Skip to content
Start for free

Systematic or narrative review: which one does your assignment need?

A systematic review answers one focused question with a search and selection method planned in advance, so someone else could repeat it; a narrative review explains and interprets a topic without a fixed protocol. If your brief mentions a protocol, PRISMA or a documented search strategy, you need a systematic approach; if it asks you to review the literature on a topic, a well-organised narrative review is usually what is expected.

By Aithor.io team ·

Key takeaways

  • A systematic review follows a protocol that someone else could repeat and reach the same studies.
  • A narrative review interprets a topic and its debates; its strength is argument, not completeness.
  • The words in your brief decide which one you need, not the length of the assignment.
  • A narrative review with systematic habits (a logged search and clear inclusion criteria) suits most dissertations.

The difference in one table

Both are literature reviews: they bring together what has already been published. The difference is in how the studies are found, chosen and combined.

Systematic review Narrative review
Starting point One focused question, often framed with PICO A topic, a debate or a broad question
Plan A protocol written before searching No fixed protocol; the plan can evolve
Search Several databases, with the full search strings reported Whatever finds the key work: databases, citation chains, books
Selection Inclusion and exclusion criteria applied to every result, usually by two people The author chooses what is relevant and should explain why
Quality appraisal Formal assessment of each study’s risk of bias Critical discussion, usually less formal
Synthesis Structured, sometimes statistical (meta-analysis) Interpretive: themes, arguments, history of ideas
Reporting PRISMA checklist and flow diagram Clear structure; no standard checklist required
Can someone repeat it? Yes, that is the point Not exactly, and it doesn’t claim to
Typical time Months for a published review Weeks for a dissertation chapter

There are many more review types than these two. One well-known typology describes 14, each defined by how it searches, appraises, synthesises and analyses [2]. For a student, though, the systematic and narrative ends of the spectrum are the two that matter most, with the scoping review in between.

What a systematic review involves

A systematic review tries to find all the studies that answer a specific question, so that the conclusion doesn’t depend on which papers the author happened to know. That aim drives every step:

  • A protocol before you start. The question, the databases, the search terms, the inclusion criteria and the analysis are written down first, so you can’t adjust them after seeing the results. Published reviews often register their protocol.
  • A reproducible search. Several databases, each with its full search string and the date it was run.
  • Screening against criteria. Every result is checked against the inclusion and exclusion criteria, first by title and abstract, then in full text. In published reviews two people do this independently.
  • Appraisal and synthesis. Each included study is assessed for risk of bias, and the results are combined, in a table, a narrative synthesis or a meta-analysis.
  • Transparent reporting. The PRISMA 2020 statement sets out a 27-item checklist and a flow diagram that shows how many records were found, screened, excluded and included [1].

The Cochrane Handbook is the standard reference for the method when the question is about the effect of an intervention [3]. If your brief uses any of these words (protocol, PRISMA, risk of bias, search strategy), your marker expects this approach, even if on a smaller scale.

What a narrative review involves

A narrative review sets out what is known about a topic and makes sense of it. It might trace how ideas developed, compare competing explanations, identify a problem the field hasn’t solved or build a case for a new study. A classic paper on writing them lists aims like these: developing or evaluating theory, surveying the state of knowledge, identifying problems and giving a historical account [4].

Freedom doesn’t mean anything goes. The same paper warns against common failings: covering the evidence unevenly, summarising studies one after another without integrating them and drawing conclusions the evidence doesn’t support. A quality scale for narrative reviews, SANRA, checks six things: why the review matters, whether its aims are stated, whether the literature search is described, the referencing, the scientific reasoning and the presentation of evidence [5]. Notice the third one: even a narrative review should say how it found its sources.

Where the scoping review fits

A scoping review sits between the two. It searches systematically but asks a broader question, such as what kinds of evidence exist, how a concept is defined or where the gaps are, and it usually doesn’t appraise each study in depth. Guidance comparing the two suggests a systematic review when the question is about effectiveness or appropriateness, and a scoping review when you need to map a field or clarify concepts first [6]. For a dissertation on a new or messy topic, it can be the most honest choice.

Decision table: which one does your assignment need?

Read your brief and the marking criteria, then find the line that matches best.

If your brief… You probably need… What to check with your supervisor
Mentions a protocol, PRISMA, a search strategy or risk of bias A systematic review (possibly a reduced one) How many databases, whether one reviewer is acceptable
Asks “what is the evidence that X works?” for a defined group A systematic review Whether a meta-analysis is expected or optional
Asks what research exists on a broad or new topic A scoping review Which framework to report against
Asks you to “review the literature” for an empirical dissertation A narrative review with a documented search How much of the search to report in the method
Asks you to critically discuss theories or debates A narrative review How many sources are expected
Gives no detail at all A narrative review with systematic habits Ask before you start, not after

The safest middle ground for most dissertations is the last line: a narrative review that borrows systematic habits. Log where and when you searched, write down your inclusion criteria, and keep a table of the studies you read. You don’t claim to be exhaustive, but your marker can see how you worked.

Worked example: one topic, two reviews

Take one topic, phone use and sleep among university students, and see how the review changes.

As a systematic review, the question has to be narrow enough to answer from studies: Among undergraduate students, is smartphone use in the hour before bed associated with shorter sleep duration? Example inclusion criteria: studies of undergraduates; a measure of pre-bedtime smartphone use; sleep duration measured by diary, questionnaire or device; published in a peer-reviewed journal; any year; English or Spanish. Exclusion criteria: studies of school pupils, studies of general screen time that don’t separate smartphones, and opinion pieces. The search runs in three databases with the same string, and every record is screened against those criteria.

As a narrative review, the question can be wider and more interpretive: How have researchers explained the link between students’ phone use and their sleep, and what is still disputed? You might organise it by explanation (displaced sleep time, light exposure, arousal from content) rather than study by study, and you would choose the studies that best represent each line of argument, saying how you found them.

Same topic, two different chapters. Neither is better in general; one matches your brief.

Common mistakes

  • Calling a review “systematic” because the search was careful. Without a protocol and criteria applied to every result, it is a narrative review with a good search. Call it that.
  • Claiming PRISMA without following it. If you include a flow diagram, the numbers in it have to come from a real screening process.
  • Searching only one general search engine. It is fine for exploring, not for a review that claims completeness. Our guide to using Google Scholar for your dissertation explains why.
  • Summarising papers one by one. Both types need synthesis: what the studies say together, where they disagree and why.
  • Picking the type after writing. The type decides how you search, so choose it first.

Where Aithor.io fits

The Literature review tool searches and screens published studies, then drafts the synthesis for you to check. What it produces: Includes inclusion criteria, a table of studies and a comparison matrix. That structure fits the “narrative review with systematic habits” approach well. Free: 1 literature review, only the introduction visible and no downloads.

Frequently asked questions

Can an undergraduate dissertation be a systematic review?

Yes, and some departments offer it as a format, especially in health subjects. Expect a smaller scope than a published review: one focused question, a few databases and often a single reviewer, which you should state as a limitation.

Is a narrative review less scientific?

It answers a different kind of question. A good narrative review is explicit about how it found its sources and why it included them, and it is judged on the quality of its reasoning. Tools such as SANRA exist to assess exactly that.

Do I need PRISMA for a narrative review?

No. PRISMA is a reporting guideline for systematic reviews. You can still borrow its habits, such as reporting where and when you searched, without claiming to follow it.

What is a scoping review, then?

A review that maps what research exists on a broad topic, with systematic searching but without judging the quality of each study in depth. It is a good choice when your question is “what has been studied?” rather than “what works?”.

Sources

  1. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ (2021). doi.org/10.1136/bmj.n71
  2. A typology of reviews: an analysis of 14 review types and associated methodologies. Health Information and Libraries Journal (2009). doi.org/10.1111/j.1471-1842.2009.00848.x
  3. Cochrane Handbook for Systematic Reviews of Interventions. Cochrane. www.cochrane.org/authors/handbooks-and-manuals/handbook/current
  4. Writing narrative literature reviews. Review of General Psychology (1997). doi.org/10.1037/1089-2680.1.3.311
  5. SANRA, a scale for the quality assessment of narrative review articles. Research Integrity and Peer Review (2019). doi.org/10.1186/s41073-019-0064-8
  6. Systematic review or scoping review? Guidance for authors when choosing between a systematic or scoping review approach. BMC Medical Research Methodology (2018). doi.org/10.1186/s12874-018-0611-x

Written by Aithor.io team · Editorial review: Aithor.io team (reviewed on )

Published on

How this article was made: AI-assisted draft, edited by the team · Editorial review: Aithor.io team.