Network Analysis I: Introduction
Social network analysis (SNA) is an increasingly used perspective for describing and modeling the relationships between social actors. This course will lay the groundwork behind social network analysis (SNA) from conceptual, mathematical, empirical and computational perspectives. This approach will draw from the rich multidisciplinary history that has shaped the field's development - incorporating perspectives from sociology to physics, math to public health.
SNA differs from other analytic perspectives in ways that require unique strategies for data collection, storage, descriptive and statistical analysis. The course will address each of these by sampling from a range of the most commonly used analytic concepts, and demonstrate their empirical applications, and computation (primarily in R). Those concepts will be presented around two organizing principles: (1) the two primary theoretical frameworks researchers employ to capture different reasons we think networks matter; and (2) how each class of measures can be applied across different units of analysis: individuals, groups and 'whole' networks. Within the course we will address topics including data collection, representation, ego-network composition, social balance, distance, density, centrality, bipartite (two-mode) networks, clustering, group cohesion, equivalence & social roles, along with an introduction to statistical models (e.g., ergm/p* and SABM). While by no means exhaustive, this introduction will provide the tools with which participants will be equipped to dig deeper into SNA on their own.
Please note: While we will introduce statistical models in this course, (e.g., erg/p*, SAB models) they are the focus of the advanced course offered in the second session. We will devote more of our attention in this class to whole sociometric network analysis, not ego networks; though those will make an occasional appearance.
Software: All analyses in this course will be conducted in R (using the igraph and statnet packages), with no assumed prior experience with R required.
There will be computational tutorials most class days, with periodic corresponding assignments for participants who are interested in receiving feedback on their use of software and development of research ideas.
Fees: Consult the fee structure.
Location: ICPSR -- Ann Arbor, MI
Date(s): June 26 - July 21
Time: 1:00 PM - 3:00 PM