Martin Devaux

Political Science PhD candidate at @Columbia. Political economy, immigration, far-right populism, and political methodology. 2025/2026 Job Market.

[Columbia University Political Science @Columbia_PolSci]

🎓 Our first job market spotlight is on Martin Devaux

🔗 Website: https://t.co/QtsdEoXblK
🕊️ Follow on X: @MartinDevaux

[Martin Devaux @MartinDevaux]

@AnjaliVBhatt Zotero !!!

[Martin Devaux @MartinDevaux]

Repost le matin

[Martin Devaux @MartinDevaux]

Bonjour ! Les cahiers de doléance issus du Grand Débat National en 2018 après la crise des Gilets Jaunes sont-ils numérisés et centralisés quelque part ?
#sciencepolitique #sciencessociales #scienceshumaines #sociologie #histoire @ScPoResearch

[Martin Devaux @MartinDevaux]

Bonjour ! Les cahiers de doléance issus du Grand Débat National en 2018 après la crise des Gilets Jaunes sont-ils numérisés et centralisés quelque part ?
#sciencepolitique #sciencessociales #scienceshumaines #sociologie #histoire @ScPoResearch

[Khoa Vu @KhoaVuUmn]

🚨 New working paper🚨

Been working on this for a while, so I'm really excited to share this. I propose a new estimator combining the strength of existing research designs, along w/ a Stata command. This is still very much in developing stage so feedback is welcome!

[Khoa Vu @KhoaVuUmn]

Econ seminar US
audience government
🤝🤝
Shooting down unidentified objects

[Martin Devaux @MartinDevaux]

@Pronk__ little guy with big guy complex

[Martin Devaux @MartinDevaux]

@MAranzazuRU Ah that sounds more or less like what I'm doing! Thanks!

[Martin Devaux @MartinDevaux]

@MAranzazuRU Do you integrate it with Zotero? I don't really know how to smoothly use both simultaneously

[Martin Devaux @MartinDevaux]

Yes

[Cyrus Samii @cdsamii]

I have a new "methodology big think" essay on the "problem solving" approach to social science. More here, along with a link to a PDF of the essay. I would love to hear what people think.
https://t.co/vJI1Looaml

[Martin Devaux @MartinDevaux]

In case it's useful for anyone: because it's still a random draw due to the sampling variability

[Martin Devaux @MartinDevaux]

Why should the expected value of the Q statistic be k-1, and not 0, if all the study estimates are equal?

I understand that it follows a chi-squared distribution that, in general would have mean = k-1, but wouldn't this be a conditional exp given Y_i = M for all i?

#Stats

[Martin Devaux @MartinDevaux]

@MAranzazuRU GoodNotes has been amazing for me (with pencil too)

[Martin Devaux @MartinDevaux]

@cblatts @JulianEGerez @elena_barham?

[Martin Devaux @MartinDevaux]

🔥❤️

[Omar Sy @OmarSy]

💁🏿‍♂️

[Martin Devaux @MartinDevaux]

re: interpretable machine learning. In what ways do y'all like to summarize/compare model fits?
Being agnostic about the model itself: could be tree ensemble or even "blacker box" models.

PDPs? Variable importance? What else?

#rstats #TidyTuesday #r4ds #tidymodels #DataScience

[Martin Devaux @MartinDevaux]

@CedScherer Oops number 2 didn't make it (https://t.co/E2MSVfXDFv)

[Martin Devaux @MartinDevaux]

@CedScherer Plugging mine here 🙄
https://t.co/WAPdbGzLRI
https://t.co/E2MSVfX5PX

[Martin Devaux @MartinDevaux]

Why should the expected value of the Q statistic be k-1, and not 0, if all the study estimates are equal?

I understand that it follows a chi-squared distribution that, in general would have mean = k-1, but wouldn't this be a conditional exp given Y_i = M for all i?

#Stats

[Martin Devaux @MartinDevaux]

@ChelseaParlett I am trying to understand the Q statistic in the context of meta-analysis and found your cool post (https://t.co/kJkSKaFFKi tysm for writing them all 🔥) and I have a question.

Anyone else's help is super appreciated!!