3ie Working Paper 26

Power calculation for causal inference in social science: sample size and minimum detectable effect determination

3ie Working Paper 26

Eric W Djimeu and Deo-Gracias Houndolo

This manual, developed in-house by 3ie's Eric Djimeu and Deo-Gracias Houndolo, presents the basic statistical concepts used in power calculations for experimental design. The authors draw from real world examples to calculate statistical power for individual and cluster randomised controlled trials. The manual is accompanied by the Sample size and minimum detectable effect calculator©, a free online tool developed by the authors that allows users to work directly with each of the formulae presented in the manual. 


Integrating impact evaluation and implementation research to accelerate evidence-informed action

Working paper 3ie 2018
The authors of this paper use three case studies in which researchers and implementers integrated implementation research and impact evaluation, highlighting what worked or what was missed and provide practical recommendations.


Synthesis of impact evaluations of the World Food Programme’s nutrition interventions in humanitarian settings in the Sahel

Working paper 3ie 2018

Acute malnutrition in the Sahel region affects an estimated 6 million children under the age of 5, of whom approximately 1.4 million require treatment for severe acute malnutrition.


Community-driven development: does it build social cohesion or infrastructure? A mixed-method evidence synthesis

Working paper 3ie 2018

In community-driven development (CDD) programmes, community members are in charge of identifying, implementing and maintaining externally funded development projects.

Evaluating advocacy: an exploration of evidence and tools to understand what works and why

Evaluating advocacy: an exploration of evidence and tools to understand what works and why

Working paper 3ie 2017

This working paper adds to the growing study and practice of advocacy evaluation by examining the challenges associated with evaluating advocacy initiatives.


Can we obtain the required rigour without randomisation?

Working paper 3ie 2011

Karl Hughes and Claire Hutchings describe problems development organizations face in moving beyond outcome measurement.