Course Overview

Data Analysis for Public Policy

What this course is about

Public policy increasingly depends on quantitative evidence. This course introduces students to the statistical reasoning and data-analysis tools needed to evaluate that evidence and produce careful analyses of their own.

We begin with substantive questions rather than disconnected formulas. Each method is introduced through a real public-policy problem, developed conceptually, implemented in R, and interpreted in clear language.

No prior experience with R is required. Students may enter with different levels of mathematical, statistical, and coding preparation.

Learning objectives

By the end of the semester, students should be able to:

  1. Formulate clear and answerable empirical questions.
  2. Identify appropriate data for a public-policy question.
  3. Import, clean, recode, document, and explore data in R.
  4. Summarize distributions numerically and visually.
  5. Explain probability, sampling variability, and statistical uncertainty.
  6. Construct and interpret confidence intervals.
  7. Conduct and interpret basic hypothesis tests and group comparisons.
  8. Analyze correlation and estimate simple and multiple regression models.
  9. Interpret coefficients, predicted values, and uncertainty in context.
  10. Distinguish statistical significance from substantive importance.
  11. Communicate quantitative findings to nontechnical audiences.
  12. Produce a reproducible analysis using Quarto.
  13. Evaluate the ethical and responsible use of data and AI.

How the course works

The course connects three elements throughout the semester:

  • Statistical reasoning: What does the method do, and when is it appropriate?
  • R programming: How can we implement the analysis and check the output?
  • Public-policy application: What does the result mean for the substantive question?

Coding is an important tool, but coding complexity is secondary to reasoning and interpretation.

Weekly structure

Monday: Lecture

We introduce a substantive question and the statistical concepts needed to answer it.

Wednesday: Guided lab

We apply the method together in R using real data. Most coding is completed collaboratively during class.

Friday: Lab section

Sarah Traore reviews difficult material, helps students troubleshoot code, and provides support for the lab’s On Your Own section.

Labs

Labs are cumulative and designed to build a practical data-analysis workflow. Each lab will have:

  • A detailed online teaching guide with explanations, examples, and guided code
  • A minimal downloadable Quarto template
  • Setup code and a few pre-entered examples
  • Blank R cells and spaces for written interpretation
  • A clearly labeled On Your Own section

Students begin each lab in class and complete a shorter independent section afterward.

Applied project

Students will complete an original public-policy analysis using real data. The project requires students to formulate a question, select and document appropriate data, conduct an independent analysis in R, interpret the results, and communicate the findings clearly.

Each student will submit an individual reproducible Quarto report and create a concise policy poster. The semester concludes with a poster session in which students explain their question, evidence, principal finding, and limitations to classmates and invited guests.

The project is cumulative. Intermediate checkpoints will provide opportunities for feedback before the poster session and final submission.