A systematic review starts with a question. It sets out which studies qualify, searches for and selects them, and weighs their risk of bias before any numbers are combined. A meta-analysis is the statistical step that may follow: the combination of results from two or more separate studies (Deeks et al., 2024).
Meta-analysis took shape in the 1970s in the social sciences, in the work of Gene Glass, Robert Rosenthal, and Frank Schmidt (Shadish & Lecy, 2015). Glass laid out the difference between primary, secondary, and meta-analysis of research in 1976 (Glass, 1976).
How to read one
Combining studies can sharpen an estimate and settle arguments between conflicting trials. It can also mislead when the studies differ in design, carry their own biases, or vary widely in what they did. That variation is called heterogeneity, and a careful meta-analysis reports it and looks for what explains it (Deeks et al., 2024).
In light research, studies often differ in wavelength, power, and timing, and poorly chosen settings have produced negative studies alongside positive ones (Huang et al., 2009). That is one reason two reviews of the same question can disagree.
References
Deeks, J. J., Higgins, J. P. T., Altman, D. G., McKenzie, J. E., & Veroniki, A. A. (2024). Chapter 10: Analysing data and undertaking meta-analyses. In J. P. T. Higgins, J. Thomas, J. Chandler, M. Cumpston, T. Li, M. J. Page, & V. A. Welch (Eds.), Cochrane handbook for systematic reviews of interventions (Version 6.5). Cochrane. https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-10
Glass, G. V. (1976). Primary, secondary, and meta-analysis of research. Educational Researcher, 5(10), 3–8. https://doi.org/10.3102/0013189X005010003
Huang, Y.-Y., Chen, A. C.-H., Carroll, J. D., & Hamblin, M. R. (2009). Biphasic dose response in low level light therapy. Dose-Response, 7(4), 358–383. https://doi.org/10.2203/dose-response.09-027.Hamblin
Shadish, W. R., & Lecy, J. D. (2015). The meta-analytic big bang. Research Synthesis Methods, 6(3), 246–264. https://doi.org/10.1002/jrsm.1132