October 15, Noon – 2 pm ET (Website)
Where do YOU stand?
Given the issues surrounding the misuses and abuse of p-values, do you think p-values should be used?
Do you think the use of estimation and confidence intervals eliminates the need for hypothesis tests?
Bayes Factors – are you for or against?
How should we address the reproducibility crisis?
If you are intrigued by these questions and have an interest in how these questions might be answered – one way of the other – then this is the event for you!
Want to get a sense of the thinking behind the practicality (or not) of various statistical approaches? Interested in hearing both sides of the story – during the same session!?
This event will be held in a debate type of format. The participants will be given selected questions ahead of time so they have a chance to think about their responses, but this is intended to be much less of a presentation and more of a give and take between the debaters.
So – let’s have fun with this! The best way to find out what happens is to register and attend!
Debate Host
Dan Jeske (University of California, Riverside)
Participants
Jim Berger (Duke University)
Deborah Mayo (Virginia Tech)
David Trafimow (New Mexico State University)
Register to Attend this Event Here!
Debate Host: Dan Jeske (University of California, Riverside) Participants: Jim Berger (Duke University), Deborah Mayo (Virginia Tech), David Trafimow (New Mexico State University).
Agenda
About the Participants
Dan Jeske (moderator) received MS and PhD degrees from the Department of Statistics at Iowa State University in 1982 and 1985, respectively. He was a distinguished member of technical staff, and a technical manager at AT&T Bell Laboratories between 1985-2003. Concurrent with those positions, he was a visiting part-time lecturer in the Department of Statistics at Rutgers University. Since 2003, he has been a faculty member in the Department of Statistics at the University of California, Riverside (UCR) serving as Chair of the department 2008-2015. He is currently the Vice Provost of Academic Personnel and the Vice Provost of Administrative Resolution at UCR. He is the Editor-in-Chief of The American Statistician, an elected Fellow of the American Statistical Association, an Elected Member of the International Statistical Institute, and is President-elect of the International Society for Statistics in Business and Industry.. He has published over 100 peer-reviewed journal articles and is a co-inventor on 10 U.S. Patents. He served a 3-year term on the Board of Directors of ASA in 2013-2015.
Jim Berger is the Arts and Sciences Professor of Statistics at Duke University. His current research interests include Bayesian model uncertainty and uncertainty quantification for complex computer models. Berger was president of the Institute of Mathematical Statistics from 1995-1996 and of the International Society for Bayesian Analysis during 2004. He was the founding director of the Statistical and Applied Mathematical Sciences Institute, serving from 2002-2010. He was co-editor of the Annals of Statistics from 1998-2000 and was a founding editor of the Journal on Uncertainty Quantification from 2012-2015. Berger received the COPSS `President’s Award’ in 1985, was the Fisher Lecturer in 2001, the Wald Lecturer of the IMS in 2007, and received the Wilks Award from the ASA in 2015. He was elected as a foreign member of the Spanish Real Academia de Ciencias in 2002, elected to the USA National Academy of Sciences in 2003, was awarded an honorary Doctor of Science degree from Purdue University in 2004, and became an Honorary Professor at East China Normal University in 2011.
Deborah G. Mayo is professor emerita in the Department of Philosophy at Virginia Tech. Her Error and the Growth of Experimental Knowledge won the 1998 Lakatos Prize in philosophy of science. She is a research associate at the London School of Economics: Centre for the Philosophy of Natural and Social Science (CPNSS). She co-edited (with A. Spanos) Error and Inference (2010, CUP). Her most recent book is Statistical Inference as Severe Testing: How to Get Beyond the Statistics Wars (2018, CUP). She founded the Fund for Experimental Reasoning, Reliability and the Objectivity and Rationality of Science (E.R.R.O.R) which sponsored a 2 week summer seminar in Philosophy of Statistics in 2019 for 15 faculty in philosophy, psychology, statistics, law and computer science (co-directed with A. Spanos). She publishes widely in philosophy of science, statistics, and philosophy of experiment. She blogs at errorstatistics.com and phil-stat-wars.com.
Click to access statistical-inference-as-severe-testing_flyer.pdf
David Trafimow is Professor in the Department of Psychology at the New Mexico State University. His research area is social psychology. In particular his research looks at social cognition especially in understanding how self-cognitions are organized, and the interrelations between self-cognitions and presumed determinants of behavior (e.g., attitudes, subjective norms, control beliefs, and behavioral intentions). His research interests include cognitive structures and processes underlying attributions and memory for events and persons. Additionally, he is also involved in methodological, statistical, and philosophical issues pertaining to science.
EVENT TYPE
- NISS Hosted
- NISS Sponsored
Error Statistics Philosophy 
Deborah: I’ve registered. You will do great. Stan
Stan and Pat Young 919 782 2759
I enjoyed the discussion. I must say that David seems confused about the role of modeling and counterfactual reasoning in science. He complained about “appeals to authority” but was quick to say his journal must be on the right track because the impact factor has improved. Is that the appeal to popularity?
Thanks John. It would be popularity if impact factors are determined by numbers. I have no clue. I can well imagine a journal that tells you to hide the weak statistical support behind your inferences, and gives biasing selection effects a free pass would be popular in some circles.
I enjoyed the NISS debate on p-values. Well-done Deborah!
Statistical analyses being based on an assumed model which is likely to be an incorrect model seemed to be David’s major issue. I don’t see it as a major problem. While some researchers report results based on some assumed model, there are often others who are testing the validity of aspects of that model or who seek to replace the model with a better model. Once a model becomes known to be deficient in some respect, the research community tends to move on to a more appropriate model. This activity in the marketplace of ideas seems to work well to advance the scientific enterprise.
I also don’t understand Trafimow’s critique that a hypothesis test only tests a hypothesis H but not a model M (although I suppose he’d be correct if we were just choosing any ol random model M). Model testing is done (normality, homogeneity, independence, etc.) quite regularly, and Mayo has a chapter I believe with input by Spanos on assumption checking and mentions that testing H entails testing H & M.
(so the answer to Trafimow’s critique is more hypothesis testing is needed)
That a model is never true is a poor critique, since we need some rough models to proceed in understanding the world. By rejecting or failing to reject models we move forward. I doubt Trafimow would have any issue using a bernoulli model with p=.5 as a fair coin model, for example, and going by inferences from using that model, even though we all know it isn’t exactly correct in every detail.
Justin
“I doubt Trafimow would have any issue using a bernoulli model with p=.5 as a fair coin model, for example, and going by inferences from using that model, even though we all know it isn’t exactly correct in every detail.”
I wouldn’t be so sure of that. He could express worries about whether the coin was perfectly balanced at the mint, etc., and quibble over unrealistic requirements. Models are among the greatest tools in science, but no one ever said they have a 1 to 1 match to reality. I can only imagine the quagmire that is created by adopting the view that you really cannot use statistics to falsify hypotheses. That will undo all of the progress of the 20th century and turn journals into post-modernist fashion magazines.
John: I’m afraid that if David had his way, that’s what might happen. All this should make people appreciate error statistical methods more, not less. That’s because they only require rather coarse properties–the error probabilities– hold approximately, and conservatively. Moreover, they are key tools for testing assumptions.
Justin and Graham: Yes, I don’t know why David thinks this is some kind of knock down argument against statistical testing. For starters, I think he misunderstands the nature of the implicationary assumptions in testing: they may be used simply to draw out consequences. That is why RCTs work, where the assumption needed for the statistical significance level is vouchsafed by the carrying out of the experimental design itself.
Justin: Can you tell me what’s happened to your blog? Is it back?