• Synthesis

• Feedback Cycles

• Shared responsibility and sustainable cooperation

Meta-analyses have increasingly become a prominent way of summarizing and integrating scientific studies. In many cases, meta-analyses are seen as constituting the highest standard of scientific evidence. They are also considered to be more objective compared to other ways of integrating scientific evidence, such as narrative reviews. In this chapter I argue that the statistical analyses which characterize meta-analyses are highly flexible and provide the authors a large degree of influence over the results. I argue that this is the case even if other aspects of the meta-analysis -- such as inclusion criteria and the coding of the literature -- are held constant. Furthermore, I argue that there are reasons to expect that the authors of meta-analyses may be biased. Given the flexibilities present even implicit bias may strongly influence the results.

To what extent does a choice in statistical analysis influence the results of a meta-analysis?

• -

-

—

—