Long, Harry;
(2024)
Do the kinds of standards applicable in statistical reasoning apply in philosophical reasoning? The example of Simpson’s paradox.
Masters thesis (M.Phil.Stud), UCL (University College London).
Preview |
Text
HDL_MPhil_Thesis__Revised_26_July_2024_.pdf - Accepted Version Download (858kB) | Preview |
Abstract
My research intersects the philosophy of science and metaphilosophy, with two overarching key aims. First, to investigate the extent to which, how, and when the kinds of standards we apply in statistical and scientific reasoning apply in instances of case-based philosophical reasoning putatively analogous to controlled scientific experimentation. Second, to discover — in those instances in which those standards do apply — how exactly to meet those standards, so that philosophers executing such reasoning may do better, by their own lights. This thesis represents the beginning of my research. In particular, it focusses on the problem of Simpson’s paradox, for which many solutions have been developed, and are frequently deployed, in statistical reasoning. The aim of this thesis is twofold. First, to elucidate the relevance of Simpson’s paradox to philosophy. This is by three means: by providing a necessary condition for the paradox, by relating it to a collection of other problems of keen philosophical interest, and by proposing solutions to these problems — where available — derived from the tools employed in statistical reasoning. These problems include the problem of base rate neglect, the Monty Hall problem, the Newcomb problem, counterexamples to the Dominance and Sure-Thing Principles, and problems with the ex ante Pareto principle. In each case, I aim to clarify the relationship between Simpson’s paradox and the problem at hand. Additionally, I clarify the relationship between Simpson’s paradox and the more general reference class problem, as well as the relationship between the reference class problem, Bertrand’s paradox, and the problem of (vindicating) induction. Second, to propose a condition that any philosopher employing case-based reasoning putatively analogous to controlled scientific experimentation must satisfy in order for her reasoning to be valid. The condition requires such reasoners to do two things. First, to check that their reasoning is not vulnerable to an instance of Simpson’s paradox. Second, if there is a suspected instance of Simpson’s paradox, to handle it by using the appropriate tools developed in statistical reasoning. If these conditions turn out to be difficult to satisfy, then the aspiration that the reasoning is analogous to controlled scientific experimentation may need to be dropped or attenuated.
| Type: | Thesis (Masters) |
|---|---|
| Qualification: | M.Phil.Stud |
| Title: | Do the kinds of standards applicable in statistical reasoning apply in philosophical reasoning? The example of Simpson’s paradox |
| Open access status: | An open access version is available from UCL Discovery |
| Language: | English |
| Additional information: | Copyright © The Author 2024. Original content in this thesis is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) Licence (https://creativecommons.org/licenses/by-nc/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request. |
| Keywords: | epistemology, metaphilosophy, philosophy of statistics, thought-experiments, Simpson's paradox, reference class problem |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL SLASH UCL > Provost and Vice Provost Offices > UCL SLASH > Faculty of Arts and Humanities UCL > Provost and Vice Provost Offices > UCL SLASH > Faculty of Arts and Humanities > Dept of Philosophy |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10197116 |
Loading...
Loading...Loading...Loading...Archive Staff Only
 |
View Item |
