Systematic Review vs Scoping Review vs Literature Review

Start with the question, not the label

Most people choose a review type by prestige — "systematic" sounds more rigorous, so they write "systematic review" in the proposal. Then, months later, the method doesn't fit the question and the whole thing has to be relabelled.

Work in the other direction. Write your question first, then read it:

• Does it ask does X work, or which is better, X or Y? → systematic review.
• Does it ask what evidence exists, how is this studied, or where are the gaps? → scoping review.
• Does it ask what do we currently think about this topic, as background to your own study? → narrative literature review.

The narrative literature review

This is the review most students actually write: the chapter or opening section that establishes context, defines the concepts, and argues that a gap exists which your study will fill. It is selective by design. You choose the most relevant and authoritative work rather than everything ever published, and you organise it by theme or debate rather than by study.

There is no mandatory reporting checklist and no registration. That freedom is also its weakness: without a documented search, a reader can't tell whether you found the balance of evidence or just the papers that agreed with you. Even in a narrative review, saying which databases you searched and over what years costs you two sentences and buys a lot of credibility.

The systematic review

A systematic review answers one narrow, pre-specified question using explicit, reproducible methods, so that someone else running your method would reach your result. The machinery around it is what makes it slow:

• A structured question, usually framed with PICO — Population, Intervention, Comparison, Outcome.
• A protocol written and registered before screening starts. PROSPERO, run by the Centre for Reviews and Dissemination at the University of York, registers systematic reviews with a health-related outcome; reviews are no longer eligible once data extraction has begun.
• An exhaustive, documented search across multiple databases, with the full strategy reported so it can be re-run.
• Screening by two people independently, with a third resolving disagreements.
• Risk-of-bias assessment of every included study — this is not optional.
• Reporting against PRISMA 2020, a 27-item checklist plus a flow diagram, which replaced the 2009 statement and should be used in its place.

Budget accordingly. Borah and colleagues analysed 195 completed reviews registered in PROSPERO and found a mean of 67.3 weeks from registration to publication, with a mean of five authors (BMJ Open, 2017). A systematic review is not a solo semester project.

The scoping review

A scoping review is systematic in its methods but broad in its question. It maps a field rather than answering a clinical question. Munn et al. (2018) list six legitimate indications: as a precursor to a systematic review; to identify the types of available evidence in a field; to identify and analyse knowledge gaps; to clarify key concepts or definitions; to examine how research is conducted on a topic; and to identify key characteristics or factors related to a concept.

The methodological differences that follow from that:

• Question framing uses PCC — Population, Concept, Context — instead of PICO. There is no required outcome, because you aren't measuring an effect.
• Broader inclusion. Any study design, plus grey literature, can be in scope.
• Critical appraisal is generally not required, since the point is to show what exists, not to grade it.
• The method comes from Arksey and O'Malley (2005) — five stages (identify the question; identify relevant studies; select studies; chart the data; collate, summarise and report), plus an optional sixth consultation stage with stakeholders, which Levac et al. (2010) argued should be treated as essential. JBI has since built on this framework.
• Reporting follows PRISMA-ScR (Tricco et al., Annals of Internal Medicine, 2018): 20 essential items and 2 optional ones, the optional pair covering critical appraisal if you chose to do it.
• Registration goes elsewhere. PROSPERO does not accept scoping reviews; the Open Science Framework is the usual home for the protocol.

Results are usually charted — a table of study characteristics plus a narrative and numerical summary — rather than pooled. No meta-analysis, and no "the evidence shows treatment X works". A scoping review can conclude that the evidence is thin, fragmented, or dominated by one design, and that is a genuinely useful finding.

Choosing in under a minute

Three checks, in order:

1. Can you name the outcome you want to measure? If yes, and you want to synthesise it across studies, you are doing a systematic review. If the honest answer is "I don't know what's out there yet", you are doing a scoping review.
2. Who is on the team, and for how long? Independent double screening and risk-of-bias assessment need at least two people and, realistically, a year. Alone with three months, a well-executed scoping review or narrative review beats an abandoned systematic one.
3. What will the output be used for? Informing practice or a guideline points to a systematic review. Justifying a PhD topic, framing a grant, or clarifying a messy concept points to a scoping review — or, for a thesis chapter, a narrative one.

A common and perfectly respectable sequence is scoping review first, systematic review second, once the scoping stage has shown that enough comparable studies exist to synthesise.

The failure mode all three share

Whatever label you use, a review is only as trustworthy as its sources. Every included study has to be one you actually retrieved and read — not a plausible-looking reference that arrived from a search-result snippet, a secondary citation, or an AI assistant. Language models are known to generate references that look perfect and do not exist, and in a review, where the reference list is the evidence, one fabricated entry undermines everything around it.

Frequently asked questions

Keep reading

• How to Write a Literature Review Without Fabricating Sources
• How to Search Google Scholar Effectively
• How to Write an Abstract for a Research Paper
• Free Citation Generator (APA, MLA, Chicago, Harvard, IEEE)