Revealed or Concealed? Transparency of Procedures, Decisions, and Judgment Calls in Meta-Analyses
系统审查1995-2008年间11本顶级期刊的198篇元分析，发现平均仅报告52.8%的复制所需信息，近期和顶级期刊的透明度略高，但整体不足，影响知识积累和实践应用。
The authors examined the degree to which meta-analyses in the organizational sciences transparently report procedures, decisions, and judgment calls by systematically reviewing all (198) meta-analyses published between 1995 and 2008 in 11 top journals that publish meta-analyses in industrial and organizational psychology and organizational behavior. The authors extracted information on 54 features of each meta-analysis. On average, the meta-analyses in the sample provided 52.8% of the information needed to replicate the meta-analysis or to assess its validity and 67.6% of the information considered to be most important according to expert meta-analysts. More recently published meta-analyses exhibited somewhat more transparent reporting practices than older ones did. Overall transparency of reporting (but not reporting of the most important items) was associated with higher ranked journals; transparency was not significantly related to number of citations. The authors discuss the implications of inadequate reporting of meta-analyses for development of cumulative knowledge and effective practice and make suggestions for improving the current state of affairs.