How to Commit the Gambler's Fallacy and Get Away with It


In a recent article Ian Hacking argues that there can be cases where no probabilities may correctly be ascribed to individual members of a population, while probabilities are correctly ascribable to the population as a whole. In this paper a simple artificial coin-flipping model for such probabilities, not 'grounded from below' is constructed. The inferences licensed by this model and a consequence of the model for the theory of statistical tests is explored.

Download options


    Upload a copy of this work     Papers currently archived: 72,660

External links

Setup an account with your affiliations in order to access resources via your University's proxy server

Through your library


Added to PP

17 (#641,687)

6 months
1 (#388,784)

Historical graph of downloads
How can I increase my downloads?