Behaviour analysis and evidence-based education

Education can have a powerful and long-term effect on people's lives and therefore should be based on evidence of what works best. This assertion warrants a definition of what constitutes good research evidence. Two research designs that are often thought to come from diametrically opposed fields – single-subject research designs and randomised controlled trials – are described and common features, such as the use of probabilistic assumptions and the aim of discovering causal relations, are delineated. Differences between the two research designs are also highlighted and this is used as the basis to set out how these two research designs might better be used to complement one another. Recommendations for future action are made accordingly.