For evaluators, the term "Theory of Change" is not just jargon but a critical element for our work.  As evaluators we are often asked: what will happen once an evaluation is done? What difference is it going to make, how will we know people learned, changed behaviors, or became more successful? And, are these changes commensurate with the money we spend on evaluation?

In the case of the World Bank Group, IEG have seen a correlation in World Bank Group projects between strong results-based management (results frameworks, M&E, work quality at entry and during supervision) and good outcome ratings. What explains that correlation? Does the theory of change apply to our work? If so, how?

In this blog, I try to explain how evaluation influences change, and why a theory of change matters in evaluation.

Simple Theory of Change. In its simplest form, the theory of change of an evaluation is that it will shed light on success and failure. By doing so it helps us make better decisions about what should be replicated and what should be stopped or avoided. It facilitates taking corrective actions and hence achieve better results.

A Different Theory for Independent Evaluation? Not really! Both self- and independent evaluation aim to influence change through critical reflection on what works and why to inform immediate and future actions and increase development effectiveness. The advantage of self-evaluation lies in shorter feedback loops. Lessons are generated and owned within the program management and self-evaluation process, and learning might take place faster. For independent evaluation, feedback loops are longer and learning can be impeded when evaluation findings are resisted. But, self-evaluation might have blind spots for things that are not working, or be bound by institutional norms that impede questioning certain issues. Independent evaluation can overcome these with a dispassionate, arms-length assessment that is not subject to undue pressure. In short: both have advantages and disadvantages that can be balanced out in a combination of both.

Reality Is More Complex Than Theory (of Change). Embedded in such a simple theory of change are factors that are more complex and affect the way in which change will occur as a result of evaluation. Assumptions made along the results chain are often not made explicit. For instance, we assume that enough evidence exists to explain the causal relationship between an intervention and observed changes. Or that evidence weighs heavily in decision-making processes, which however are often dominated by other considerations. Just like a development intervention intervening factors play an important role in explaining the influence of an evaluation. Many of them do not follow a linear results-chain but a complex system of interrelated feedback loops.

Turning Theory to Higher Value Added. These factors that can turn an evaluation into something highly influential or make it fall flat belong into four categories. Two within and two outside the control of the evaluation: those that affect the cost of the evaluation and those that affect its value. In each of these "boxes" the factor might have a positive or negative effect.

Clarity about what the evaluation tries to achieve, the assumptions we make about the channels of influence, and factors that might help or hinder its value-added, or increase or lower its costs, help improve evaluation effectiveness, and with that its value for money.

How does that work?

Let's take a leaf out of our book of advice to colleagues on the operational side of the business. When we design an evaluation, let's:

• Address issues that are of importance to stakeholders when defining the objectives of an evaluation, while sticking to the evaluation business of assessing past experience.
• Be explicit about decision-points, processes, or behaviors that the evaluation aims to influence.
• Be clear about our assumptions how the evaluation will exercise influence. Do any of them require changes in the evaluation design (methods and/or process) to ensure the evaluation makes as convincing a case it needs to?

Just like our assessment of World Bank Group projects - where high quality design and proactive management during implementation are essential contributors to success - I believe these measures increase chance that an evaluation will influence change. They can directly affect the "value" side of the equation, albeit that increasing "value-for-money" will require finding efficient ways to carry out the evaluation.