Systematic review and meta analysis

Boyle, James and Connolly, Michael and MacKay, Tommy (2016) Systematic review and meta analysis. Educational and Child Psychology, 33 (3). ISSN 0267-1611

[thumbnail of Boyle-Connolly-MacKay-ECP2016-systematic-review-and-meta-analysis]
Text. Filename: Boyle_Connolly_MacKay_ECP2016_systematic_review_and_meta_analysis.pdf
Accepted Author Manuscript

Download (816kB)| Preview


Aim: This paper provides an overview of the research methodologies of systematic review and meta-analysis and uses a commentary on the analysis of data from a previously published study to illustrate the procedures and decision-making involved for consumers and those who may be considering carrying out a systematic review. Rationale: Systematic review and meta-analysis are located within a hierarchy of evidence-based practice, and their underlying epistemological and theoretical basis considered. The advantages of systematic review over traditional narrative reviews are discussed, together with the case for the use of meta-analysis to synthesise research findings. The feasibility of the use of these methodologies by educational psychologists is also considered. Findings: The worked example details the steps necessary to carry out a systematic review and meta-analysis and the commentary addresses key issues such as specifying inclusion/exclusion criteria and determining relevance; specifying the literature search strategy and coping with the ‘grey’ literature; extracting and coding data from the included studies and the importance of reliability checks; study quality; selecting the most appropriate effect size; selecting the most appropriate model for meta-analysis (fixed-effect versus random-effect), combining and averaging effect sizes across studies; running weighted ANOVAs or meta-regression analyses to investigate heterogeneity; checks for publication bias; and sensitivity analysis to deal with outliers. Conclusions: Future developments in these methodologies, details of available software and resources, and implications for educational psychologists who may wish to carry out systematic reviews and meta-analysis are discussed.


Boyle, James ORCID logoORCID:, Connolly, Michael ORCID logoORCID: and MacKay, Tommy;