In its pre-inception phase, CEDIL identified four main gaps in evaluation methods to inform the research agenda for the inception phase:
- Thematic and geographic evidence gaps: much of the evidence produced by impact evaluations of development projects over the last 15 years is concentrated on specific geographic areas and in particular sectors of interventions, like health and education. Various sectors have been neglected by impact evaluations including climate change; conflict and humanitarian settings; governance and infrastructure. Study areas have also neglected marginalised populations and geographical regions such as North Africa.
- Methods gaps for primary studies: there are many evaluation questions in international development for which there is no consensus as to the best available approach to answering the question. For example, in rapid onset emergencies, when the number of project observations is very small, and in complex interventions which may evolve over time with emergent outcomes. New technologies and evaluation methods are being developed by different disciplines that need to be adapted to the evaluation of development programmes and tested, for example, process tracing, QCA, integrative mixed method analysis and predictive analysis.
- Synthesis gaps: summaries of evidence through systematic reviews and other synthesis products are relatively new in international development. In addition, the synthesis of evidence on the effectiveness of development interventions presents several challenges related to the greater complexity of interventions and heterogeneity of contexts in comparison to other fields. Methods of synthesis need to be developed or adapted to the development context.
- Evidence translation gaps: producing rigorous evidence is not sufficient to ensure its use, nor are the passive dissemination strategies of the past. CEDIL will develop models of stakeholder engagement to increase the relevance, interpretation and application of study findings for policy and practice. Policy uptake also depends on the timeliness of evaluations and CEDIL will explore approaches such as adaptive learning impact evaluations, predictive analysis and real-time data-based approaches to produce more timely and relevant study findings.