全球智库信息集成服务系统

Evaluations of development projects are conducted to assess their net effectiveness and, by extension, to guide decisions regarding the merits of scaling-up successful projects and/or replicating them elsewhere. The key characteristics of ‘complex’ interventions – numerous face- to-face interactions, high discretion, imposed obligations, pervasive unknowns – rarely fit neatly into standard evaluation protocols, requiring the deployment of a wider array of research methods, tools and theory. The careful use of such ‘mixed methods’ approaches is especially important for discerning the conditions under which ‘successful’ projects of all kinds might be expanded or adopted elsewhere. These claims, and the practical implications to which they give rise, draw on an array of recent evaluations in different sectors in development.