Conflict, Climate Change, and Childhood Malnutrition

As part of a team of scientists working on a DFID and Action Against Hunger project, I am working on modeling childhood malnutrition in the context of armed conflict and climate change. I am leading the design and execution of a subnational analysis, which combines geocoded data on violence and satellite imagery of climate conditions with machine learning algorithms to predict malnutrition within African countries.

Data Visualization

In addition to my work on the political economy of development, I also design visualizations that communicate patterns in complicated data to the public. At this link, you’ll find a few of my recent contributions, inlcuding an article on the makeup of the new US Congress featured in USA Today.

IDP Camps, Land, and Vote Buying

I use a randomized response experiment to show that returned IDPs in Northern Uganda are often targeted with vote buying.

The Demand for Aid

I combine an original conjoint experiment with a formal model to understand citizen demand for different types of development in Uganda.


Skills and experience

Data science:

  • Statistical programming:
    • R (advanced)
    • Stata (advanced)
    • Python (familiar)
  • Methods:
    • Econometrics (regression, maximum likelihood, panel data)
    • Bayes (hierarchical models, stan, brms)
    • Machine learning (various methods for regularization and feature selection)
  • Data visualization:
    • ggplot2 (advanced)
    • Shiny (intermediate)
    • d3.js (familiar)

Social science:

  • Causal inference and research design
  • Conjoint experiments
  • Experiments for sensitive topics
  • Survey design and implementation
  • Field work (Uganda, Ghana, Burkina Faso)


  • As teaching assistant:
    • Politics of the Developing World
    • Religion, Beliefs, and World Affiars
    • International Negotiations
  • As a workshop instructor:
    • Causal Inference using Conjoint Analysis
    • Introduction to Spatial Data and GIS
    • Document Processing with LaTeX and knitr