Fall 2026

PUBH 6310 — Statistical Analysis in Health Policy

Follow the suggested learning path for each session, then open any module for guided exploration and practice.

1. Before Aug. 27

   Session 0

   Pre-Course Setup — Building Your AI-Data Environment

   Asynchronous technical-readiness tutorial completed before Session 1.

   Course recordings

   • Brief Intro ↗
   • Stata Installation ↗
   • VSCode, Stata MCP, Codex Installation ↗
2. Aug. 27

   Session 1

   Building Your AI-Data Command Center

   Suggested learning path

   1. Mapping the Research Process
   2. AI Is an Assistant, Not an Investigator
   3. Context Is Part of the Analysis
   4. The Research Plan Is the Analytical Contract
   5. The AI Audit Loop
   6. Build Your Research Specification with AI
   7. Research Integrity in AI-Assisted Analysis Optional

   Course recordings

   • Class Recording ↗
3. Sept. 3

   Session 2

   AI-Assisted Data Management

   Suggested learning path

   1. Unit of Analysis
   2. Unit of Treatment
   3. How a Dataset Represents the Population
   4. Combining Datasets for Analysis
   5. One-to-One Merge
   6. One-to-Many Merge
   7. joinby and Pair Generation
   8. Append Data Files
   9. Recoding Variables
   10. Generating Variables from Multiple Fields
   11. Missing Data Decisions

   Course recordings

   • Class Recording ↗
4. Sept. 10

   Session 3

   Levels of Measurement, Hypothesis Testing, and Univariate Analysis

   Suggested learning path

   1. Measurement and Representation
   2. Sampling Variability
   3. Confidence Intervals
   4. Survey Weights and Clustering
   5. Hypothesis Testing as Evidence

   Course recordings

   • Class Recording ↗
   • Homework 1 Tuitorial ↗
5. Sept. 17

   Session 4

   Bivariate Analysis of Categorical Data

   Suggested learning path

   1. Categorical Denominators
   2. Chi-Square as Evidence of Association
   3. Fisher's Exact Test Optional
   4. Effect Magnitude Before Significance
   5. Choosing a Categorical Effect Measure
   6. Statistical Significance vs Policy Importance

   Course recordings

   • Class Recording ↗
6. Sept. 24

   Session 5

   Causal Reasoning, Confounding & Research Design & Review of HW #1

   Suggested learning path

   1. Association vs Causation
   2. Variable Roles
   3. DAG Reasoning for Adjustment
   4. Confounding and Adjustment
   5. From DAG to Analysis Plan

   Course recordings

   • Class Recording ↗
7. Oct. 1

   Session 6

   Multivariate Analysis of a Dichotomous Outcome

   Suggested learning path

   1. Probability, Odds, and Log-Odds
   2. The Logistic Curve
   3. Introducing Multivariable Logistic Regression
   4. An Odds Ratio Is Not a Risk Ratio
   5. Predicted Probabilities
   6. Same-Sample Adjustment
   7. Logistic Model Assessment and Selection

   Course recordings

   • Class Recording ↗
8. Oct. 8

   Session 7

   Bivariate Analysis with a Continuous Outcome

   Suggested learning path

   1. Reading a Distribution
   2. Shape Before Correlation
   3. Pearson and Spearman Correlation
   4. Mean, Median, and Extreme Values
   5. Independent or Paired?
   6. Choosing a Two-Group Comparison
   7. Comparing Two Means
   8. Comparing Two Medians Optional
   9. Comparing Groups with Ranks
   10. ANOVA as a Comparison of Variation

   Course recordings

   • Class Recording ↗
9. Oct. 15

   Session 8

   Integration, Homework #2 Review & Preparing for Linear Regression

   Suggested learning path

   1. Choose an Analysis Starting Point
   2. Choosing the Method Is an Investigator Decision
   3. The AI Audit Loop
   4. Match the Question to the Method
   5. The Two-Group t-Test as Regression
   6. Correlation and the Standardized Slope
   7. Linear Regression as a Model
10. Oct. 22

    Session 9

    Multivariate Analysis of a Continuous Outcome

    Suggested learning path

    1. Linear Regression as a Model
    2. Holding Other Variables Constant
    3. Residuals
    4. Linearity of the Conditional Mean
    5. Heteroskedasticity
    6. Influential Observations
    7. Collinearity
11. Oct. 29

    Session 10

    Model Refinement, Robustness & Effect Heterogeneity

    Suggested learning path

    1. Functional Form
    2. Raw and Log Outcome Scales
    3. Sensitivity Analysis
    4. Specification Searching
    5. Effect Heterogeneity and Interaction
    6. Investigator Defense
12. Nov. 5

    Session 11

    Reproducibility Audit, Practical Exam Studio & Final Project Clinic

13. Nov. 11–13

    Session 12

    Individual Practical Exam

    Remote individual exam window; no regular class meeting on Nov. 12.

14. Nov. 19

    Session 13

    Feedback on Preliminary Findings & Volunteer Presentations

15. Dec. 3

    Session 14

    Final Project Research Conference & Course Synthesis