AEA365 | A Tip-a-Day by and for Evaluators



SNA TIG Week: Maryann Durland on the Social Network Analysis Application Standard

Hi, I’m Maryann Durland, an independent evaluator and Social Network Analysis (SNA) practitioner. In this post I will address the requirements for doing an SNA application, particularly in evaluation, and which we could also call the standard for an application. I will use the early literature that formed and grounded SNA thinking and that continues to be relevant.

Early on in the history of defining SNA, Linton C. Freeman, described four requirements for completing a social network analysis in his book, The Development of Social Network Analysis: A Study in the Sociology of Science:

  1. A structural perspective
  2. Empirical data
  3. Graphics
  4. Mathematical models with analysis

I believe and promote, particularly in evaluation applications, that these are still the requirements for meeting the standard for doing SNA. Evaluations using SNA are distinct from research on SNA theories and measures, which may have different requirements.

In evaluation applications structural perspective means that we can define relationships within the program and these relationships create a structure through which information flows, resources are found, barriers are identified, spaces are found that need connections, and so on. Data is the existence or non-existence of a relationship between two elements. Empirical data refers to verifiable data collected on the relationship between any two elements, also called the members of a set.

Just like traditional data collection, we collect relational data through a variety of methods from observations to surveys about experiences. The data we collect populates matrix cells, indicating the presence or degree of a relationship between two members. Graphics indicate that we can visualize the network, results and/or analysis in graphs such as sociograms. Mathematical models with analysis allow us to calculate SNA measures which are measures of the network, not attributes assigned to individuals. Models called algorithms or a set of procedures, are as much a description of the relationship as they are the algorithm for how to calculate a measure.

Lesson Learned: Clearly, these four requirements delineate a specific methodological basis that is different from traditional quantitative and qualitative analysis. These requirements mean evaluators must think differently, ask questions for a different purpose, and conceptualize an evaluation differently.

Rad Resource: Early literature on SNA sought to develop what we could call standards for applications and one of the most important resources is the work of Linton C. Freeman. Freeman’s work continues to set the standards for SNA applications and the reference for the requirements.

The American Evaluation Association is celebrating Social Network Analysis Week with our colleagues in the Social Network Analysis Topical Interest Group. The contributions all this week to aea365 come from our SNA TIG members. 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. Would you like to submit an aea365 Tip? Please send a note of interest to aea365 is sponsored by the American Evaluation Association and provides a Tip-a-Day by and for evaluators.

No tags


  • Susan Sloan · July 22, 2015 at 7:11 am

    What you’ve said is fascinating but could you provide a specific example of how stakeholder expectations can be incorporated in an innovative way? Are you talking about the analysis itself or about conversations with stakeholders that should be occurring once the SNA is produced?


  • rick davies · July 21, 2015 at 7:58 am

    The four requirements for completing a social network analysis” may be sufficient for SNA research but I don’t think they are sufficient for SNA use in an evaluation context. Notably they miss out what is is a crucial element in many evaluations i.e. people’s participation. Stakeholders expectations about network membership and structures is vitally important when tying to make sense of SNA data. In research uses of SNA this is less likely to be the case. My exposure to SNA literature over the last decade or so suggests to me that this is where innovative evaluators could make a significant contribution to SNA practice.


Leave a Reply




To top