James Coleman Symposium

When:
April 13, 2018 @ 1:00 pm – 5:00 pm
2018-04-13T13:00:00+02:00
2018-04-13T17:00:00+02:00
Where:
Utrecht University Hall - Belle van Zuylenzaal
Domplein 29
3512 JE
Utrecht

The symposium will take place in the Academy building in Utrecht, in the “Belle van Zuylenzaal” and is titled:

Is that true? Replication and Meta-analysis in the social sciences

Science has currently been confronted with a so-called ‘replication crisis’, meaning that scientists have found that the results of many scientific studies cannot be replicated upon subsequent investigation. This crisis not only concerns the social sciences, but many other disciplines such as economics, biology, and medicine. It raises fundamental issues on the importance of replications, the generalizability of findings and theories, study design, the use of statistics to combine and interpret results, and the publication culture of empirical findings.

Five speakers address these fundamental issues raised by the replication crisis. Daniel Lakens of Eindhoven University of Technology and member of the programme committee of the NWO Grant Replication Studies, will address the importance of replications and the difficulty of combining and interpreting findings of original and replication studies. Jelte Wicherts of the meta-research center at Tilburg University will discuss how significant and non-significant replication findings should affect our belief that a hypothesis is true, and how these findings actually affect researchers’ beliefs of this hypothesis being true. Eric-Jan Wagenmakers of the University of Amsterdam and founder and director of the free statistical software JASP will tell about his experiences with pre-registered replication studies and how to interpret their findings. Fred Hasselman of Radboud University Nijmegen and co-author of several papers on ManyLab projects will discuss the implications of these projects for the generalizability of findings and the heterogeneity or contextual dependence of effects. Marcel van Assen of Utrecht University and the meta-research center at Tilburg University will demonstrate that the optimal design of replication studies, but of original studies as well, includes heterogeneous contexts rather than a few experimentally controlled conditions.

The program starts at 13:00 (Belle van Zuylenzaal Utrecht) and ends with drinks at 17:00. Please register via Saskia Simon (s.simon@rug.nl).