by Stephen Tall on May 1, 2014
I never tire of reminding liberals suspicious of evidence-based policy informed by Randomised Controlled Trials that they are literally are a liberal idea: they’re founded on father-of-liberalism John Stuart Mill’s ‘method of difference’ through which the observed difference between two groups can be measured to identify the ‘active ingredient’ which works.
However, it turns out I might have under-estimated the historical precedent for control group analysis to guide decision-making:
Assignment of treatment to different groups and subsequent comparison of outcomes dates as far back in history as the Old Testament, in which King Nebuchadnezzar is said to have ordered a group of his subjects to eat rich meat and drink wine while another group was made to adhere to vegetarianism in order to evaluate the merits of the two diets. (1 Daniel 11-16, New International Version)
(‘Risk and Evidence of Bias in Randomized Controlled Trials in Economics’, Eble, Alex and Boone, Peter and Elbourne, Diana, Risk and Evidence of Bias in Randomized Controlled Trials in Economics (January 6, 2014). Mimeo, (2013), Brown University. Available at SSRN: http://ssrn.com/abstract=2272141). Hat-tip: @TimHarford.
And, incidentally, the first RCT in this country wasn’t a medical one: it was in the field of education in 1929:
(Hat-tip: A brief history of RCTs in Education (over the last 100 years) – Professor Carole Torgerson, Durham University).