I think this paper sets a great standard: controlled survey or lab experiment is not enough, we need to connect effects to population data to understand implications. Here, the population patterns are so stark that they overturn the naive inference from the experiment.

I am now preparing some DiD lecture slides to give a talk at Amazon, and I have created this DiD Checklist. I am sharing them here in case some of you find it useful!

Too many good conferences happening in October folks. There are other months of the year! This is creating too much FOMO!

Best piece I’ve read on NYC’s congestion pricing debacle: nytimes.com/2024/06/07/opi…

📣 We're excited to announce our 2024 Membership Round is open! We’re looking to expand our network w/ new individual researchers & institutional members. Deadline: Sept 30, 2024 For more details & to apply, see here: forms.gle/eFfNQFQzZR9UMX… Please RT & share w/ networks!

Loss of an intellectual giant. His work was an inspiration to me as a grad student and remains ever relevant.

Interested in becoming an EGAP member? We’re expanding our network! Learn more and apply today: forms.gle/eFfNQFQzZR9UMX… Deadline: Sept 30, 2024

There is not enough constancy in the social world for "cumulating understanding of recurrent macro phenomena" to be a reasonable goal for social scientists. A better goal is providing "micro navigation tools" for selecting good actions in ever-evolving macro circumstances.

Interesting analysis: “sample sizes typical of many real-world clinical trials datasets often constrain regression adjustment methodologies from attaining their theoretical potential in practice.” Proposes methods to “borrow observations from large-scale observational datasets.”

NYU's Politics Department is seeking outstanding and innovative scholars for two positions at the assistant professor level: International Relations and US Politics. For full descriptions and to apply visit apply.interfolio.com/152515 (IR) and apply.interfolio.com/152513 (US).

Junior faculty recruitment in IR and American Politics at NYU:

Fortunately there is a pre-analysis plan …

The seamless interface between estimatr and modelsummary in R is such an efficiency boost, not to mention the brilliant performance of each respective package. Vincent Arel-Bundock and estimatr team (no longer here?) bravo