A literature review is an account of what is known about a question, organized so that a reader can see where the agreement is, where the disagreement is, and what has not been settled. It is a piece of reasoning about a body of work rather than a description of it, which is why the same twenty papers can produce an excellent review and a useless one. This page is the working definition, what the review is for depending on where it sits, and the three things it is commonly confused with.
What it is for depends on where it sits
In a thesis it does two jobs: it shows you know the field, and it earns the gap your own study will fill. In a journal article the background section does a compressed version of the same thing in a few hundred words. In a systematic review the literature is not the background, it is the data. Knowing which of the three you are writing settles almost every question about length, structure and how explicit the method needs to be.
The three things it is mistaken for
An annotated bibliography, which describes each source in turn and never compares them. A summary, which reports what was found without asking what it means together. And a justification, which starts from the conclusion the author wanted and cites only what supports it. All three are recognisable from the outside, and all three are what a reader means when they say a review is descriptive rather than critical.
The work happens before the writing
Almost everything that makes a review good is decided before a paragraph exists: what question it is answering, what counts as relevant, what was searched and where, what was read and what was set aside, and what was pulled out of each source. Doing that in a spreadsheet is possible and doing it in your head is not, which is why the extraction matrix exists and why the free worksheet on this site costs the work out in hours before you start.
How it differs from the background section of a paper
A background section does the same job in a few hundred words and with a single destination: it establishes just enough for the reader to understand why this study was done. A review chapter has to survey the field whether or not every part of it bears on your study. People writing their first chapter often write a long background instead, which reads as thin coverage rather than as concision.
When to write it
Earlier than the thesis timetable suggests, and again at the end. The early version is how you find the gap your study claims to fill, and it is nearly always wrong in useful ways. The final version is written when you know what you found, because the field looks different once your own result is in it. Treating it as a chapter written once, early, is why so many reviews sit oddly against the discussion.
What a supervisor is actually reading for
Three things: whether you know the field, whether you can tell important work from abundant work, and whether the gap you claim is real. The first is coverage, the second is judgement, and the third is the argument the whole chapter exists to make. A review that lists thoroughly and concludes weakly fails on the third, which is the one that matters at the viva.
Other questions people ask
What is the difference between a literature review and an essay?
An essay argues a position using sources as evidence. A review argues about the sources themselves: what they collectively show, where they disagree, and what is missing. The subject of a review is the body of work rather than the topic.
Does a literature review have to be systematic?
No. A systematic review follows a protocol and reports its selection in detail; a narrative review is a reasoned reading of a field. Both are legitimate, and the difference to be honest about is how reproducible your selection was.
How many sources does a literature review need?
Enough to cover the field, which depends entirely on how large the field is. Coverage is the measure, not the count: a reader who knows the area should not be able to name an important paper you missed.
When should I write my literature review?
Draft it early to find your gap, then rewrite it at the end when you know what you found. The field genuinely looks different once your own result is in it, and reviews written once and early sit oddly against the discussion.
How is it different from the background of a paper?
A background section establishes just enough for one study in a few hundred words. A review chapter surveys the field whether or not every part bears on your study, which is why a long background is not a short review.