第81页 Bare-Bones Meta-Analysis: Correcting for Sampling Error Only

啊，看书看得兴致来了手打一段。作者argue为什么要correct for study artifacts。
This section also presents the mathematics of a meta-analysis in which sampling error is the only artifact corrected. Alas, there are those who do not believe that other artifacts exist. There are also those who believe that if there are artifacts in the original study, then those same srtifacts should be reflected in the meta-analysis. They believe that the purpose of meta-analysis is only to describe observed results and that meta-analysis should not correct for known problems in research design. However, most scientists believe that the goal of cumulative research is to produce better answers than can be obtained in isolated studies. From that point of view, the purpose of meta-analysis is to estiamte the relationships that would have been observed if studies had been conducted perfectly, that is, to estimate construct-level relationships that does not correct for as many artifacts as possible is an unfinished meta-analysis. We hold that view. If a meta-analysis corrects only for sampling error, then it is the mathematical equivalent of the ostrich with its head in the sand: It is a pretense that if we ignore other artifacts then their effects on study outcomes will go away. 
每一个做meta-analysis的时候做了artifacts correction的都是理想主义者！