When Was Random Allocation First Used To Generate Comparison Groups In Experiments To Assess The Effects Of Social Interventions?

Random allocation is an important feature of experiments designed to assess the effects of interventions: it ensures that, in respect of measured and unmeasured factors of prognostic importance, the comparison groups generated will differ only by chance. It has been asserted that random allocation in experiments to assess the effects of social and educational interventions was introduced at least as early as the first quarter of the 20th century. However, because the term 'experiment' and words with the root 'random-' have not been adequately defined and kept apart in these accounts, there is still confusion regarding the studies that have been cited as examples on early randomisation. We examined these putative examples and found that they were not randomised trials. It seems that matching on prognostic variables was the predominant method used to generate comparison groups in social and education intervention studies. The earliest description of a random allocation procedure in a social or educational intervention study that we were able to identify, was published in 1928.