[Colloquium] Seminar Announcement: False-positives and inflated effects: a meta-research view

[Ninfa Mayorga]

~Reminder~

Computation Institute-Computational Intelligence

Speaker: John P.A. Ioannidis, MD, DSc, C.F. Rehnborg Chair in Disease Prevention, Director, Stanford Prevention Research Center
Host: James Evans 
Date: April 26, 2012
Time: 12:00 PM - 1:30 PM
Location: University of Chicago, Searle 240A/B, 5735 S. Ellis Avenue

False-positives and inflated effects: a meta-research view

Abstract:
Many fields of scientific investigation are plagued by very high rates of false-positive claims of discoveries and/or inflated, exaggerated effects for discoveries that are not entirely false. There are several reasons for this pattern, including both mathematical expectations for new discoveries (the winner's curse phenomenon) and an array of selective reporting and other biases. Meta-research approaches can offer insights about these diverse mechanisms by examining large numbers of datasets and studies in different fields of scientific inquiry, where there is large-scale evidence on their replication patterns. Examples will be drawn during the talk from diverse fields of biomedical research (ranging from genetics, prognostic studies, clinical trials, neurosciences, and traditional epidemiology), and other fields. The talk will also discuss potential solutions for improving the reproducibility and credibility of research findings that stem from this evidence. 

John P.A. Ioannidis, MD, DSc
C.F. Rehnborg Chair in Disease Prevention
Director, Stanford Prevention Research Center
Professor of Medicine, Health Research and Policy, and Statistics
Stanford University


Information:  Those interested in meeting with Professor Ioannidis can contact James Evans (jevans at uchicago.edu).

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20120426/2b3b042e/attachment.htm