General Sessions - 2026 ICPSR Summer Program Registration

Format/Modality: All General Session classes will be offered in a dual-mode/hybrid format, and participants can choose to participate either in-person or online via Zoom livestream and/or asynchronous lecture recordings.

Class Dates & Times: Math and computing Lectures will be held during Week 1 of each General Session. Methods Courses will be held during Weeks 2-4 of each Session. All classes meet Monday through Friday, unless stated otherwise. All times are Eastern Time (ET).

Class Selection Guides: Unsure which Courses and Lectures to take? Consult our guides for selecting classes in the First Session and the Second Session, and watch our class selection video.

Registration Deadlines: Registration for the First Session will close on Friday, May 29, 2026. Registration for the Second Session will close on Friday, June 26, 2026.

Dates: June 8-12, 2026

Time: 8:15 a.m.-12:15 p.m.

Instructor: Anne Cizmar

Goals:

To define core mathematical concepts through theory and examples.
To apply mathematical relationships to statistical terms and theories.
To analyze and review solutions to mathematical problems within statistical context.

Learning Outcomes:

Describe introductory/intermediate mathematical concepts.
Solve introductory/intermediate mathematical problems.
Synthesize mathematics and statistics.
Develop skills for advanced concepts.

Prerequisites: This introductory-level course has no prerequisites. The course assumes no prior knowledge in statistics. The course is open to participants from every discipline and level. No software is required for the course, although some examples will be shown in R and Stata and the codes will be shared with the participants.

Difficulty Level: Beginner

Software: None

Dates: June 8-12, 2026

Time: 8:15 a.m.-12:15 p.m.

Instructor: Sara Tomek

Goals:

To review and learn statistical and mathematics concepts prior to seeing them in the formal courses.
Increase your statistical and mathematical vocabulary in order for a deeper understanding of the statistics courses taken during the first session.

Learning Outcomes:

Participants will become familiar with concepts of Matrix Algebra, Calculus, and Categorical and Continuous Probability Distributions and how this math is used throughout statistics.

Prerequisites: Participants should have some mathematics knowledge and experience. It is an intense week but will provide students with greater confidence to learn more in their more formal courses.

Difficulty Level: Intermediate

Software: None

Dates: June 8-12, 2026

Time: 1:30-5:30 p.m.

Instructor: Michael Hawthorne

Goals:

Introduce participants to core computing concepts and solutions applicable to analysis packages, particularly directed to participants having limited experience.
Introduce participants to the dataset development process, from beginning to end, examining real-world issues in creating and maintaining datasets through examples across disciplines.

Learning Outcomes:

Understand core procedures in three statistical software packages
How to implement the dataset development process
Understand connections between concepts in developing datasets and practical issues.

Prerequisites: There are no prerequisites required for the course. The course will be particularly beneficial for participants who have had limited experiences with creating their own datasets and/or using statistical software, or have been dependent upon having others assist in creating datasets and software commands for analysis, or are more advanced users now confronting new issues with datasets they wish to develop.

Difficulty Level: Beginner/Intermediate

Software: SAS, SPSS, STATA/MP

Dates: June 8-12, 2026

Time: 1:30-5:30 p.m.

Instructor: Kerem Ozan Kalkan

Goals:

Introduce participants to the R statistical environment through applied examples.
Familiarize participants with the fundamentals related to coding in R.
Cover introductory and intermediate tasks that social scientists frequently use in R.

Learning Outcomes:

Recognize introductory/intermediate coding in R.
Code in R to manipulate data, run introductory/intermediate statistical routines, and generate graphics.
Substantively interpret R output.

Prerequisites: There are no prerequisites for this course. The course assumes no prior knowledge in R.

Difficulty Level: Beginner/Intermediate

Software: R, Rstudio/Posit (Cloud)

Dates: June 8-12, 2026

Time: 5:45-6:30 p.m.

Instructor: Kerem Ozan Kalkan

Goals:

Introduce participants to the LaTeX markup language through applied examples.
Cover introductory to advanced environments, functions, and practices commonly observed among quantitative social scientists.
Prepare participants to use LaTeX in their assignments and projects.

Learning Outcomes:

Identify introductory to advanced environments and functions in LaTeX.
Practice coding in LaTeX through applied examples.
Typeset their own documents in LaTeX

Prerequisites: There are no prerequisites for this course. The course assumes no prior knowledge in LaTeX.

Difficulty Level: Beginner/Intermediate

Software: R, Rstudio/Posit (Cloud), LaTeX, TeXShop

Dates: June 15-July 2, 2026

Time: 9:00 a.m.-12:00 noon

Instructors: Jake Bowers & Thomas Leavitt

Goals:

Understand Fisher’s and Neyman’s counterfactual models of random assignment studies, the utility of these models in clarifying causal inferences’ scope and prior commitments, and their contribution to the rise of the randomized controlled trial as a model research design
Practice the use of R to estimate or test hypotheses about treatment effects in experiments and non-confounded observational studies
Understand and practice the combination of matching with hidden variable sensitivity analysis to address both overt and hidden confounders in non-randomized studies.

Learning Outcomes:

Articulate, differentiate and critique essential assumptions of causal estimation and hypothesis testing that flow from research design.
Apply forms of randomization inference appropriate to randomized trials featuring clustered or stratified treatment allocation and/or imperfect compliance. Appropriately adapt the same inference strategies to observational studies, and to diagnostics including covariate imbalance.
Implement, critique and defend statistical adjustments for non-randomized studies using the linear model (i.e. “controlling for”) and/or post-stratification (i.e. “matching”).
Implement testing, estimation, and adjustment methods covered in the course using R, in replicable scripts combining R and markdown.

Prerequisites: The course assumes basic familiarity with R and previous knowledge of statistical concepts, such as probability distributions, statistical inference, estimation, and hypothesis testing. The course will rely on R for computation and to demonstrate concepts.

Difficulty Level: Beginner/Intermediate

Software: R

Dates: June 15-July 2, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Clay Webb

Goals:

Introduce statistical models for time series and panel data
Offer a comprehensive guide to estimation and inference using time series and panel data models
Provide practical guidance on how to conduct time series and panel data analyses using R and Stata

Learning Outcomes:

How to measure dynamic phenomena over time
How to build statistical models of dynamic phenomena
How to use dynamic models to test hypotheses about dynamic phenomena

Prerequisites: Regression 1; A basic understanding of ordinary least squares (OLS) regression

Difficulty Level: Intermediate

Software: R, Rstudio/Posit (Cloud), STATA/MP

Dates: June 15-July 2, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Chris Hare

Goals:

Introducing participants to the mechanics of machine learning methods, particularly focusing on supervised learning techniques.
Demonstrating how these techniques can be applied to social science research to enhance model prediction accuracy.
Providing hands-on experience in estimating and interpreting machine learning models using R.

Learning Outcomes:

The theory underlying of machine learning
How to implement various machine learning techniques (focusing on tree-based methods) in R
Strategies to interpret results from "black box" methods.

Prerequisites: The workshop assumes a basic understanding of statistical concepts and the R programming language. While not strictly necessary, prior experience with regression analysis and data manipulation in R would be beneficial.

Difficulty Level: Intermediate

Software: R

Dates: June 15-July 2, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Bob Lupton

Goals:

Introducing the method of maximum likelihood for statistical modeling
Understanding the usefulness of generalized linear models for statistical analysis
Learning how to specify, and interpret the results of, limited dependent variable models

Learning Outcomes:

Benefits and limitations of ML for statistical inference
Specify, interpret, apply and communicate the the results of theoretically informed models
Data visualization techniques in R and Stata

Prerequisites: Participants should be familiar with the linear regression model. Some prior exposure to Stata or R is useful but neither is assumed.

Difficulty Level: Intermediate

Software: R, STATA/MP

Dates: June 15-July 2, 2026

Time: 9:00 a.m.-12:00 noon

Instructors: Kerem Ozan Kalkan & Naomi Nubin-Sellers

Goals:

Provide an overview of theories and methodological approaches to the study of race and ethnicity in policy and public opinion.
Introduce participants to methodological tools that are commonly observed in the study of race and ethnicity.
Develop strategies toward developing and executing original research ideas related to race and ethnicity.

Learning Outcomes:

Theoretical debates in the study of race/ethnicity.
Methodological approaches in the study of race/ethnicity.
Replication/extension.
Develop a comprehensive research approach.

Prerequisites: None

Difficulty Level: Beginner/Intermediate

Software: R, STATA/MP

Dates: June 15-July 2, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Robert Cooper

Goals:

Empower participants to grow comfortable with and master introduction and intermediate regression analysis.

Learning Outcomes:

Learn how to perform linear regression and test assumptions through regression diagnostics, correcting for violations of assumptions
Participants will also be introduced to MLE.

Prerequisites: There are no prerequisites, although some basic statistical background would be helpful.

Difficulty Level: Beginner/Intermediate

Software: R

Dates: June 15-July 2, 2026

Time: 2:30-5:30 p.m.

Instructor: Justin Esarey

Goals:

How to apply Bayesian models to the study of social scientific questions and interpret the results
Practical Bayesian implementations of the (hierarchical) generalized linear model
Learn to use R for programming, data management, and visualization with RStudio as an IDE.

Learning Outcomes:

A basic understanding of Bayesian inference
How to implement Bayesian regression and hierarchical models in R/STAN
How to interpret and troubleshoot Bayesian model outcomes"

Prerequisites: A course on basic probability theory and linear regression / the generalized linear model would be helpful.

Difficulty Level: Intermediate/Advanced

Software: R, Rstudio/Posit (Cloud)

Dates: June 15-July 2, 2026

Time: 2:30-5:30 p.m.

Instructor: Dave Armstrong

Goals:

Conceptualize the idea of robustness beyond sign and significance.
Develop ways to measure robustness of statistical quantities (model parameters, predicted values, first differences, marginal effects) across different model contexts.
Apply robustness evaluation across a wide set of models and contexts.

Learning Outcomes:

Understand robustness in different domains - functional form, omitted variables, measurement, causal heterogeneity and sample-population mismatches.
Measure robustness of statistical quantities in each domain.
Learn to use general tools to independently apply robustness evaluation in any model setting.

Prerequisites: Students should have a solid understanding of the linear model. We will discuss other modelling contexts, but I will provide some basic information about the extension of the linear model to these other contexts.

Difficulty Level: Intermediate

Software: R, Rstudio/Posit (Cloud)

Dates: June 15-July 2, 2026

Time: 2:30-5:30 p.m.

Instructor: Adam Enders

Goals:

Provide participants the tools to properly measure social scientific constructs.
Learn multiple specific models and general strategies for developing new measures of social scientific constructs, ensuring the reliability and validity of such measures, and investigating the dimensionality of such measures.

Learning Outcomes:

How to explore the dimensionality of a dataset
How to produce statistically reliable and valid measures
How to present results, including using graphs
How to develop a new measure from the bottom up

Prerequisites: Participants should be familiar and comfortable with ordinary least squares regression and more basic statistics. Familiarity with matrix algebra and maximum likelihood estimation will serve participants well, but is not strictly required for participation or necessary for participants to understand course material.

Difficulty Level: Intermediate

Software: R, STATA/MP

Dates: June 15-July 2, 2026

Time: 2:30-5:30 p.m.

Instructor: Sarah Shugars

Goals:

Study systems that are interconnected, including networks of friendship, social media sharing, political ties, organizations, and more
Introduce fundamentals of network analysis.

Learning Outcomes:

Conceptualize and model research in which observations are connected
Describe properties of networks and the methods for studying them
Employ network analysis in R

Prerequisites: This course has no prerequisites and is open to people from all disciplines and research interests.

Difficulty Level: Suitable for all levels

Software: R

Dates: June 15-July 2, 2026

Time: 2:30-5:30 p.m.

Instructor: Jane Sumner

Goals:

This workshop provides a basic introduction to statistics, probability, and data analysis. The intention is to introduce students to all steps of the research process between theorizing and statistical modeling, and provide the foundational knowledge students need to learn statistical modeling. We will cover issues relating to measuring and collecting data, summarizing data, basic probability and statistics, and confidence intervals and preliminary hypothesis testing.

Learning Outcomes:

Explain general principles for how to measure a concept and collect data
Describe, visualize, and discuss distributions of individual variables
Articulate how probability is used for inference
Analyze and interpret bivariate relationships for discrete and continuous variables

Prerequisites: No prerequisites; Need to know what software they will plan to use for engaging with the lab session portion of the course.

Difficulty Level: Beginner

Software: R, RStudio

Dates: July 6-10, 2026

Time: 8:15 a.m.-12:15 p.m.

Instructor: Tim McDaniel

Goals:

Provide a broad and generalizable foundation in Matrix Algebra, Calculus, and Probability that will enhance learning in subsequent ICPSR Summer Program courses.

Learning Outcomes:

Introductory and advanced Matrix Algebra.
Differential and Integral Calculus.
Discrete and Continuous Probability distributions.

Prerequisites: Proficiency in basic mathematics (e.g., rudimentary algebra and summation notation) is assumed but not required. We will begin with a brief review of fundamental mathematics and probability.

Difficulty Level: Suitable for all levels

Software: None

Dates: July 6-10, 2026

Time: 1:30-5:30 p.m.

Instructor: Michael Hawthorne

Goals:

Introduce participants to core computing concepts and solutions applicable to analysis packages, particularly directed to participants having limited experience.
Introduce participants to the dataset development process, from beginning to end, examining real-world issues in creating and maintaining datasets through examples across disciplines.

Learning Outcomes:

Understand core procedures in three statistical software packages
How to implement the dataset development process
Understand connections between concepts in developing datasets and practical issues.

Prerequisites: There are no prerequisites required for the course. The course will be particularly beneficial for participants who have had limited experiences with creating their own datasets and/or using statistical software, or have been dependent upon having others assist in creating datasets and software commands for analysis, or are more advanced users now confronting new issues with datasets they wish to develop.

Difficulty Level: Beginner/Intermediate

Software: SAS, SPSS, STATA/MP

Dates: July 13-31, 2026

Time: 9:00 a.m.-12:00 noon

Instructors: Mark Nieman

Goals:

Learn the theory and (some of) the math behind categorical data analysis, understand how it works (assumptions), and how to interpret the results.

Learning Outcomes:

How to apply statistical estimators with limited dependent variables
When to use specific estimators given data characteristics
How to interpret statistical output in substantive terms.

Prerequisites: Helpful to have some familiarity with linear regression.

Difficulty Level: Beginner

Software: R

Dates: July 13-31, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Yaoyao Dai

Goals:

This course provides an introduction to data science for social data and problems, with a focus on utilizing text as data.
Learn through hands-on projects the fundamental tools of data science and apply them to a wide range of political and policy-oriented questions.

Learning Outcomes:

Have a good understanding of the landscape and frontier of text analysis and ML methods
Be able to identify and conduct the appropriate methods for their research questions.

Prerequisites: Students should have taken a research design and introduction level statistics class. Prior knowledge of coding is recommended but not required.

Difficulty Level: Suitable for all levels

Software: Python

Dates: July 13-31, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Desmond Wallace

Goals:

Practical learning and application of methods to answer important research questions.
Understanding appropriate methods for modeling spatially/temporally dependent data.
Investigating the role of maximum likelihood estimation concerning modeling dependent data.

Learning Outcomes:

Modeling data with a nesting structure, events occurring in time, and spatial/temporal characteristics.
Understand the issues modeling dependent data via ordinary least squares (OLS).

Prerequisites: Participants should have a background in matrix algebra, basic statistics, and linear regression. Previous experience in maximum likelihood estimation is recommended, but not required.

Difficulty Level: Intermediate/Advanced

Software: R, Rstudio/Posit (Cloud), STATA/MP, Python

Dates: July 13-31, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Tim McDaniel

Goals:

Have an applied and intuitive understanding of OLS Regression.
Apply and use basic and more advanced OLS Regression.
Critically identify potential problems related to OLS Regression.

Learning Outcomes:

Specify an OLS Regression model.
Test the assumptions of OLS Regression.
Use and interpret interactions and categorical variables.
Deal with nonlinearity.

Prerequisites: It is assumed that students are comfortable with very basic algebra and introductory-level statistics and now want to learn OLS Regression for their own research and to understand the work of others.

Difficulty Level: Intermediate

Software: Examples are in SPSS, but students can use any software package of their choosing.

Dates: July 13-31, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: Shane Singh

Goals:

Participants will review the basics of statistical inference and measures of bivariate association.
Participants will learn the theoretical underpinnings, assumptions, and mechanics of Ordinary Least Squares (OLS).
Participants will develop the ability to utilize OLS in their own research projects confidently and using modern software techniques and statistical graphics.

Learning Outcomes:

A fundamental understanding of the theory and math behind OLS.
How to interpret OLS results—and when such interpretation can be causal.
How to apply OLS in one's research.
Modern software techniques for OLS, including statistical graphics.

Prerequisites: Only familiarity with introductory statistics, probability, and basic algebra is assumed.

Difficulty Level: Beginner/Intermediate

Software: R, STATA/MP

Dates: July 13-31, 2026

Time: 9:00 a.m.-12:00 noon

Instructor: David Peterson

Goals:

Give students an understanding of how to incorporate theoretical modeling into their research process.
Introduce students to the mathematical and coding foundations of modeling
Establish frameworks for matching theoretical models to empirical data in coherent, motivated ways.

Learning Outcomes:

How to turn theoretical intuitions into models.
The math behind game theory and agent-based modeling.
How to match models to data and empirical methods.

Prerequisites: None

Difficulty Level: Beginner/Intermediate

Software: NetLogo (though no prior experience is required)

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructor: Marc Ratkovic

Goals:

Move from a solid understanding of Bayesian methods to cutting-edge applied techniques

Learning Outcomes:

Work with simple Bayesian models analytically
Implement Bayesian models computationally
Conceptualize Large Language Models in NLP
Implement basic chatbots and RAGs

Prerequisites: Familiarity with the topics covered in Bayes I; some calculus and basic probability theory will be introduced but prior exposure is helpful; coding in R will be helpful

Difficulty Level: Intermediate/Advanced

Software: R, Rstudio/Posit (Cloud), Python, Docker

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructor: Sebastian Calonico & Nicolás Idrobo

Goals:

Provide a thorough introduction to recent methodological developments to study causal inference in non-experimental settings.

Learning Outcomes:

Methodological and practical training in reweighting methods, regression discontinuity designs, difference-in-differences, synthetic controls and causal machine learning.

Prerequisites: It is assumed that participants have elementary working knowledge of statistics, econometrics and policy evaluation. It would be useful, but not required, if participants were familiar with basic results from the literature on program evaluation and treatment effects. This course is nonetheless meant to be self-contained and hence most underlying statistical concepts and results are introduced and explained in class.

Difficulty Level: Intermediate/Advanced

Software: R, STATA/MP

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructor: Rob Franzese

Goals:

Teach empirical modeling of/for social science theory (& why "causal-inference" testing is not enough.
Learn many useful empirical-modeling techniques/methods, incl. linear & nonlinear & interactions, random-coefficient/multilevel models, dynamic models including VARs, spatial-econometrics, instrumental variables & systems of equations, etc.

Learning Outcomes:

Difference testing for existence of causal effects v. estimating causal responses
How methods appropriate to latter differ from former
Many specific theory/substance-motivated models & methods

Prerequisites: The approximate background expected is: equivalent to a first graduate social science course in statistical data analysis through regression.

Difficulty Level: Intermediate/Advanced

Software: R, STATA/MP

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructor: John Poe

Goals:

Give everyone a solid grounding in the basic toolkit of models for multilevel data structures
Give everyone practical code to analyze complex multilevel data
Give everyone the ability to break incredibly complex problems down into manageable pieces.

Learning Outcomes:

How to analyze multilevel data in R.
How to talk about and think about assumptions and when/why/how models break
How to create an analysis plan for complex multilevel research questions

Prerequisites: Students should have taken an introductory class on multilevel modeling OR longitudinal data analysis. A background with maximum likelihood models will be helpful but isn't strictly required.

Difficulty Level: Intermediate/Advanced

Software: R, STATA/MP, We mostly use R but I can provide Stata code when needed.

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructors: Shahryar Minhas & Olga Chyzh

Goals:

Introduce students to advanced statistical models for network analysis.
Build hands-on skills applying these models using R.
Enable critical evaluation and application of these methods in research.

Learning Outcomes:

Researchers in social sciences, political science, or sociology working with network data.
Anyone with a foundational knowledge of network analysis and statistical modeling.
Scholars interested in applying advanced network models to their research questions.

Prerequisites: Basic understanding of network analysis concepts, familiarity with statistical modeling principles, and proficiency in R programming. Regression and MLE are helpful.

Difficulty Level: Intermediate

Software: R

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructor: Kenicia Wright

Goals:

Provide an overview of definitions and measurements used in academic research related to race and ethnicity
Explore the connection between theory and Research Design using data available through the ICPSR repository
Offer participants the opportunity to make progress, present, and receive feedback on a research proposal.

Learning Outcomes:

Definitions, theories, and measures of race and ethnicity across disciplines, publicly available datasets, development of feedback and presentations
Tools for fact-based, respectful discourse.

Prerequisites: Some familiarity with empirical methods and R or Stata. I will adjust course expectations, content, and structure based on student familiarity (will be accessed in the first few days of the Session).

Difficulty Level: Beginner/Intermediate

Software: R, STATA/MP

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructor: Sandy Marquart-Pyatt

Goals:

Provide participants with the skills necessary to utilize structural equation modeling with latent variables in their own research
Critically evaluate contemporary social science research using advanced quantitative approaches.

Learning Outcomes:

Recursive & nonrecursive models
Measurement/latent variable models (CFAs)
Model specification & identification, model fit, model evaluation
Mediation, multiple groups, latent class, longitudinal

Prerequisites: Participants should have a solid understanding of linear regression. Some background in matrix algebra, covariance algebra, and even factor analysis would be helpful but are not required.

Difficulty Level: Intermediate/Advanced

Software: R, Rstudio/Posit (Cloud), STATA/MP, MPlus

Dates: July 13-31, 2026

Time: 2:30-5:30 p.m.

Instructors: Mark Pickup & Andy Philips

Goals:

Gain an overview of prominent approaches to working with panel and longitudinal data
Develop skills needed to diagnose threats to inference and to identify appropriate analysis strategies for particular panel/longitudinal datasets common in the social sciences
Develop R and Stata skills to implement the approaches covered

Learning Outcomes:

When using panel and longitudinal data, understand: threats to inference; prominent approaches to modeling; how to apply different identification strategies using statistical software.

Prerequisites: A basic linear regression course is preferred, but not required. Experience with R and/or Stata is desirable, but not required.

Difficulty Level: Intermediate

Software: R, STATA/MP