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
Oleg Urminsky Profile picture
Oleg Urminsky
@OlegUrminsky
Prof @ UChicago Booth. Trying to learn how people think. Not a statistician but not exactly a psychologist either. You kids keep me young.
Sep 16, 2023 • 7 tweets • 2 min read
I'm seeing lots of discussion about this post that kinds of sounds like debating whether to burn everything down before tenure or wait until after tenure. I'd suggest another way to think about it... 🧵 Have, live by and express the same values and principles before and after tenure. Before tenure, you are asked to play a different role in the field and in your institution than after. The tradeoff is being shielded from some difficult decisions but having less info and input.
Apr 13, 2022 • 4 tweets • 1 min read
One aspect of academia it took me a long time to appreciate: self-selection means that you're mostly presenting research to peers who care about the research question less than you do. The people who are most interested in Question A study question A. Then they present their findings to (and are evaluated by) people who didn't choose to study Question A. I think this explains a lot of the "I love research but hate peer-review" feeling in academica.
Dec 22, 2021 • 15 tweets • 3 min read
This thread yesterday got some really thoughtful replies and discussion, that I'd like to highlight: how social science could have been better prepared with relevant answers, what we did have to offer and the questions we were able to answer about the pandemic. 1/n There seemed to be more agreement than I expected on my main point. Many of the failings were “behavioral” & psychology can identify issues to consider. But we don’t have clear answers that were ready to be implemented or models that made non-ambiguous and correct predictions.
Dec 21, 2021 • 23 tweets • 6 min read
1/n Seeing exasperated "You figured this out NOW?" retweeting by social scientists. And, yes, part of me is also frustrated that some policy makers seemingly thought they could ignore the human element.

But maybe we social scientists have some hard lessons to learn too... Let's take a look in the mirror: if Collins had come to us and said "Tell me what you know," what could we actually have said, with a solid scientific basis?
Honest question. I'm curious if anyone can point to consensus conclusions that translate into actionable policy advice.
Sep 24, 2021 • 14 tweets • 3 min read
Meta-analysis is a terrifically useful tool, but these are important questions: what's the point of meta-analysis if a literature is biased? A brief thread. What meta-analysis does is provide summary estimates from potentially conflicting data. Just like a mean and standard deviation provide useful summaries of individual people's data, meta-analysis does the same thing across studies.
Sep 3, 2021 • 10 tweets • 2 min read
A case study in how non-scientific conclusions are spread in the mainstream media.

Title: "The Downsides of Masking Young Students Are Real. The educational cost of face coverings is far better established than the benefits of mandates."

What does the article actually say? Article: "Many studies support the importance of babies seeing caregivers’ faces, and prior to the arrival of COVID-19, many American professional organizations, including the AAP, strongly agreed."
Sep 1, 2021 • 6 tweets • 2 min read
Great to see this. But it's embarrassing that this even needed to be done. Once we knew it was airborne, it was obvious, from everything we know about *other* airborne diseases based on 100s of years of medical advances, that masks would almost certainly be very beneficial.🧵 Nevertheless, really prominent researchers (including some affiliated with my institution) took the position (even a few weeks ago!) that because there was no RCT proof, it was irresponsible or even harmful to require masks.
Aug 29, 2021 • 25 tweets • 5 min read
A twitter kerfluffle erupted yesterday over some claims in a hyperbolic essay about the 🙄“death of behavioral economics”.

For the big picture, I agree with this thread by my colleague @alexoimas:


But what about the specific claim made? Long🧵 The claim:
“findings rebutting their view of loss aversion were carefully omitted...and other findings that went against their model were misrepresented so that they would instead support their pet theory...any data that didn't fit Prospect Theory was dismissed or distorted.”
Jun 6, 2020 • 12 tweets • 3 min read
Hello @MZHemingway! I hear you're interested in the wisdom of Czechoslovakian dissidents. My mother was nearly shot by Soviet snipers during the 1968 invasion & I was teargassed as a baby on the 1 year anniversary. I also think Havel is relevant to current events! So, that Havel story you quoted, from The Power of the Powerless:
en.wikipedia.org/wiki/The_Power…
He's talking about how displaying the slogans of the regime because of the fear of state repression, perpetuates the power of the state and one's own humiliation by the state.
Jan 4, 2020 • 7 tweets • 3 min read
@Moshe_Hoffman I agree that it's limited, the question is why. I think it boils down to how much irreducible complexity you think there is in human decision-making. @Moshe_Hoffman It could be that there's little -- taking the right perspective will yield a simple and accurate model of decision making. If so, then we have seemingly failed to find the answer thus far, but there is an answer out there to find.
Jan 21, 2019 • 10 tweets • 3 min read
This article is crucial to understanding our political situation:
buzzfeednews.com/article/hnsgra…

I actually worked a bit on polling for the Peres campaign mentioned in the article. It was a heartbreaking loss, but we had no idea what it would lead to. /thread In politics (and in political consulting) there is a fundamental conflict between the mercenaries and the idealists. People sometimes start out as idealists and turn into mercenaries. In the US, on the Democratic side there are plenty of both.