Welcome to the project

Valid Methods for Meta-Analyses with Few Studies and Small Sample Sizes

The wide field of research on meta-analysis received more and more attention in the last decades. Especially against the background of a large number of false positive results, the aggregation of quantitative heterogeneous findings to one research question has become of great interest.
Different clinical trials with the same issue even with identical starting conditions usually provide varying results. Besides this natural sample variance differences in study populations and the conduct of the studies can be found in practice. A valid interpretation of research results depends therefore on adequate modelling of this heterogeneity. Small sample sizes and insufficient validated research results can be found in many areas of science. This circumstance highlights the necessity of meta-analysis methods that provide valid results even if:

are present.
The outcome of this project will be flexible and valid statistical methods for different meta-analytic models that do not require strong model assumptions and allow for a clear interpretation of the inference results. In order to be able to give recommendations for the correct application of the new methods, these will be thoroughly examined in extensive simulation studies and will be applied to current data sets. The methods will furthermore be made freely accessible in open-source software and provided with easy to understand documentation.

This project will be a joint venture of the Department of Medical Statistics of the University Medical Center Göttingen and the Chair Mathematical Statistics and Applications in Industry at the TU Dortmund University and is funded by the German Research Foundation (DFG).