Retelling the John Snow Story: The Logic of Scientific Inquiry and Causality

Keywords: scientific reasoning, John Snow, education
Session Type: Virtual Paper Abstract
Day: Monday
Session Start / End Time: 2/28/2022 09:40 AM (Eastern Time (US & Canada)) - 2/28/2022 11:00 AM (Eastern Time (US & Canada))
Room: Virtual 47

Abstract

The case of John Snow, the Broad Street pump, and the search for how cholera was transmitted in mid-19th-century London is a classic example in spatial epidemiology and medical geography. This presentation retells the story of this case, with a focus on the logic of scientific inquiry and causality. The dominant frameworks used to teach causality in data science courses in the social sciences are from statistics (potential outcomes) and computer science (graph theory). These frameworks focus on measuring the effects of causes, as in the identification of treatment effects in randomized control trials. This presentation compares this narrower statistical causal framing of the problem today to an approach that focuses on the causes of effects or underlying theories and on the broader history and logic of scientific discovery, thus providing a valuable template and guide for how to obtain more reliable scientific evidence. It is an example of how to reason with data and potential explanations in a way that seeks to repeatedly reject or falsify explanations in an abductive process. The presentation also demonstrates why today’s re-analyses of Snow’s data to show that water was causal using a potential outcomes framework do not tell the whole story: By 1852, it was widely accepted that water was a cause in cholera transmission. However, Snow had to differentiate between alternate theories, which required both a broader perspective and a broader range of data and analysis than for a simple treatment effect.