Share this page!

Enter URL or ID to Unroll

You can paste full URL like: https://x.com/threadreaderapp/status/1644127596119195649
or just the ID like: 1644127596119195649

How to get URL link on X (Twitter) App

  1. On the Twitter thread, click on or icon on the bottom
  2. Click again on or Share Via icon
  3. Click on Copy Link to Tweet
  4. Paste it above and click "Unroll Thread"!
  5. More info at Twitter Help
Rafe Meager (they/them) Profile picture
Rafe Meager (they/them)
@economeager
Aspiring wastrel, applied econometrician. Life, stats and poverty. Also writes https://t.co/UBlzYddnTE about art, death and emotions 🏳️‍⚧️🏳️‍🌈🇦🇺
Feb 14 • 6 tweets • 2 min read
let's see what DOGE has been cancelling at the department of education. @stuartbuck1 went through the list. will link to his post at the end of the thread. i will start with his 4th example: foundational research to understand how american kids are doing at school Image DOGs has also cancelled an impact evaluation of after-school programs in high-poverty schools with the goals of reducing drugs and violence. the program is intact, but the *evaluation* of the effect of the program, ie what it does or doesn't do to help kids, is cancelled! Image
Nov 27, 2024 • 15 tweets • 4 min read
We still have a relatively poor understanding of the relationship between research and policy. Program evaluation in particular is often motivated by a desire to make policy better. But how effective is program evaluation itself? Michelle Rao's JMP tackles this question. Image You can find this paper and more at her research homepage here assuming Elon does not sink this thread like a stone with his algorithm.

Let me now explain in my view what's so exciting about Michelle's paper.michellerao.com/research
Mar 31, 2023 • 4 tweets • 1 min read
bringing this back why not and this!
Mar 28, 2022 • 4 tweets • 1 min read
procrastinating on my real work by writing simulations of the bad controls problem in R it's the last week of our semester my brain is fried and i just want to have fun with my best friend the rnorm() function
Feb 18, 2022 • 5 tweets • 1 min read
Good day to reread "The Awesomest 7 year Postdoc ever" One thing I found depressing about the discussion we had here last time is people started debating whether the author of this essay is sufficiently morally good as to be worthy of our attention. i'm worried for us if this is where we are going.
Dec 17, 2021 • 4 tweets • 1 min read
Something that has stuck with me from Kristen Neff's excellent book "self compassion" is the exercise where you list 3 socially-valued traits at which you are above-average, average and below average. This is such a good exercise to demonstrate the difference between self-compassion and self-esteem or self-regard. if you love and accept yourself, it's not a big deal to acknowledge that you're average at some stuff, and also have weaknesses alongside your strengths.
Dec 8, 2021 • 4 tweets • 1 min read
sagest advice i can give grad students is to remember table reformatting always takes 1 whole day to do, regardless of how trivial the reformatting initially appears i tweet this stuff because i know how discouraging it feels to experience yourself as "unproductive" over days like this, but some stuff - especially formatting, admin, etc - just actually takes much longer than you would think it does, and people don't like to talk about it
Sep 21, 2021 • 33 tweets • 7 min read
Ok ok so we finally have the new working paper up and I think I have time to do a little thread about it! The paper, joint with @MikeGechter, offers a new method to combine RCTs with observational studies in meta-analysis. We provide both nonparametric identification results and a parametric small-sample implementation for applications! personal.psu.edu/mdg5396/MGRM_C… Image
Jul 28, 2021 • 6 tweets • 1 min read
people misunderstand what it means for a research project to be finished. the project is not finished when i have results or I write them up after reflecting on them. the project is finished when i have reflected on the results so deeply that I become confused by my own work. publishing a working paper should always feel like "here, YOU try and make sense of this"
Jul 15, 2021 • 4 tweets • 1 min read
this is really common in academia and, as the thread accurately goes on to demonstrate, the problem is emotional and psychological, not intellectual or moral. at this point i basically don't think laziness as we conceive of it really exists. If it does exist, then lazy people are HAPPY when they don't do tasks, and a self actualised lazy person would say no to doing tasks. if you're not happy about not doing tasks, you're not lazy.
Jun 17, 2021 • 18 tweets • 4 min read
I've been dismayed by how Forstater vs CGD is being misrepresented by the media, including the BBC in this article bbc.co.uk/news/uk-508589… . Even the twitter moment summary last week was very misleading. As far as I am able to discern, none of the key points below are true: In fact, Maya was not fired from her job, the main problem was not tweets but workplace conduct, and the objectionable speech in question is not the claim "that biological sex is real".
May 21, 2021 • 5 tweets • 2 min read
Ok so chapter 12 of “Noise” by Kahneman, Sibony and Sunstein contains a repeated, incorrect claim about stats / causality: the claim that no correlation implies no causation. Easy to disprove this claim, no math needed, see my next tweet. Imagine driving a car, reaching a hill and pumping the gas as you begin to go up so that your speed is constant. The correlation between pressing on the gas and the speed of the car is zero but they’re obviously causally related, it’s that the agent is optimizing speed!
May 21, 2021 • 7 tweets • 1 min read
Isaiah at the Chamberlain Seminar! T minus 3 minutes!! they dont pay me for this and i am not affiliated with them i'm just a mikusheva / andrews / co groupie :)
May 6, 2021 • 4 tweets • 1 min read
me in grad school: who gives a fuck about the properties of positive semi definite matrices though. nobody. useless nonsense.

me now, programming a simulation of multivariate selection models with jointly gaussian errors: well fuck t minus 5 minutes projected until @rgiordan comes to twitter to tell me off for EVER failing to appreciate the deep importance of the properties of positive semi definite matrices
May 5, 2021 • 8 tweets • 2 min read
listen to this to hear me and mike swear a bunch haha i forgot i said "nothing is robust to all kinds of misspecification, that would be insane"
Jan 26, 2021 • 7 tweets • 2 min read
Just finished teaching and reluctantly forced myself to open up a paper i'm slightly late in reviewing (because, pandemic) -- but got immediately drawn in by the excellent introduction! Now excited to spend my day on this. this is the thing an introduction has to accomplish. i feel that you can really tell when a paper is written for you -- the reader -- versus being written for the author's own joy or whatever. if you want to communicat effectively, you have to think about your interlocutor! "what's my reader thinking now? what's she feeling now?"
Dec 16, 2020 • 13 tweets • 2 min read
statistics is the most beautiful and egalitarian discipline of all of those i have studied, because when we are confused about anything in statistics we are almost always confused about the elementary, foundational concepts in my personal experience this is not true when learning languages, history, economics, or other branches of mathematics
Dec 14, 2020 • 4 tweets • 1 min read
i think we can acknowledge it's unfair that attacks on academic titles / expertise will tend to target women and minorities (& that we naturally wish to defend ourselves) while also not believing that anything on this earth is or can be earned because merit is a broken concept not here for people whose expertise is never questioned (nor time devalued and service assumed available) pontificating about how titles are sooo silly, but definitely also not here for "i earned this"
Dec 11, 2020 • 24 tweets • 4 min read
So some very smart folks have asked about how we would apply the AMIP metric to studies of rare events. This kicked off a discussion of what robustness checks are really for, and I want to take that set of questions seriously in this thread. I think robustness checks mainly (ought to) function to illuminate how variation in the data is being used for inference, and we should then be able to discuss whether we think this is a reasonable situation and adjust our confidence in the results.
Dec 10, 2020 • 4 tweets • 1 min read
Applied econometrics squad assemble!!!! Guys this paper is super important. Arnold, Hull and Dobbie are among the most careful applied econometricians we have, and the explosion of algorithmic decision making means this method -- and their finding of pervasive discrimination -- could hardly be more timely.
Dec 9, 2020 • 5 tweets • 2 min read
"average person eats 3 spiders a year" factoid actualy just statistical error. average person eats 0 spiders per year. Spiders Georg, who lives in cave & eats over 10,000 each day, is the AMIS. we've decided to show quantiles of the spider consumption distribution instead. This was my friend @averyflinders idea, cowriting credit where it is due!!! #amipmemes