Network Evaluation: Avoiding the “Straitjacket”

March 2010

In our experience, people who build networks for social change are deeply curious about their network’s performance, but they are wary of the conventional evaluation “straitjacket.” They can’t imagine how a rigid assessment framework could be usefully applied to the dynamic, self-organizing network they are nourishing. And they wonder how an evaluation approach designed to assess organizational practice could possibly capture the far more complex practice of network organizing.

When we design an evaluation for a network, we do draw on conventional evaluation principles but we also use a unique network evaluation framework to track and document a network’s evolution and outcomes. We look at things having to do with networks as a distinct organizing form, such as network structure and composition: Who is connected to whom? What is transacted through these links? We also track value creation (What value does the Network produce both for individual members and for the broader constituency it serves?) and internal network conditions that contribute tonetwork health (such as complementary capacities and diversity).  Although it is sometimes difficult to tease out contributing factors, we try to design evaluations that allow network builders to assess the relationship between network organizing and network impact. What difference did network organizing make and why?

One way to document the difference that network organizing makes is to compare performance across networks with similar goals and different network organizing practices. Have you had any experience with this approach? What do you think of it? What approaches are you using?

When we first started thinking about network evaluation we found the work of two Canadians very helpful:  Heather Creech and Terri Willard through the International Institute for Sustainable Development.