MIXED METHODS AND CROSS-CUTTING APPROACHES
20 Theory-based Evaluation
Agathe Devaux-Spatarakis
Abstract
Theory-based evaluation was developed in response to the limitations of experimental and quasi-experimental approaches, which do not capture the mechanisms by which an intervention produces its impacts. This approach consists of opening the “black box” of public policy by breaking down the different stages of the causal chain linking the intervention to its final impacts. The hypotheses thus formulated on the mechanisms at play can then be tested empirically.
Keywords: Qualitative methods, theory, causal chain, intervention logic, theory of change, impact pathway
I. What does this approach consist of?
Formalised in the 1970s and 1980s in the United States, theory-based evaluation (TBE) is not a method, but rather a logic of evaluative research, an analytical approach that can mobilise several evaluation methods or tools. The term “theory” is to be understood here as the decomposition and explanation of a causal chain linking the public intervention to its expected results and impacts on the targeted and beneficiary publics. This takes the form of a graphic representation which then becomes the “scaffolding” underlying the evaluation investigations. Depending on its use, this tool may be called an “intervention logic”, “theory of change” or “impact pathway”. This approach is widely used in evaluation practice, although it covers a diversity of practices of varying quality and using more or less rigorous approaches.
Theory-based evaluation, contrary to what this name might suggest, seeks to be as close as possible to the reality on the ground and the context of the intervention. The term theory takes on a more or less scientific meaning depending on usage. At the very least, a theory-based approach is concerned with the theory of implementation of an intervention, i.e. the actions that are put in place for the proper implementation of the intervention in order to achieve the first results (intervention logic). In most cases, it tends to go a step further and make explicit the hypothetical causal links between an intervention and its intended outcomes up to and including final impacts (theory of change). This analysis is generally structured in several categories:
- Outputs represent the implementation of actions by public authorities,
- The results are the first immediate effects, directly linked to the achievements and which can be observed on the public participating in the actions,
- Intermediate and final impacts describe the expected impacts of the outputs on the final beneficiaries whose situation is to be improved.
This process systematically begins with the development of the ‘theory’. This is generally based on a literature review and discussions with the stakeholders of the intervention in order to identify the key components of the intervention, the target audiences, the main expected results and the final impacts targeted by the inter