Software

Open-source tools and R packages for quantitative social science built to make experimental design, data visualization, and archival data more accessible.

Interactive Dashboards

Shiny App 29 countries · 1980–1995 · 76,000+ events

European Protest & Coercion Dashboard

Browse, visualize, and model Ron Francisco's European Protest & Coercion Data — the most comprehensive daily-resolution record of political contention and state repression in modern Europe. Four tabs: summary statistics, annual trends, a searchable event table, and interactive regression estimation.

Data: ronfran.ku.edu · Cite as: Francisco, R. (2000). European Protest and Coercion Data. University of Kansas.

Web Dashboard 497 leaders · 123 countries · 1946–present

Personalism in Dictatorships Dashboard

Explore a cross-national dataset measuring personalism in dictatorships via 21 binary indicators and bifactor 2PL IRT estimation. Six tabs: leader rankings, time trends, regional comparisons, indicator details, model diagnostics, and a full codebook.

With Lee Morgenbesser · Manuscript in progress

R Packages

All packages are available on GitHub. CRAN packages include live download counts.

plotrr CRAN Active

Functions for visual exploratory data analysis with nested and hierarchical data. Coauthored with Michael J. Nelson.

colorr CRAN Active

Color palettes for EPL, MLB, NBA, NHL, and NFL teams.

nomine CRAN Active

Classifies names by probable race/ethnicity and gender using the NamePrism and NamSor APIs. Useful for large-scale audit studies and correspondence experiments where respondent identity needs to be inferred from name data.

repllm GitHub Active

A research methodology toolkit for social scientists using LLMs to annotate and classify text. Built on top of ellmer, it adds the methodological layer needed for peer-reviewed research: structured coding runs, reliability assessment (Cohen's κ, Krippendorff's α), sensitivity analysis across prompts/models/temperatures, gold-standard validation, prompt pre-registration, and automated reporting.

R CMD check
validatednamesr GitHub Active

A database of 600 names with 44,000+ evaluations across race, income, education, and citizenship — designed for experimental studies on race and ethnicity. Coauthored with Jae Yeon Kim, S. Michael Gaddis, John B. Holbein, Connor Guage, and William W. Marx.

Crabtree et al. (2023). Scientific Data, 10, 130.

Using one of these tools in your research? I'd love to hear about it — bug reports, feature requests, and coauthorship inquiries welcome.

Get in touch →