A Note on 'Rethinking John Snow’s South London study: A Bayesian evaluation and recalculation'

Abstract

Koch and Denike (2006), in examining Snow's analysis of cholera in south London in 1854 (Snow (1856)) have claimed that "Snow made ... conceptual mistakes that adversely affected his results." Koch and Denike's claims, analysis, and conclusions are wrong: they mis-read or mis-interpret the underlying data; they incorrectly apply an inappropriate statistical test; they alter the underlying primary-source data; and some of their published test results cannot be reproduced using their published data. Although Snow's analysis fails to meet modern standards of statistical rigor (primarily due to the lack of appropriate tools in Snow's era) Snow's approach and his claims about the strength of the effect of water do survive re-analysis using modern statistical methods.

Keywords: John Snow, Cholera, Causal Inference, Epidemiology, Statistical Methodology, History of Science

JEL Classification: C18, N33, N93, B40, C52