Intermediate SNA (UCINET)

Overview

This course is taught by Steve Borgatti. It is a more technical and in-depth workshop than the Introductory workshop, but covers many of the same concepts. It focuses on the concepts and methods of SNA, particularly as they apply to specific research objectives. In this course, everything is related back to the research questions -- how the network analysis relates to consequences of interest. In addition, the mathematics and algorithms behind the measures and techniques are explained. Prior familiarity with network analysis and research in general is assumed.

This workshop uses the UCINET software package extensively. Important note: UCINET is Windows-only software. Please visit our software page in advance of the workshop.

The course meets for four consecutive days starting June 14: Monday, Tuesday, Wednesday, and Thursday from 10:00-12:00 EDT and 12:30-2:30 EDT for a total of 12 contact hours. There is a half-hour lunch break at 12. TAs (scroll down for contact info) will be available during the sessions to catch people up, answer questions, and bring questions to the instructor's attention. The TAs will also be available for an hour before and after each session for both group and 1-on-1 work, including help with homework assignments.

Main Schedule

1 Monday - Intro

  • Overview of network analysis: the network perspective

  • Working with network data in UCINET

  • Visualizing network data

2 Tuesday - Groups

  • Characterizing groups / whole networks

  • Community detection

3 Wednesday – Node-level constructs

  • Ego network measures

  • Centrality

4 ThursdayStatistical models

  • MR-QAP & LR-QAP

TA contact information

Recommended Readings

  • Borgatti, SP and Everett, MG. 2021? “Three Perspectives on Centrality.” In The Oxford Handbook of Social Networks, edited by James Moody. Oxford University Press [pdf]

Special Data

  • In case some people have issues with using the "data for learning ucinet.xlsx" file, some key datasets in ucinet format can be found here