Best of #econtwitter - Week of August 7, 2022 [3/3]
Welcome readers old and new to this week’s edition of Best of Econtwitter. Please submit suggestions — very much including your own work! — over email or on Twitter @just_economics.
This is part three of three.
Paper summaries
Jon Hartley@Jon_Hartley_
1: Excited to share my new paper with Abel Brodeur, Nikolai Cook (@nikolaimcook) and Anthony Heyes: “Do Pre-Analysis Plans and Pre-Registration Reduce p-Hacking and Publication Bias?” A THREAD (1/N)
papers.ssrn.com
Do Pre-Registration and Pre-Analysis Plans Reduce p-Hacking and Publication Bias?
4:45 PM · Aug 4, 2022
64 Reposts · 161 Likes
Jon Hartley@Jon_Hartley_
4: We find no meaningful difference in the distribution of test statistics of pre-registered RCTs compared to the distribution of test statistics of non-pre-registered RCTs (4/N)
4:45 PM · Aug 4, 2022
7 Reposts · 25 Likes
Jon Hartley@Jon_Hartley_
5: However, we do find that pre-registered RCT studies with a complete pre-analysis plan (PAP) are significantly less p-hacked compared to pre-registered RCT studies without a PAP (5/N)
4:45 PM · Aug 4, 2022
11 Reposts · 46 Likes
Augustine Denteh@austin_denteh
What happens when the treatment variable is misclassified in DID designs? This could occur when exact treatment timing is unknown, or treatment groups are estimated. In our new WP, we show that the DID estimand does not have a clear causal interpretation (1/3)
3:45 PM · Aug 4, 2022
21 Reposts · 97 Likes
Alex Hollingsworth@ajhollingsworth
Traditional power analyses rely on formulas and are only available for simple settings. Many modern/complex research designs don't have a corresponding power formula. How can you estimate power without a formula? Simulation! A 🧵 re: our recent paper
Alex Hollingsworth @ajhollingsworth
If you're interested in simulated power analyses. You can check out a paper with @leticiafcnunes @KosaliSimon and Bernie Black here
https://t.co/m8rt6j3LTp
1:39 PM · Aug 1, 2022
26 Reposts · 85 Likes
Brandyn Churchill@bf_churchill
Grateful for all the helpful feedback that we (@LEHenkhaus, @EmilyCLawler) received on our paper at #NBERSI ! A big thanks to our great discussant, @btshapir, and all the participants (1/9)
9:00 PM · Jul 29, 2022
3 Reposts · 30 Likes
^not a covid vaccine paper; here’s a covid vaccine paper
Alexander Berger@albrgr
Interesting paper using UMETRICS data from a bunch of universities to look at what happens when a researcher receives an NIH grant: science.org/doi/10.1126/sc…
They hire more people, who produce more research. But magnitudes debatable; authors argue not a mechanical hiring effect.
10:41 PM · Aug 3, 2022
5 Likes
Jessica Cohen@jessicaleecohen
What do u you think influences your decision to get screened for cancer? Family history? Age? Nerves? Probably not top of mind: Cancer rates of your primary care doc’s *other* patients.
Short🧵on our paper in @JAMANetworkOpen @AnnabelZWang @ml_barnett
ja.ma/3zsMHih
ja.ma
Changes in Cancer Screening Rates Following a New Cancer Diagnosis in a Primary Care Patient Panel
11:40 PM · Aug 3, 2022
13 Reposts · 64 Likes
Kevin Williams@mrequalsmc
🚨New Paper Alert🚨 What are the welfare effects of dynamic pricing in oligopoly markets? We offer new theoretical insights and empirical estimates for the airline industry. Spoiler alert: Little intuition from single-firm settings carry over; the welfare effects flip sign. (1/n)
8:24 PM · Aug 2, 2022
71 Reposts · 349 Likes
More: India housing; job postings; allocation of immigrant talent; sovereign default; Geography of Human Capital Management; rent regulation; sports club vouchers null
Interesting discussions
Adam Ozimek@ModeledBehavior
If you are an active undergraduate on Econ twitter and become friends with a whole lot of leading economists at great phd schools I feel like that has to increase your odds of getting into those programs even if those professors insist it doesn’t
2:05 PM · Aug 2, 2022
2 Reposts · 203 Likes
^discuss. Worked for n=1:
John Ruf@JohnRuf6
@ModeledBehavior I made an engineering substack post one time about the mechanics of caterpillar swarms and @AnthonyLeeZhang liked it and got me an initial interview for my current job as a predoc.
Now here I am waiting for my code to compile while eating lunch.
5:42 PM · Aug 2, 2022
1 Repost · 12 Likes
quan le 🍚🥢@qlquanle
pining for an econ textbook in the tradition of math texts like "Counterexamples in Topology" where we just go through nice simple true-sounding statements in econ and cleverly tearing them apart
4:35 AM · Aug 8, 2022
5 Reposts · 110 Likes
Prof. Paul Musgrave, Ph.D.@profmusgrave
your perpetual reminder that the "9-month-contract" is a useful fiction that allows academics to consult and do other work while being employed full time. you would not earn 3 extra months' salary if you switched to a "12-month contract".
7:09 PM · Aug 7, 2022
18 Reposts · 300 Likes
Andrea Matranga 🇺🇦🌻@andreamatranga
@leightjessica @annalilharvey @mosenkis @tom_wein @caitlin_tulloch @3ieNews @NSF Something I keep fantasizing about is like regional RCT conferences where local admin and researchers can meet to design El Cheapo RCTs.
3:11 PM · Aug 3, 2022
1 Like
Nick Arnosti@NickArnosti
This is a great example of market design: journals.uchicago.edu/doi/abs/10.108…
It is also a great example of the failed econ publishing process. The "new" mechanism started in 2005. Data from 2005-2011. I saw Canice's talk in 2013. What good does it do for the JPE to "publish" it in 2022?!?
journals.uchicago.edu
The Allocation of Food to Food Banks | Journal of Political Economy: Vol 130, No 8
1:17 PM · Aug 2, 2022
35 Reposts · 214 Likes
Paul Novosad@paulnovosad
I like these article highlights — Elsevier showing they can still add value to the publication process.
sciencedirect.com/science/articl…
8:33 PM · Aug 3, 2022
102 Reposts · 847 Likes