Replicability in science
In recent decades, the quality of the published scientific literature in various scientific disciplines has been seriously questioned. To minimise this crisis of confidence, it is crucial to organise science in such a way that scientists can independently assess whether a given phenomenon is supported by evidence. Replication is one of the most obvious tools for verifying findings within science.
Our research deals with challenges related to replication. One such challenge is how to quantify replication success. In other words, how does one determine whether the results of the replication study are consistent with the results of the original study? Although a variety of methods have been proposed and used to quantify replication success, there is not much knowledge about how they function. The first objective of this thesis is to gain knowledge about the functioning of methods that have been used to quantify replication success. A complicating factor when comparing methods for quantifying replication success is publication bias, as it leads to (more) false positive findings in the literature, which in turn leads to lower replication success. Therefore, the second aim of this thesis is to conduct a thorough investigation of publication bias. Publication bias is largely a consequence of dichotomous thinking towards results. The ultimate goal of this thesis is to delve deeper into the conditions under which academics exhibit such dichotomous thinking.