Analyzing Network Data

Overview

Location: BPC 152

This is a 4-day (Mon-Thu) workshop that covers much of the same material as Introduction to Social Networks, but is more software intensive. It is also more technically oriented in terms of learning the formulas behind measures. It assumes that the participant has read articles that use network concepts, but wants to learn the nuts and bolts of how to really do network analysis. No prior experience using UCINET is assumed. In the labs, participants work independently to complete a set of exercises, and may also choose to work on their own data.

The first day provides a brief refresher about the network perspective which leads into a discussion of the mathematical foundations for social network analysis, particularly graph theory and matrix algebra. The day concludes with an extensive "hands on" tutorial of the two software tools used during the workshop: UCINET and NetDraw. The second day begins with a brief discussion of social network data, but focuses primarily on entering data into UCINET. Then, from Tuesday afternoon until Thursday morning, each half day session will cover a new analytic technique used in social network analysis, specifically: Measuring Centrality, Cohesive Subgroups, Network & Dyadic Cohesion, and Social Capital & Brokerage. Thursday afternoon is devoted to the special consideration in testing hypotheses based on network data. A schedule of topics is given at the bottom of this page.

The goal of this workshop, as compared to the "Introduction to Social Network Analysis" module, is to focus more on the analytic processes, focus a little more on the software tools (and also move at a slightly faster pace here), and allow more independent work. The workshop will assume fluency in using Windows programs, although not UCINET or NetDraw. For example, after the first day, the goal is for participants to work on the lab exercises independently (or in small groups of students) rather than taking time for the instructor to demonstrate how to calculate various measures in the software. (TAs will circulate helping people who get stuck.) During these labs, participants are encouraged to run the relevant analyses on their data if they have any, either instead of or in addition to running them on the sample data sets.

Topics

Monday
- (AM) Overview of basic concepts
- (PM) Installing UCINET; Introduction to UCINET and NetDraw
- Tuesday
  - (AM) Graph theory; Network data and data entry.
  - (PM) Centrality
- Wednesday
  - (AM) Brokerage and social capital
  - (PM) Cohesion
- Thursday
  - (AM) Hypothesis testing with QAP
  - (PM) Special topics: 2-mode data, data collection, etc.