Öppen för allmänheten
Using replications and prediction markets to estimate the reproducibility of science
There is increasing concern about the fact that many scientific results have been impossible to reproduce. This is problematic, since reproducibility is a central cornerstone of scientific research. Indeed, if a finding cannot be reproduced, there are good reasons to question the finding itself. Factors contributing to a lack of reproducibility include low statistical power, publication bias, and testing of hypotheses with low prior probability of being true.
In a number of projects, Anna Dreber Almenberg (and coauthors) assess the reproducibility of published results by performing large replications of experiments in psychology and economics. In addition, they test whether researchers in the field are able to guess which results can be reproduced and which cannot. This is tested by using direct survey questions, and also by letting the researchers earn money through so-called prediction markets if they are correct.
Föreläsare: Anna Dreber Almenberg is the Johan Björkman professor of economics at the Stockholm School of Economics. Her research in behavioral economics is mainly interdisciplinary on for example variation in economic preferences between and within individuals. In her research, Anna is currently focusing on reproducibility and the use of prediction markets and other tools to predict replication results. Anna¿s research has been published in for example Nature, Science, Proceedings of the National Academy of Sciences, American Economic Review, Psychological Science and Management Science. Anna is also a member of the Young Academy of Sweden and she is a Wallenberg Academy Fellow.
Tid: 15:00 - 17:00
Arrangör: Handelshögskolan vid Göteborgs universitet
Handelshögskolan, Vasagatan 1
Kontaktperson: Marie Örninge
Telefon: 031-786 5607