Coordinates  Thursdays from 14 to 16 PM in room 021 of Ludwigstrasse 31. 
Lecturer  Tom Sterkenburg. Contact me at tom.sterkenburglmu.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 longrunning controversies about their foundations. These foundational debates—indeed, "statistics wars"—have only gained more prominence in recent years, for instance in light of the socalled replication crisis. In this course we will cover the main themes in the philosophy of statistical inference, in particular the opposition between the classical and the Bayesian outlook. 
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 reading material is not yet set in stone; depending on participants' interests, we might make some changes as the course progresses.) The first half of the course (the first six 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 overview articles by Romeijn (2014) and Sprenger (2014). In the second half of the course, we first look into more detail at two important specific bones of contention between classical statisticians and Bayesians (meetings 7 and 8). We will then discuss suggestions for some reconciliation between or some pragmatic pickandchoose from the two contenders (meetings 9 and 10). We conclude with a look at broader issues around statistical methodology and practice, namely the replication crisis and the advent of data science (meetings 11 and 12). 
Prerequisites  This is a philosophy course, and our focus will be on conceptual issues. Nevertheless, it will be helpful to have some knowledge of elementary probability and (classical) statistics. We will quickly go through these basics in the first meetings, but this will inevitably be too quick and too little. I would therefore recommend to from the start try and familiarize yourself with (for instance) Wasserman (2004), chapters 1 to 3 and 6. 
Assessment  The course is worth 9 ECTS. Your grade will be determined by a term paper at the end of the course. The term paper treats of a theme we have discussed in the course, and has a length of about 50006000 words. In addition, everyone who is taking the course for credits will be required to give a brief presentation about the readings in one of the meetings. 
Schedule
Date  Topic  Material  Assignment 

Thu 20 Apr  Introduction. Probability and interpretations.  Romeijn (2014), sects. 1 and 2. [Wasserman (2004), chs. 1 to 3.] 

Thu 27 April  Classical statistics: Motivation and methods.  Romeijn (2014), sect. 3.1. Sprenger (2014), sects. 2, 3, and 4 until 4.1. [Wasserman (2004), ch. 6.] 

Thu 4 May  Classical statistics: Challenges.  Romeijn (2014), sect. 3.2. Sprenger (2014), sects. 4 and 5. Schneider (2015).  
Thu 11 May  A better philosophy of classical statistics? Error statistics.  Mayo (2018), sect. 1.I. Mayo, Spanos (2011), sects. 1 and 2. Sprenger (2014), sect. 6.  
Thu 18 May  NO CLASS: Ascension Day.  
Thu 25 May  Bayesian statistics: Motivation and methods.  Romeijn (2014), sect. 4.1. Sprenger (2014), sect. 1. Lindley (2000).  
Thu 1 June  Bayesian statistics: Challenges.  Romeijn (2014), sects. 4.2 and 4.3. Mayo (2018), sect. 6.I. [Efron (1986), Gelman (2008).] 

Thu 8 June  NO CLASS: Corpus Christi.  
Thu 15 June  Bone of contention: The likelihood principle.  Grossman (2011), excluding sects. 6 and 7.27.3. [Mayo (2018), sect. 1.5.]  
Thu 22 June  Bone of contention: Optional stopping.  Sprenger (2014), sect. 7. Grünwald & de Heide (2021).  
Thu 29 June  Reconciliation.  Gelman & Shalizi (2013). [Morey, Romeijn & Rouder (2013).] 

Thu 6 July  Eclecticism.  Senn (2011). [Gigerenzer & Marewski (2015). Box (1983).] 

Thu 13 July  The replication crisis and statistical reform.  Feest (2019). [Romero (2019).]  
Thu 20 July  A new paradigm? Statistics and data science.  Frické (2015). [Kitchin (2014).]  
Fri Sep 22  Deadline term paper. 
Material
 Box (1983). An apology for ecumenism in statistics. Scientific Inference, Data Analysis, and Robustness
 Feest (2019). Why the replication crisis is overrated. Philosophy of Science.
 Frické (2015). Big data and its epistemology. Journal of the Association for Information Science and Technology.
 Gelman & Shalizi (2013). Philosophy and the practice of Bayesian statistics. British Journal of Mathematical and Statistical Psychology.
 Gigerenzer & Marewski (2015). Surrogate science: The idol of a universal method for scientific inference. Journal of Management.
 Grossman (2011). The likelihood principle. Handbook of the Philosophy of Science: Philosophy of Statistics.
 Grünwald & de Heide (2021). Why optional stopping can be a problem for Bayesians. Psychonomic Bulletin & Review.
 Kass (2011). Statistical inference: The big picture. Statistical Science.
 Kass (2021). The two cultures: Statistics and machine learning in science. Observational Studies.
 Lindley (2000). The philosophy of statistics. Journal of the Royal Statistical Society D.
 Mayo (2018). Statistical Inference As Severe Testing: How to Get Beyond the Statistics Wars.
 Mayo, Spanos (2011). Error statistics. Handbook of the Philosophy of Science: Philosophy of Statistics.
 Morey, Romeijn & Rouder (2013). The humble Bayesian: Model checking from a fully Bayesian perspective. Comment on Gelman and Shalizi. British Journal of Mathematical and Statistical Psychology.
 Romero (2019). Philosophy of science and the replication crisis. Philosophy Compass.
 Romeijn (2014). Philosophy of statistics. Stanford Encyclopedia of Philosophy. [link].
 Schneider (2015). Null hypothesis significance tests. A mixup of two different theories: the basis for widespread confusion and numerous misinterpretations. Scientometrics.
 Senn (2011). You may believe you are a Bayesian but you are probably wrong. Rationality, Markets, Morals.
 Sprenger (2014). Bayesianism vs. frequentism in statistical inference. The Oxford Handbook of Probability and Philosophy.
Background material and further reading
 Van Dongen, Sprenger, Wagenmakers (2023). A Bayesian perspective on severity: Risky predictions and specific hypotheses. Psychonomic Bulletin and Review.
 Efron (1986). Why isn't everybody a Bayesian? Journal of the American Statistical Association.
 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.
 Kitchin (2014). Big Data, new epistemologies and paradigm shifts. Big Data & Society.
 Kruschke (2013). Posterior predictive checks can and should be Bayesian: Comment on Gelman and Shalizi. British Journal of Mathematical and Statistical Psychology.
 Mayo, Hand (2023). Statistical significance and its critics: Practicing damaging science, or damaging scientific practice? Synthese.
 Royall (1997). Statistical Evidence: A Likelihood Paradigm.
 Wasserman (2004). All of Statistics.