Please note: This course will be taught online only. In person study is not available for this course.

John Barry Ryan is an associate professor at the University of Michigan with a joint appointment in the Departments of Political Science and Communication and Media. He studies political campaign communication, social influence, and public opinion. His latest co-authored books are The Other Divide and Partisan Hostility and American Democracy.

Course Content

The focus of this course is studying social influence. Standard statistical models and survey measures implicitly assume that individuals are atomistic and come to their attitudes without influence from others individuals – who might also be in the dataset. There are various techniques for overcoming these assumptions and studying individuals as interdependent members of a social group. Social network analysis considers the entire network – a population of connected people. This methods requires a different conceptualization of what is a “case” in a dataset. It also requires a particular dataset that can be hard to obtain. For this reason, we will also examine alternative means for studying social influence using surveys and experimental methods.

Course Objectives

At the end of the course, participants should be able to:
• Collect social network data for studying public opinion.
• Calculate summary statistics for a network.
• Produce network graphs in R.
• Design surveys for studying influence without using social network analysis.

Course Prerequisites

Course participants who want to take part in the hands-on analysis components of the course will benefit from having some familiarity with R, but it is not required. Participants should be comfortable with descriptive statistics and should be familiar with OLS regression. No additional background readings are required for this course.

Background knowledge required


Linear regression – moderate


OLS – moderate

Maximum likelihood – elementary






Methodological Issues in Studying Social Influence

Studying interdependence violates standard statistical assumptions. We discuss the issues this causes and what can be done about them.


Social Network Datasets

How to organize your data to perform social network analysis.


Describing Networks

Concepts and measures used to summarize networks and actors within networks.


Creating Network Graphs

How to visualize social networks using R.


Models in Networks

Methods to predict ties in networks and for comparing networks.


Alternatives to SNA: Ego Networks and Snowball Surveys

Often it is not possible to obtain the whole network; we discuss alternative survey methods for studying social influence.


Alternatives to SNA: Measurement in Ego Networks

How to write survey questions to measure social context.


Alternatives to SNA: Group Based Experiments

Designing experiments to study social influence in a lab or online.


Alternatives to SNA: Analyzing Group Based Experiments

Group-based experiments violate standard statistical assumptions, but there are ways to relax those assumptions