Mini Meta‐Analysis of Your Own Studies: Some Arguments on Why and a Primer on How

TLDR

Mini meta‑analyses within manuscripts are uncommon, largely because researchers doubt their legitimacy for few studies and perceive them as too complex without expert guidance. The paper aims to encourage and guide researchers to perform mini meta‑analyses within their own manuscripts by debunking misconceptions and providing practical instructions. The authors supply effect‑size formulas, metric conversions, annotated Excel spreadsheets, and a step‑by‑step guide to conduct a simple meta‑analysis on few studies. Mini meta‑analyses strengthen and clarify related studies when included in a manuscript.

Abstract

Abstract We outline the need to, and provide a guide on how to, conduct a meta‐analysis on one's own studies within a manuscript. Although conducting a “mini meta” within one's manuscript has been argued for in the past, this practice is still relatively rare and adoption is slow. We believe two deterrents are responsible. First, researchers may not think that it is legitimate to do a meta‐analysis on a small number of studies. Second, researchers may think a meta‐analysis is too complicated to do without expert knowledge or guidance. We dispel these two misconceptions by (1) offering arguments on why researchers should be encouraged to do mini metas, (2) citing previous articles that have conducted such analyses to good effect, and (3) providing a user‐friendly guide on calculating some meta‐analytic procedures that are appropriate when there are only a few studies. We provide formulas for calculating effect sizes and converting effect sizes from one metric to another (e.g., from Cohen's d to r ), as well as annotated Excel spreadsheets and a step‐by‐step guide on how to conduct a simple meta‐analysis. A series of related studies can be strengthened and better understood if accompanied by a mini meta‐analysis.