Regression in Social Science

Build and interpret linear, categorical-outcome, fixed-effects, and multilevel regression models.

Further development planned

Course 2 · 12 modules · About two hours of study per module

Before you start

Prerequisites: Basic statistics and working knowledge of R; equivalent foundation skills are acceptable.

Module sequence

Choose a module below. Planned modules contain curriculum information only; their lessons are not yet available.

Module Focus Learning goal Availability
1 Foundational Knowledge Connect research questions, variables, statistical summaries, and uncertainty. Planned
2 Linear Regression Interpret a fitted slope and prediction in their original units. Pilot
3 Categorical Predictors and Interactions Interpret reference categories and relationships conditional on another variable. Planned
4 Model Specification and Diagnostics Assess functional form, residual patterns, and influential observations. Planned
5 Logit and Probit Regression Relate binary outcomes to logit/probit models and predicted probabilities. Planned
6 Predicted Probabilities and Marginal Effects Translate nonlinear model results into probabilities and marginal effects. Planned
7 Ordinal and Multinomial Logit Models Distinguish ordered and unordered categorical outcomes and model assumptions. Planned
8 Conditional Logit and Choice Data Represent choice sets and interpret alternative-specific predictors. Planned
9 Fixed-effects Models Explain within-unit comparisons and unit/time fixed effects in panel data. Planned
10 Multilevel Models Distinguish within-group and between-group variation in hierarchical data. Planned
11 Regression Presentation Present regression estimates and predictions with clear substantive interpretation. Planned
12 Regression Project Choose and assess a regression model for a bounded research question. Planned