We have been working with TASO to develop guidance on evaluation with small cohorts and it is now available on their website. TASO’s focus is widening participation and improving student outcomes in higher education but the guidance can be applied to many areas of social policy evaluation.

Increasingly, public bodies aspire to draw on evidence that supports a causal interpretation to inform their work. The need for such evidence is beyond question – public bodies want to engage in activities that are effective and proven to be so – but the search for causal evidence also leads them into the difficult and contested area of ‘causal’ inference and the related paradigm wars in the social sciences.

One increasingly common challenge is the need to reconcile the nature of the programmes and policies for which causal evidence is required with the conventional ‘what works’ toolkit, comprising systematic review, randomised controlled trial and quasi-experimental design. Many interventions are not compatible with these forms of causal inference, due to their complex nature, resource issues, emergent or developmental features or small scale. Faced with these challenges, evaluators, service-deliverers and policymakers are increasingly seeking alternative understandings of causality to fill the perceived gaps in actionable evidence.

Some of these alternatives are the focus of this report, specifically designed for impact evaluation that can be used with small cohorts: so-called ‘small n’ impact evaluations.