Replication materials for Automated Reminders Reduce Incarceration, 2025

This repository contains replication materials for our paper, Automated Reminders Reduce Incarceration for Missed Court Dates: Evidence from a Text Message Experiment (Science Advances, 2025).

If you are not viewing this README on GitHub, a more recently updated version may be available at our online paper repository.

Users can set up the correct R environment by running renv::restore() in the root project directory. This will install the packages necessary to run the scripts in this repository.

Data access

Before you begin, download the replication data from the Harvard Dataverse and place the CSVs in the data/ directory.

You have the option of either a) immediately downloading a public aggregate version of the data from our experiment or b) requesting access to the full set of de-identified restricted data required to replicate all analyses in the paper. Instructions for requesting access to the restricted data are described on Dataverse.

Running the replication on public data

To replicate core findings using public aggregate data, download the public datasets from the Dataverse, place them in the data/ folder, and run src/public_data_analysis/evaluation.R.

Running the replication on restricted data

To replicate all statistics and figures in the paper, you must request access to the restricted data from the Dataverse. After your access is approved, download the restricted datasets and place them in the data/ folder. Once the environment is restored and the restricted data are downloaded and placed in data/, run:

source("make_restricted.R")

This will render a LaTeX file output/stats.tex from the R Markdown analysis in src/calculate_stats.Rmd. The LaTeX file contains all statistics reported in the paper.

Repository contents

The main scripts are:

• make_restricted.R – Script to calculate all paper statistics and export figures and LaTeX macros to output/ using restricted data.
• src/public_data_analysis/evaluation.R – Loads public aggregate data and fits regression models for replication of main findings.
• src/restricted_data_analysis/calculate_stats.Rmd – Primary R Markdown file generating descriptive stats, regression results, and LaTeX macros.
• src/restricted_data_analysis/evaluation.R – Loads data, fits regression models, and defines model objects used in the analysis.
• src/restricted_data_analysis/utils.R – Helper functions for factorization, summary statistics, formatting output, and LaTeX-friendly presentation.