Literature review

You will turn a pile of sources into a review organised by theme or debate that ends in a clearly stated gap your study fills.

Do this first

Build a synthesis matrix for your ten most important sources: one row per source, one column per theme or question, plus columns for method and limitation. Fill it in from your notes, then read down each column and write one sentence on what the sources say together.

Steps

0 of 6 steps done

Define the review's job
Write your research question at the top of the page. A literature review is not a record of everything you read: it shows what is known, disputed and missing in relation to that question. Sources that do not help answer it go in a separate file, not the chapter.
Build a synthesis matrix
List sources as rows and themes, concepts, methods or findings as columns. Filling the cells forces you to compare sources on the same point, which is what synthesis is. Empty cells and clashing entries show you where the debates and gaps are.
Organise by theme or debate, not by author
Treat each column of your matrix as a candidate section or paragraph. Order sections so they build towards your question, for example from settled consensus to the narrow unresolved problem. Use chronology only when change over time is the point.
Open paragraphs with claims about the field
Start each paragraph with a sentence about the literature, not about one author: “Studies using panel data consistently find…” rather than “Smith (2019) found…”. Then group sources that agree, show where others differ, and explain why (method, sample, context or theory). Close by saying what this means for your question.
Evaluate as you go
Tell the reader how much weight the evidence can bear: designs, sample sizes, settings and theoretical assumptions. Phrases such as “largely based on small samples” or “rarely tested outside” do the evaluating for you and set up the gap.
State the gap and your contribution
End the review, or each major section, with a precise statement of what is missing and how your study addresses it. A good gap is specific: a population, setting, method, period or theoretical question that existing work has not covered or resolved. “Little research exists” is weak unless you show why the absence matters.

Templates

Synthesis matrix

Research question: [your question]

Columns: Source | Theme 1: [name] | Theme 2: [name] | Theme 3: [name] | Method and sample | Limitation

Row: [Author, year] | [what it says] | [what it says] | [leave blank if silent] | [design, n, setting] | [weakness]
Row: [Author, year] | [what it says] | [what it says] | [what it says] | [design, n, setting] | [weakness]

Reading down each column:
Theme 1: The sources agree that [point]. They differ on [point], probably because [method, sample, context or theory].
Theme 2: [same pattern]

Synthesis paragraph

[Claim about the field]. Several studies support this: [Source A] found [result], and [Source B] and [Source C] reported similar [results] in [contexts]. However, [Source D] found [contrary result], possibly because [difference in method, sample, context or theory]. Taken together, these studies suggest [your interpretation], although [limitation of the evidence]. This matters for the present study because [link to your question].

Gap statement

Although [what is established] ([Author, year]; [Author, year]), [specific issue] remains [untested / unclear / contested] because [existing studies focus on X / rely on Y / overlook Z]. This matters because [consequence for theory, policy or practice]. This study addresses the gap by [what you do], using [data, sources or approach].

Check your draft

0 of 8 done

Does each body paragraph open with a claim about the literature rather than an author's name?
Do most paragraphs cite and compare more than one source?
Can I name the theme or debate each section addresses in a few words?
Have I explained why studies disagree, not only that they do?
Have I evaluated the strength of the evidence, not only reported it?
Is the gap stated in one or two specific sentences that lead directly to my research question?
Have I cut sources that do not bear on my question?
Is my own voice present, telling the reader what the sources add up to?

In your field

Reviews are usually short and folded into the introduction, moving quickly from established findings to a specific technical gap. Citation is often numbered (Vancouver, IEEE) or author-date, and systematic and scoping reviews follow reporting guidelines such as PRISMA.

Common mistakes

Writing one paragraph per source, so the review reads like an annotated bibliography.
Starting sentences with author names, which keeps the focus on who said what rather than on what the field knows.
Claiming that “no research exists” on a topic without a thorough search to back it up.
Including everything you read to show effort, which buries the sources that matter.
Leaving the gap implicit and expecting the reader to work out why your study is needed.

Go further

UNC Writing Center: Literature ReviewsExplains the difference between summary and synthesis and how to organise a review chronologically, thematically or by method.
UC Merced Library: How to Write a Literature Review (synthesis step)Shows how to sort sources with a synthesis matrix and compares weak and strong synthesis paragraphs written by students.
University of Toronto Writing Advice: The Literature ReviewA short guide with self-check questions on scope, critical comparison and organising the review around your research question.
Academic Phrasebank (University of Manchester): Referring to sourcesSentence patterns for describing the literature, signalling agreement and disagreement, and choosing verb tenses when citing.
The Literature Review: A Step-by-Step Guide for Students, by Diana RidleyPractical guidance on searching, reading critically and writing a review in your own voice, with many examples from theses.