Multiple studies and weak evidential defeat

Research output: Contribution to journalArticle

Colleges, School and Institutes

Abstract

Having read a study showing statistically significant correlation between an exposure and an outcome our credence of a real connection between the two increases. Should that credence remain the same when we discover that further independent studies between the exposure and other independent outcomes were conducted? Kotzen argues that it should remain the same, even if we discover the results of those studies. We argue that it can differ dependent upon the results of the studies.

Details

Original languageEnglish
Pages (from-to)353–366
Number of pages14
JournalTheoretical Medicine and Bioethics
Volume38
Issue number5
Early online date4 Jul 2017
Publication statusPublished - Oct 2017

Keywords

  • Philosophy of epidemiology, Multiple testing, Multiplicity, Bayesian, Evidential defeat