? Dramatic growth in racialized words in legislation, but concentrated in symbolic resolutions, not substantive bills drive.google.com/file/d/11WPw... - new JMP from Amber Mackey (on poli sci market from Penn, MIT predoc)