| Coordinates | Thursdays from 10 to 12 PM in room 028 of Ludwigstrasse 31. |
| Lecturer | Tom Sterkenburg. Contact me at tom.sterkenburg lmu.de; visit me in room 126 of Ludwigstrasse 31. |
| Course description | Despite the central role of statistical methods in many branches of science, there are various long-running controversies about their foundations. These foundational debates—sometimes indeed referred to as "statistics wars"—have only gained more prominence in recent years, prompted by the replication crisis and calls for statistical reform. In this seminar we will cover some main themes in the philosophy of statistical inference. We will discuss literature about the philosophical foundations of statistical methods, particularly the opposition between the classical and the Bayesian schools. Moreover, we will discuss literature from the sciences and the philosophy of science about the replication crisis. |
| Contents and material | We will read and discuss a number of texts on the philosophy of statistics. See the below schedule and material for details. The references between square brackets are optional background reading. The first four meetings will be more expository, introducing the classical (frequentist) and the Bayesian paradigm and their (perceived) strenghts and weaknesses. Here we rely for good part on the book by Otsuka (2023) and the overview article by Sprenger (2014). The next two meetings we read two papers that in different ways explore the world beyond the narrow Bayesianism v. frequentism "wars." The final part of the seminar will focus on the replication crisis. Here we will discuss a number of recent papers in the philosophy of science dealing with this topic. |
| Prerequisites | This is a philosophy seminar, and our focus will be on conceptual issues. Nevertheless, it will only be helpful to have some knowledge of elementary probability and (classical) statistics. Chapter 1 of the Otsuka book provides the essentials, but if the field is new to you it could be profitable to browse through some more material, like Wasserman (2004), chapters 1 to 3 and 6. |
| Assessment | All students are requested to attend the seminar sessions, carefully study the readings, and participate in the discussions. To receive a grade for the seminar, you are required to complete a presentation and a writing assignment. There are two distinct modes of assessment, depending on whether you hail from the relAI or the LMU Philosophy program. Presentation (all students). You are required to give one 20-minute presentation on one of the readings. Further, you will be paired with a fellow student to act as reviewer/moderator for their presentation, which means you will initiate and moderate (together with me) the discussion after their presentation. Writing (relAI). Writing assignment on the topic of your presentation of approximately 3,000 words (more details will follow). Deadline t.b.a. Writing (Philosophy). Term paper of approximately 6,000-7,000 words, on a topic related to the seminar but of your own choosing. Deadline t.b.a. Final grade (relAI, 6 ECTS). Presentation & participation (30%); writing assignment (70%). Final grade (Philosophy, 9 ECTS). Term paper (100%). |
Schedule
| Date | Topic | Material | Assignment |
|---|---|---|---|
| Thu 24 Apr | Introduction. Probability and interpretations. | Otsuka (2023), ch. 1. [Romeijn (2014), sects. 1-2.] | |
| Thu 1 May | NO CLASS: Labour day. | ||
| Thu 8 May | Bayesian statistics. | Otsuka (2023), ch. 2; Sprenger (2014), sect. 18.1. [Romeijn (2014), sect. 4.] | |
| Thu 15 May | Classical statistics: Motivation and methods. | Otsuka (2023), ch. 3; Sprenger (2014), sects. 2-4 (until "p-values and posterior probabilities"). [Romeijn (2014), sect. 3.1.] | |
| Thu 22 May | Classical statistics: Challenges. | Sprenger (2014), sects. 4-5, 7-8; Schneider (2015).[Romeijn (2014), sect. 3.2.] | |
| Thu 29 May | NO CLASS: Ascension Day. | ||
| Thu 5 June | Beyond the statistics wars? | Gigerenzer & Marewski (2015). | |
| Thu 12 June | Beyond the statistics wars? | Van Dongen et al. (2019). | |
| Thu 19 June | NO CLASS: Corpus Christi. | ||
| Thu 26 June | The replication crisis. | Romero (2019). | |
| Thu 3 July | The replication crisis. | Bird (2021). | |
| Thu 10 July | The replication crisis. | Lavelle (2023). | |
| Thu 17 July | The replication crisis. | Fletcher (2022). | |
| Thu 24 July | The replication crisis. | Feest (2023). | |
| t.b.a. | Deadline writing assignment. |
Material
Readings, first part.
- Van Dongen et al. (2019). Multiple perspectives on inference for two simple statistical scenarios. The American Statistician.
- Gigerenzer & Marewski (2015). Surrogate science: The idol of a universal method for scientific inference. Journal of Management.
- Otsuka (2023). Thinking About Statistics: The Philosophical Foundations.
- Schneider (2015). Null hypothesis significance tests. A mix-up of two different theories: the basis for widespread confusion and numerous misinterpretations. Scientometrics.
- Sprenger (2014). Bayesianism vs. frequentism in statistical inference. The Oxford Handbook of Probability and Philosophy.
Readings, second part.
- Bird (2021). Understanding the replication crisis as a base rate fallacy. British Journal for the Philosophy of Science.
- Feest (2023). What is the replication crisis a crisis of? Philosophy of Science.
- Fletcher (2022). Replication is for meta-analysis. Philosophy of Science.
- Lavelle (2023). When a crisis becomes an opportunity: The role of replications in making better theories. British Journal for the Philosophy of Science.
- Romero (2019). Philosophy of science and the replication crisis. Philosophy Compass.
Background material and further reading
- Efron (1986). Why isn't everybody a Bayesian? Journal of the American Statistical Association.
- Gelman & Shalizi (2013). Philosophy and the practice of Bayesian statistics. British Journal of Mathematical and Statistical Psychology.
- Gelman (2008). Objections to Bayesian statistics. Bayesian Analysis.
- Gelman, Hennig (2017). Beyond subjective and objective in statistics. Journal of the Royal Statistical Society Series A.
- Hájek & Hitchcock (2014). Probability for everyone—even philosophers. The Oxford Handbook of Probability and Philosophy.
- Kass (2011). Statistical inference: The big picture. Statistical Science.
- Mayo (2018). Statistical Inference As Severe Testing: How to Get Beyond the Statistics Wars.
- Romeijn (2014). Philosophy of statistics. Stanford Encyclopedia of Philosophy. [link].
- Royall (1997). Statistical Evidence: A Likelihood Paradigm.
- Senn (2011). You may believe you are a Bayesian but you are probably wrong. Rationality, Markets, Morals.
- Wasserman (2004). All of Statistics.