Cause and effect in small n impact evaluations

Addressing attribution of cause and effect in small n impact evaluations: towards an integrated framework

February 1, 2012

Speaker: Dr. Howard White, 3ie

The drive to demonstrate results is leading to an increased focus on impact evaluation. There has been a substantial rise in the use of ‘large n’ experimental and quasi-experimental impact evaluations. But what can be done when there are insufficient units of assignment to apply statistical tests to determine the difference in outcomes between the treatment and comparison group? The need for such ‘small n’ approaches can arise when investigating one or a few cases, such as capacity building in a single organisation or a policy introduced at the national level, or possibly when an intervention has significant heterogeneity, with different recipients receiving different forms of support. This presentation will discuss some of the methodologies proposed for tackling causal inference in small n cases, including Realist Evaluation, Contribution Analysis, Outcome Mapping and Most Significant Change. Dr. White examines the possibility of drawing out common elements from these different methodologies and developing an integrated framework for small n analysis.

Scroll to Top