The Institute for Mathematical Behavioral Sciences Colloquium Series presents
“A Bayesian Perspective on the “Crisis of Confidence” in Psychological Science”
with Eric-Jan Wagenmakers, Department of Psychology, University of Amsterdam
Thursday, January 16, 2014
Social Science Plaza A, Room 2112
The “crisis of confidence” refers to the recent realization that many findings from psychological science may not be replicable. This presumed lack of replicability is caused by a combination of factors such as hindsight bias and the incentive structure of academia (“publish or perish”). This presentation will focus on two factors that contribute to the crisis of confidence but have received rather scant attention thus far; the first factor refers to the fact that “the weight of evidence for a claim should be proportioned to its strangeness” (i.e., extraordinary claims require extraordinary evidence) and the second factor refers to the fact that “p values are violently biased against the null”. The importance of these factors will be demonstrated by means of concrete examples from the literature.
For further information, please contact Joanna Kerner, email@example.com or 949-824-8651.