Research is an integral part of the work at Tsao Foundation. As pathfinders, we are  deeply committed to understanding whether our programme innovations work and how they can be improved, replicated and scaled up. Our latest project - the Community for Successful Ageing (ComSA) - is no less subject to the same evaluation and scrutiny.

In the past, using such traditional methods as randomised controlled trial to assess the outcomes of an innovation had at times provided a 'verdict' on its effectiveness without nessarily much accompanying insight into its 'why' and 'how'. As ComSA is an open system that comprises multiple complex interventions, research on it can be expected to be greatly challenging.

The term 'black box evaluation' has been used to highlight the inadequacy of approaches that do not capture the complexities and organisational structure of an intervention. To avert it, ComSA will be using a theory-based approach called ‘Realist Evaluation’.

Realist Evaluation is not a method or a technical procedure, rather, it is a logic of inquiry that attempts to answer the question: ‘What works, for whom, in what circumstances and why?’ Realist Evaluation is gaining momentum among researchers and evaluators of complex interventions to improve healthcare. However, this theory-driven research methodology still goes largely unknown and/or is perceived with scepticism in some quarters. The purpose of this page, therefore, is to consolidate and share learning about Realist Evaluation with our partners and colleagues. We will upload all lectures, workshops and seminars hosted by us and also provide web links to useful learning resources elsewhere.

Title: Realist Evaluation: Opening the black-box evaluation
Short abstract: This is a lecture conducted by Prof Bert Vrijhoef at Tsao Foundation on 27 Feb 2017. The lecture focuses on: why do we need Realist Evaluation, in what circumstances, for whom and how does it work? Examples are presented from the field of health services and policy research and implementation science.