Dashboards
Packages
An R package aimed at making specification curve analysis easy, fast, and pretty. It improves upon existing offerings with additional features and tidyverse integration. Users can easily visualize and evaluate how their models behave under different specifications with a high degree of customization.
Features include: support for generalized linear models, parallelization, fixed effects, standard error comparison and more to come.
Available for install via CRAN or check it out on GitHub.
See here for an in-depth write up on the motivation behind and process of developing the package.
Replications
This paper uses historical delegate vote data from the Constitutional Convention to estimate the effect of delegates' family composition on their voting behavior. Originally done in STATA, I replicate the analyses in R with additional diagnostics and counterfactuals considered.
In this paper, the authors model the Rehnquist-era Supreme Court as a system of magnetic dipoles in an attempt to understand when unanimity emerges on the Court. I partially replicate the analyses using Python.
Replication materials here.