In nutshell, meta-analysis can be described as a set of statistical methods for aggregating, summarizing, and drawing inferences from collections of thematically related studies. The key idea is to quantify the size, direction, and/or strength of an effect, and to cancel out sampling errors associated with individual studies.
Meta-analysis has become the standard method for aggregating the results from thematically related studies in the health/medical and social/behavioral sciences. Its importance in economics and management science has also increased considerably. Furthermore, some published meta-analyses may have exerted substantial effects on company value (i.e., company stock prices). For instance, following the publication of a meta-analysis on the impact of a specific drug exclusively offered by GlaxoSmithKline (see Nissen & Wolski, 2007), suggesting an increased risk of myocardial infarction due to the intake of the drug ´Rosiglitazone´, GlaxoSmithKline faced an immediate and sharp drop in their stock value.
Given this example, are there perhaps other/similar cases? How large and how homogeneous are these effects? Is there perhaps a systematic relation between certain aspects of the published meta-analytic reports (e.g., topic, author affiliation, publication outlet, etc.) and the the size of the presumed effect?
The overall goal of theses under this thematic umbrella is to (a) find similar cases, (b) summarize them systematically, and (c) to develop and test models explaining the impact of published meta-analyses on company value (i.e., stock prices, for instance).
Candidates are expected to introduce themselves into the foundations of meta-analysis, e.g. using the sources summarized here:
http://www.meta-analysis.eu
The theses will be mentored in collaboration with Prof. Dr. Wolfgang Viechtbauer (U Maastricht, NL).
Literature cited:
Nissen, S. E., & Wolski, K. (2007). Effect of Rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. New England Journal of Medicine, 356, 2457-2471.