Nathan Morse, PhD

Statistician • Social scientist • Educator

I develop datasets, metrics, and visualizations that drive improvements in the Census Bureau’s statistical methods.

Mathematical Statistician
at the U.S. Census Bureau

Washington, DC

nathanamorse

Data visualization

interactive charts, animated graphs, dashboards, web apps
Longitudinal research methods

time series analysis, survival analysis, panel survey design and analysis
Automating data analysis

package development, prompt engineering, random forests, natural language processing

I have a PhD in quantitative political science from Penn State. My doctoral studies concentrated in political methodology, including advanced statistics, econometrics, and research design. My interactive doctoral dissertation uses rich visualizations, videos, and web apps to explore how constitutional design affects polarization and inequality.

PhD, Political Science

The Pennsylvania State University • 2024
MA, Political Science

The Pennsylvania State University • 2022
BA, Statistics & Political Science

University of Florida • 2019

While in graduate school, I taught Introduction to Political Research at Penn State—an undergraduate course with an overview of research design, basic statistics, and AI writing tools—for which I received the political science department's 2022-2023 Friedman Award for Excellence in Teaching. I then served as the department's Methods Preceptor, an assistant instructor position for graduate-level statistics courses.

PLSC503: Multivariate Analysis for Political Research

Preceptor • Spring 2024
PLSC504: Topics in Political Methodology

Preceptor • Fall 2023
PLSC502: Statistical Methods for Political Research

Preceptor • Fall 2023
PLSC308: Introduction to Political Research

Instructor • Spring 2022

My academic research focuses on how democracy and capitalism interact over time, with particular attention to constitutional design and economic inequality. I primarily present my work using interactive visualizations, web apps, and dynamic articles, and I develop tools to help others do the same.

American Politics in Perspective

Interactive doctoral dissertation • March 2024
Visualizing Social Patterns

Data visualization blog • November 2023 – Present
AI is transforming how we write, and educators should embrace it

Op-ed article • January 2023
tsatools package

R library • May 2022

This app lets users explore economic, social, and political patterns across the 38 member nations of the OECD, the world's developed democracies. The statement at the top of each graph describing the relationship between the two variables is based on a simple linear regression model controlling for the categorical variable selected in the “Color by” input (unless “United States” is selected). The models can only describe correlation, not causation. Chapter 3 of my dissertation reviews literature with causal analysis of some of these patterns. Produced in R with shiny and highcharter.

These animated graphs were posted on TikTok and are built on data from the Economic Policy Institute and the World Inequality Database. The first video displays the tight relationship between union membership and incomes. Studies consistently find that increased participation in unions was a primary driver behind the shift toward a fairer income distribution in the mid-twentieth century and that tighter restrictions on union activity reversed this trend. Since then, workers have been keeping a smaller and smaller share of the value of their labor, as the second video shows. There are many more reasons for these patterns than two graphs can display, so see the accompanying articles for more details and analysis. Produced in R with ggplot2 and gganimate.

Thomas Jefferson argued constitutions should only last 19 years to prevent corruption. I investigated whether this is a meaningful threshold by estimating the effect of a country's constitution outliving 19 years on metrics of political and economic inequality. This coefficient plot shows the results of a model of a democracy index (V-Dem). The sample includes more than 600 national constitutions from the Comparative Constitutions Project. Variables with red bars are associated with poorer democracy, blue bars indicate a positive link with democracy, and gray bars mean no relationship. The results suggest that 19 years is an arbitrary cutoff. I also modeled several alternative propositions of the optimal lifespan for a constitution. Across all models, there is no evidence of any metric that can reliably predict how long a constitution will perform effectively. Data sources and methodological details are discussed in Chapter 5 of my dissertation. Produced in R with plm and highcharter.

Nathan Morse, PhD

Statistician • Social scientist • Educator

Mathematical Statistician at the U.S. Census Bureau

Washington, DC

nathanamorse

nathanamorse

Expertise

Education

Teaching

Research

nathan.a.morse@gmail.com

Mathematical Statistician
at the U.S. Census Bureau