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.comDo Pre-Registration and Pre-Analysis Plans Reduce p-Hacking and Publication Bias?Randomized controlled trials (RCTs) are increasingly prominent in economics, with pre-registration and pre-analysis plans (PAPs) promoted as important in ensuri

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)

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)

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)

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

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)

^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.

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.maChanges in Cancer Screening Rates Following a New Cancer Diagnosis in a Primary Care Patient PanelThis cohort study examines whether primary care physicians’ exposures to patients with new diagnoses of breast or colorectal cancer were associated with changes in screening rates for other patients subsequently visiting the affected physicians.

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)

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

^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.

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

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".

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.

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.eduThe Allocation of Food to Food Banks | Journal of Political Economy: Vol 130, No 8Feeding America allocates donated food to over 200 food banks. In 2005, it transitioned from a queueing mechanism to one where food banks use a specialized currency to bid for food. Food banks chose very different food than they received before. Small food banks acquired 72% more pounds per client t…

Paul Novosad @paulnovosad

I like these article highlights — Elsevier showing they can still add value to the publication process. sciencedirect.com/science/articl…