A literature review is the chapter examiners read first — and the one they judge the rest of your dissertation against. A weak review signals that you have not engaged with your field; a strong review proves you can think critically about what is known, what is contested, and what is missing. This 2026 guide walks you through every step of writing a literature review that does the second.
What Is a Literature Review?
A literature review is a critical synthesis of the scholarly work relevant to your research question. It is not a summary of what each paper said — it is an argument about what the field collectively shows, where the gaps are, and why your study is needed.
Three things separate a strong review from a weak one: (1) it argues rather than summarises, (2) it identifies a clear gap that your study fills, and (3) it builds the theoretical framework your methodology and discussion will rely on.
Types: Narrative, Systematic, Scoping, Integrative
Picking the wrong review type is one of the most expensive mistakes a candidate can make. The four most common types each suit different research questions:
Narrative review — critically synthesises mature literatures around a theoretical argument. Common in humanities and qualitative PhDs. No formal protocol required.
Systematic review — uses a pre-registered protocol, exhaustive search, formal screening (usually two reviewers), and quality appraisal. Suits clearly-bounded clinical or social-science questions. PRISMA 2020 reporting standard.
Scoping review — maps the breadth of an emerging literature to identify research gaps. PRISMA-ScR reporting standard.
Integrative review — combines empirical and theoretical work to produce a new conceptual framework.
If you are unsure which type your question needs, our [literature review writing service](/services/literature-review-writing-service) starts with a free 30-minute scoping call.
Building a Defensible Search Strategy
Examiners scrutinise the search strategy because it determines what your review could possibly have found. A defensible search has four elements:
1. Database selection. Cover the major databases in your field — typically 3–6. Health: PubMed, CINAHL, PsycINFO, Cochrane Library. Education: ERIC, Education Source, British Education Index. Business: Business Source Premier, EconLit, ABI/INFORM. Engineering: Scopus, Web of Science, IEEE Xplore.
2. Boolean search strings. Combine subject headings (MeSH for PubMed, thesaurus terms for ERIC) with free-text keywords using AND / OR / NOT. Document every string and every database.
3. Inclusion / exclusion criteria. State these before screening — date range, language, study design, population. Apply them consistently.
4. Grey literature. For policy-relevant or applied research, search OpenGrey, Google Scholar (first 200 results), and key organisational websites. Pre-registered protocols should specify this in advance.
Screening, Quality Appraisal & Extraction
For systematic and scoping reviews, screening happens in two stages: title/abstract, then full text. Both stages should ideally be done by two reviewers independently with a third to resolve disagreement. Record every decision in a screening log.
Quality appraisal tools vary by study type: CASP (qualitative), JBI critical appraisal (mixed designs), Cochrane Risk of Bias 2 (RCTs), AMSTAR-2 (existing reviews). Extract findings into a structured table.
Generate a [PRISMA 2020 flow diagram](/tools/prisma-flow) using our free tool — your examiner will expect to see one.
Synthesis: Beyond Summarising
This is where most reviews fail. A summary lists what 50 papers said. A synthesis identifies what those papers collectively show. Four established synthesis approaches:
Thematic synthesis (Thomas & Harden) — code findings line-by-line, develop descriptive themes, derive analytical themes. Most common in qualitative meta-synthesis.
Framework synthesis — code findings against an a priori framework (e.g. a theoretical model). Useful when an established framework already exists.
Meta-ethnography (Noblit & Hare) — translate concepts across studies to build a higher-order interpretation. Demanding but powerful for qualitative PhDs.
Narrative synthesis — group findings by sub-topic, identify patterns, explore relationships textually. The standard for mixed-design reviews.
How to Structure the Chapter
A typical literature review chapter follows this structure: introduction (1 page), methods (search strategy + screening), thematic findings (organised by sub-theme, not by paper), critical synthesis (what the field collectively shows and where it disagrees), theoretical framework (if relevant), gap identification (explicit), and conclusion that motivates your study.
Master's reviews are typically 4,000–8,000 words; PhD narrative reviews 12,000–25,000 words; standalone systematic reviews 6,000–10,000 words depending on the journal.
10 Common Mistakes Examiners Flag
1. Summarising rather than synthesising. 2. No clear search strategy. 3. Missing seminal works. 4. Over-reliance on textbooks. 5. Citing 2014 sources when 2023 systematic reviews supersede them. 6. No critical engagement with methodological quality. 7. Theoretical framework parachuted in unrelated to the review. 8. Gap claim that is not actually a gap. 9. Inconsistent citation style. 10. No explicit link between the review and the research question.
APA Literature Review Example
See our companion guide [APA Literature Review Example](/blog/apa-literature-review-example) for a full annotated sample showing structure, in-text citation, and reference list formatting.
If you would prefer expert PhD support building your literature review, our [literature review writing service](/services/literature-review-writing-service) covers every review type — from a Master's narrative review to a Cochrane-style meta-analysis.