SNA TIG Week: Stacey Friedman on Defining Network Boundaries in Social Network Analysis

My name is Stacey Friedman and I am co-chair of the Social Network Analysis (SNA) TIG. I work for a nonprofit agency that supports health professions educators around the world. One of our aims is to create and sustain an active international community of individuals interested in improving the education of doctors, nurses, dentists, and other health professionals.

One challenge with SNA in evaluation is defining the members of a network by setting network boundaries. This is critical with SNA, as often you need information from all members of a network and sampling a subset of the network is not an option.

For example, my organization supports fellowship programs for health professions educators. The network of interest could be all fellows. However, depending on the evaluation question, we might use more narrow or broad boundaries, such as fellows from a given geographic region, or all those involved in health professions education (such as faculty, students, school administrators) in a given geographic region. The possibilities are seemingly endless!

Hot Tip #1: Stay focused and realistic when defining your network boundaries. If you are studying student relationships, do you need to know about students in an entire school system, one particular school, or just one classroom? In addition to network size (and resources needed to collect data from a large network), consider accessibility of data from or about network members. Missing data can be problematic for SNA (for more on this, see the AEA365 post this week by Todd Honeycutt). It is important to choose your network boundaries wisely!

Hot Tip #2: Does your network have naturally occurring boundaries? Knowing the answer will help you decide if you need to create boundary criteria.

  • Examples of naturally occurring (a priori) boundaries:

–          Classroom (e.g., students in a third grade classroom)

–          Organization (e.g., staff of a public health agency)

–          Kinship group (e.g., three generations of a family)

–          Community (e.g., residents of a small town)

  • Examples of boundaries defined by the investigator:

–          Spatial area (e.g., those living within a 10 km radius of a city center)

–          Demographic criteria (e.g., those with household incomes below the poverty level)

Rad Resource: The SNA e-text by Hanneman & Riddle (“Introduction to Social Network Methods,” which can be found at includes a discussion of network boundaries in Chapter 1.

The American Evaluation Association is celebrating SNA TIG Week with our colleagues in the Social Network Analysis AEA Topical Interest Group. The contributions all this week to aea365 come from our SNA TIG members and you can learn more about their work via the SNA TIG sessions at AEA’s annual conference. Do you have questions, concerns, kudos, or content to extend this aea365 contribution? Please add them in the comments section for this post on the aea365 webpage so that we may enrich our community of practice. aea365 is sponsored by the American Evaluation Association and provides a Tip-a-Day by and for evaluators.

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.