Meta-analysis: what it is and who is doing it right.

ESA's open access journal, Ecosphere, recently ran a nice article on what meta-analysis is, what it isn't and how ecologists use and sometimes abuse the term. Daniela Vetter et al.'s "Meta-analysis: A need for well-defined usage in ecology and conservation biology" is a nice, concise description of what meta-anaysis is, isn't and how ecologists have not yet converged on consistent language to describe studies pulling from multiple data papers. There are lots of systematic reviews parading as meta-analyses, but they lack a couple things, namely accounting for effect sizes. Vetter found that most ecologists do not apply proper meta-analytical techniques in their research, perhaps leading to biased results or misinterpretation of results. 




Hot on the heels of this paper, Christopher Lortie et al. have posted a pre-print over at PeerJ for their manuscript: Practical Interpretation of Ecological Meta-Analyses. Unlike Vetter, who provides criteria and assesses how ecologists have been doing in making inference with meta-analyses, Lortie et al. discusses some cosntraints on how ecological data may make inference difficult. Studies at different scales, studying different suites of environmental filters and taxonomic groups, is not the standard in psychology and the medical sciences, both of which have rigorous and standardized procedures for conducting experiments and meta-analyses. This divergent data is however, common in ecology and Lortie provides some cautionary tales on how to move forward with including studies in meta-analyses and interpreting their results using multiple test statistics.

Fortunately for the scientific community, both of these articles are open access and available at the links above. Rather than have to provide a spoiler alert, I you directly check out Vetter's and Lortie's manuscripts.

Greatest Hits