In March 2021 an influential study found no statistically significant difference in COVID rates in school districts with 3 vs 6ft of distancing.

In our new reply we question their core result, which changed due to errors uncovered in the original data academic.oup.com/cid/advance-ar…
1/n tl;dr: if you use the corrected data and adjust for known confounds (demog. & community case rates), the case for claiming a null result for 6ft of distancing becomes very thin

This thread explores some of the data and methodological concerns in the original work.

2/n

NEW: Since Mar '20, the state prison population fell by at least 16%—the largest drop in US history

Yet at the same time, the relative number of incarcerated Black people sharply increased

We look at why medrxiv.org/content/10.110… led by @elizabhinton @big_data_kane @svscarpino
🧵 We spent months scraping, FOIA-ing, and manually entering prison population data for every state and DC, for almost the last 10 years.

Our findings add new insights to what scholars & activists are referring to when discussing *structural* racism.

🧵 2/13

Recent work with @balietti and @criedl on arXiv. It's a paper about optimal experimental design, and I quite enjoyed working on it. Join me below for a fun little look at what we discuss in the paper (comments welcomed!)

[arxiv.org/abs/1807.07024]

As the kids say: thread.

1/8 The Background

Models are powerful tools for interpreting / making predictions about human behavior. This paper presents a way to design experiments that make model selection easier and faster (with fewer arguments along the way!).

2/8