In recent years nonprofits have begun to step up efforts to identify measurable outcomes, and then regularly collect and analyze their data to improve programs and services. This type of evaluation, often called performance management or performance measurement, typically utilizes internal staff and resources, not external evaluators. Nonprofit thought leaders and funders, including Bill Gates in his 2013 Annual Letter, have advocated for the importance of this type of system with Gates noting that, “setting clear goals and finding measures that will mark progress toward them can improve the human condition.”
Although more organizations engage in performance management efforts, there are some who argue that in order for high-quality evaluation to occur, nonprofits need to assess causation, i.e. did the program cause the change in participants’ attitudes, behaviors, etc. This type of scientific proof often demands more rigorous types of evaluation methodology, such as a randomized control trial. So, if an organization doesn’t assess causation, does this mean nonprofits shouldn’t bother with their own data collection and evaluation?
Earlier in my career I too struggled with this question. When I worked as a school social worker we regularly collected and reviewed data on students’ grades, behavior, and attendance. As I collected this data I often found myself thinking, “Isn’t collecting all of this data kind of a waste of time?” “How do we know if a student’s attendance improvement is a result of our program or a million other factors outside of our control?”
Over time I came to realize that while I was correct in my belief that our data collection efforts were not enough to determine causation, the performance management process was nonetheless critical to the work we were doing. With this information we could assess the progress of students and more quickly identify when a student was experiencing difficulties. From this data analysis we could also begin to identify possible reasons why the student was struggling and offer increased assistance or a different type of intervention. If we had waited until an outside evaluator came in to assess the program using more sophisticated methodologies, it would have been too late to make program adjustments for our students. Instead we used the evidence we did have to guide our practices and decisions. Mario Marino calls this “information based introspection” and it is critical in helping organizations determine what’s working well, what’s not, and where they should focus their intervention efforts.
Although I am advocating for more of a focus on results rather than scientific proof of causality, it is also critical for organizations to build a strong case for how their program contributes to the outcomes they are hoping to achieve. An organization may not be able to prove that their program caused a particular outcome, but they can work to establish a plausible link between their activities and the results achieved. Ideally, organizations will be able to utilize evidence-based studies (some of which do assess causation!) to help establish this link. Only once an organization has established a strong link, and then engaged in a regular system of performance management, might it make sense to focus on a more methodologically rigorous type of evaluation.
External evaluations which seek to determine causation are often a highly expensive and time intensive process. While the sector relies on these types of evaluations to help establish a strong evidence base, most nonprofits should consider using performance management systems first and foremost. It is this ongoing cycle of data collection and analysis that will provide organizations with the process needed to continuously improve programs and services, and hopefully achieve lasting impact.