Systematic Reviews

A systematic review differs from a narrative review in one respect that matters: every decision is specified in advance and documented, so another researcher could repeat the search and arrive at the same set of studies. That reproducibility is what gives it standing as primary research in its own right.

These guides cover the full protocol — formulating a focused question, building and recording search strings across databases, screening against explicit inclusion criteria, appraising study quality, extracting data systematically, reporting through PRISMA, and extending the synthesis into meta-analysis where the effect sizes allow it.

Most doctoral scholars meet systematic reviews in one of two ways: as the literature chapter of a thesis that a supervisor wants done 'systematically', or as a standalone paper that can be published early in the PhD while the empirical work is still under way. Both uses demand the same discipline. A review that cannot show its search strings, databases, dates and screening numbers will be read as a narrative review, however carefully the studies were chosen.

Before you start, check whether a review on your question already exists or is registered in PROSPERO or on OSF — duplicating a recent review is one of the quickest routes to desk rejection. Registering your own protocol in advance, and following the PRISMA 2020 checklist and flow diagram when you write up, signals to examiners and journal editors that your conclusions rest on a method rather than on the studies you happened to find.