What a Systematic Literature Review Actually Is
A systematic literature review is not just a thorough literature review. A narrative review — the kind most students write in a dissertation’s literature chapter — gathers the sources that seem relevant and builds an argument from them. A systematic review does something stricter: it sets out, in advance, exactly how the literature will be searched, selected and judged, and then follows that plan so completely that another researcher could repeat it and reach the same set of studies. Because every decision is recorded and reported, its conclusions can be trusted in a way a narrative review’s cannot — which is why systematic reviews sit at the top of the evidence hierarchy in medicine, nursing and the social sciences.
If you are writing a standard literature chapter rather than a systematic review, our guide to structuring a dissertation literature review covers that format. This guide covers the systematic version — the planning, the searching, the screening, and the write-up.
Start with a Clear, Answerable Question
Every systematic review starts with a question precise enough to search for. “What works for student anxiety?” is too vague to build a search strategy around. “Do mindfulness-based interventions reduce anxiety in undergraduate students, compared with no intervention or waitlist control?” is a question you can actually search for — it names the population, the intervention, the comparison and the outcome.
That structure has a name: PICO, for Population, Intervention, Comparison and Outcome. It comes from medicine, but it works well beyond it. In education or social science, the “intervention” might be a teaching method or a policy, and the “comparison” might be a different method or none at all — but the discipline is the same. A question built on PICO tells you what to search for, what to include, and what to leave out. If your question is still loose, our guide to writing a strong research question will help you tighten it before you start searching. Write the question down and freeze it before you search: changing the question halfway through is the most common way a systematic review quietly stops being systematic.
Write a Protocol Before You Search
A protocol is a short document, written before the search begins, that records every decision you are going to make: the question, the databases, the search terms, the inclusion and exclusion criteria, how you will screen studies, and how you will assess their quality. It is the single most important document in the whole review, because it is what makes the process repeatable — and what stops you from drifting. For a dissertation, agree it with your supervisor before you run a single search; registering it on PROSPERO (for health-related reviews) adds credibility.
Your inclusion and exclusion criteria deserve the most care. Decide in advance which study designs count, which date range, which languages and which populations — “studies published after 2015, in English, with undergraduate participants, using a randomised or quasi-experimental design” is a criterion a second reviewer can apply; “good studies” is not.
Build a Search Strategy That Finds Everything
The search is where a systematic review earns its name, and where most student reviews fall short. The goal is sensitivity: you would rather screen out a hundred irrelevant papers than miss one relevant one. That means searching several databases — PubMed, Scopus, Web of Science, ERIC or PsycINFO depending on your field — rather than relying on a single source, and it means building the search from keywords and controlled vocabulary together, not just typing your question into a search box.
Work with your university librarian on this step. Librarians build search strategies professionally, they know the subject headings and thesaurus terms for each database, and a thirty-minute meeting with one can save you days of missed studies. Keep a record of every search: the database, the date, the exact search string, and the number of results.
Do not stop at databases. Check the reference lists of the studies you include, search grey literature — theses, conference proceedings, government reports — and, if your field allows it, contact authors for unpublished work. A systematic review that only searches one database is, strictly speaking, not systematic at all.
Screen Studies in Two Stages
Screening is where the numbers collapse. A good search returns hundreds or thousands of records; screening reduces them to the handful that actually answer your question. Do it in two stages: first, screen titles and abstracts against your inclusion criteria; second, read the full text of everything that survives and apply the criteria again, more strictly.
Ideally, screening is done by two people independently, with disagreements resolved by discussion. For a dissertation, that second person is often your supervisor or a fellow student. Even informal double-screening on a sample of records catches the borderline cases you would have decided inconsistently on your own.
Assess Quality and Extract the Data
Once you have your included studies, two jobs remain before you can write anything: assess their quality and extract their data. Quality assessment — often called critical appraisal — asks whether each study’s methods are sound enough to trust its results. Use an established tool for your field: the Cochrane Risk of Bias tool for randomised trials, the CASP checklists for qualitative and observational studies, or the Joanna Briggs Institute tools. The Cochrane Handbook covers these tools in detail in the Cochrane Handbook for Systematic Reviews of Interventions.
Data extraction is the mechanical part: for each study, record the same set of facts — author, year, country, population, intervention, outcomes, key results — into a table. Build the table before you start extracting, pilot it on two or three studies, and then fill it in consistently. A good extraction table becomes the backbone of your results section: it is the evidence your synthesis is built on. Do not skip quality assessment because it feels harsh: a review that treats a weak study and a strong study as equals is misleading, so report the quality of each study and let it shape how much weight you give its findings.
Synthesise — Don’t Just List
The most common weakness in student systematic reviews is a results section that is really just a list: “Smith (2021) found X. Jones (2022) found Y. Patel (2023) found Z.” That is a catalogue, not a synthesis. Synthesis means grouping the studies by what they show, comparing them, and explaining where they agree, where they disagree, and why.
If the studies are similar enough, synthesis can be statistical — a meta-analysis that pools their results into a single estimate. More often in student work, the studies are too different for that, and the synthesis is narrative: organised by theme, by intervention type, or by population, with the studies’ findings woven together rather than reported one by one. Either way, the question to keep asking is: what do these studies, taken together, say about my question? Report the synthesis alongside the PRISMA flow diagram — the standard diagram showing how many records were found, screened, excluded and included. The PRISMA statement, the reporting standard for systematic reviews, sets out exactly what the diagram and the review should contain, and its guidance is available at the PRISMA statement website.
Common Mistakes (and How to Fix Them)
Starting the search before writing the protocol
The temptation is to search first and formalise later. Resist it. A protocol written after the search is a description of what you already did, not a plan — and it cannot protect you from the unconscious drift that follows from seeing the results first.
Listing instead of synthesising
As above: if your results section reads like a series of mini-summaries, stop and regroup the studies by theme or finding. The synthesis is the contribution; the list is just the raw material.
Systematic Review Checklist
Before you submit, run through this quick check:
- Is your question precise enough to search for — population, intervention, comparison, outcome?
- Did you write and date a protocol before searching?
- Did you search more than one database and record every search?
- Can you account for every record in a PRISMA flow diagram?
- Did you record reasons for excluding studies at full-text stage?
- Have you assessed the quality of every included study?
- Does your results section synthesise, rather than list, the findings?
- Would another researcher be able to repeat your search from your methods section?
That last test is the real one. A systematic review is a method, not just a document — and a method that cannot be repeated is not a method at all.
Final Thoughts
A systematic literature review is more work than a narrative review, and it should be: the extra effort is the point. The protocol, the exhaustive search, the recorded screening decisions, the quality assessment — each step exists to make the conclusions trustworthy in a way that a well-written narrative review cannot match. For a dissertation, that trustworthiness is exactly what examiners are looking for.
It is also a skill that travels. The ability to find all the evidence on a question, judge its quality, and synthesise it honestly is useful far beyond the dissertation — in research, in professional practice, and in any job that asks you to make a decision on the basis of evidence. If your dissertation includes a methodology chapter alongside the review, our guide to writing a dissertation methodology chapter covers how to report the process, and our guide to writing a dissertation abstract will help you summarise the finished review in 300 words.
If you are working to a tight deadline and your systematic review still needs a polished write-up, our team can help. We write and edit essays, dissertations and research papers for UK, US and Australian students, and every order includes free revisions. See our services for the full list, message us on WhatsApp at +49 1521 3962850 for a quick quote, or send your brief through the contact form and we will get back to you the same day.