Theory-based policy and programme evaluation

Development policies and assistance strategies are no less complex or less challenging to evaluate than public policies in domestic economic and social fields. Indeed, they are almost certainly more complex and challenging.

Nevertheless, ministers and senior officials persistently demand firm evidence that aid 'works'. Evaluators are therefore under pressure to generate palpable proof, preferably in numerical form, of the impacts on livelihoods that are attributable to aid or to particular modes of aid delivery. Modern evaluation theory suggests that such expectations of what evaluation can provide are unrealistic and – worse – damaging to its ability to deliver well what it can provide in terms of learning from experience and accounting for decisions.

We work to bring up-to-date thinking into the design and implementation of evaluations. Alongside colleagues in other ODI programmes, including the Centre for Aid and Public Expenditure and Research and Policy in Development (RAPID) Programmes, we have developed exercises that shed light on major areas of aid or significant innovations in policy, such as sector budget support modalities, voice and accountability basket funds and regional aid strategies. We have a particular interest in the evaluation of governance programmes, but also in governance dimensions of other issues. We try to ensure that political and political economy insights get due attention in any evaluation work.