Lloyd Provost, writing in BMJ Quality and Safety, argues for a distinction between enumerative and analytical studies. Enumerative studies, he says, are suitable for measurements of static samples, like the water properties in a pool, while analytic studies track changes over time, as in sampling water in a river. Analytic studies help unravel cause and effect, says this article from the Institute of Healthcare Improvement. You don’t have to be a professor of philosophy to drive a coach and horses through this distinction. Science is always concerned with causal mechanisms or, at the very least, with predictions – pick up any textbook on the philosophy of science. Philosophers talk of the ‘scandal of induction’, since there is never a fool-proof way to predict the future – think black swans. So theory is always required to help make judgements about past results and its implications over time and place. If you want to test the water in a pool to see if it is safe to swim, then that is not a scientific exercise. If you want to find out whether a certain chemical kills fish in pools, then that is a scientific exercise and you had better sample plenty of pools with live and dead fish in them.
— Richard Lilford, CLAHRC WM Director
- Provost LP. Analytical Studies: A Framework for Quality Improvement Design and Analysis. BMJ Qual Saf. 2011; 20: i92-6.