Last updated 10 min read
Key takeaways
A literature review surveys the published research on a defined question and reports what that research, taken together, shows. It appears in three common forms: as a standalone assignment, as a chapter of a thesis or dissertation, and as a short section near the start of a journal article. In every form the job is the same — show the reader what is already known, where scholars disagree, and why the question still matters.
What it is not is a string of book reports. A paragraph that summarizes one study, followed by a paragraph that summarizes the next, leaves the real work — comparing them — to the reader. Your marker wants to see you weigh sources against each other: which findings agree, which conflict, which rest on weak methods, and which questions nobody has asked yet. If you wrote an annotated bibliography earlier in the project, treat it as raw material, not a first draft.
Know which type you are writing. A narrative review, the usual coursework assignment, selects and interprets the most relevant work. A systematic review follows a pre-set search protocol and records how every source was screened. Unless your brief says otherwise, you are writing a narrative review, but borrowing the systematic habit of logging your searches will save you time later.
The most common reason a literature review stalls is a question that is too broad. “Sleep and school performance” could fill a library shelf; “how later school start times relate to sleep duration among high school students” gives you a shelf you can actually read. If your question is still fuzzy, sharpen it first with the guide to writing a good research question — the review will inherit every weakness in it.
Then write down your inclusion criteria, the rules for what gets in. Decide the date range (fast-moving fields often look at the last ten to fifteen years, while foundational older work belongs in regardless), the disciplines, the types of source you will count — peer-reviewed articles, scholarly books, government reports — and any boundaries of language or region. Writing these down before you search keeps you honest when an interesting but off-topic article tempts you sideways. Settle these questions on one page:
Here is the full process, from an empty folder to a finished draft. The step most often skipped is the fourth, and the cost shows up two steps later, when nothing in your notes can be compared.
The patterns in your notes grid are the skeleton of the review. If six studies measure the same outcome and disagree, that disagreement is a section. If a whole cluster of studies uses one method and nobody has tried another, that is a section too — and possibly your gap. This grid is often called a synthesis matrix: one row per source, one column per theme, and a few words in each cell. Then choose an organizing principle for the body:
You can combine these. A thematic review that opens with a short chronological account of how the debate began is common and works well. What you should avoid is the accidental structure, where the order of sections simply mirrors the order in which you happened to read things.
| Organization | How the body is divided | Works best when |
|---|---|---|
| Thematic | By recurring topics or findings across the literature | Most questions; the usual default for coursework |
| Chronological | By period, showing how thinking changed | Change over time is itself what you are studying |
| Methodological | By research approach, such as surveys, experiments or case studies | Different methods produce different answers, and that matters |
| Theoretical | By the frameworks or schools of thought scholars apply | Your project chooses between theories or combines them |
Whatever organizing principle you choose, the review still needs a frame around it. If your review is a section of a larger paper rather than a standalone assignment, the guide to structuring a research paper with IMRaD shows where it sits.
State the topic and why it matters, the question the review answers, and the scope — years, fields, source types. In a thesis chapter, also say how the review connects to your own study. One to three paragraphs is usually enough; a long history of the topic belongs in the body, if anywhere.
Each section takes one theme, debate or method. Open with a topic sentence that makes a claim about the literature, use several sources to support or complicate it, and close by saying what the section means for your question.
Summarize the main patterns, name the most important gap or unresolved tension, and explain how your project — or future research, if the review stands alone — responds to it. Do not introduce new sources here.
The difference between summary and synthesis shows up at the level of the sentence. A summary sentence is about one source: “Author A (2019) surveyed secondary-school teachers and found that workload predicted burnout.” A synthesis sentence is about the relationship between sources: “Survey studies consistently link workload to burnout (Author A, 2019; Author B, 2021), but the longitudinal studies suggest the link weakens once school leadership is accounted for (Author C, 2020).”
Useful synthesis moves include agreement (“several studies converge on…”), disagreement (“findings diverge on…”), qualification (“this holds for X but not for Y”), method critique (“most of this evidence comes from self-report”), and absence (“none of these studies follows participants beyond a single year”). Each of these makes a claim you, the writer, are responsible for. That is what turns a review into an argument.
Tense matters too. APA-style social science writing uses the past tense or present perfect for what researchers found (“found,” “have reported”); the humanities use the present for what a text or critic argues (“Said argues”). Keep one convention per section.
The hardest part of a literature review is holding a dozen papers in your head at once. In Cavua's Work space you can upload your own PDFs and Word files and research across several of them at the same time — ask “where do these papers disagree about how burnout is measured?” and get answers that point to the page, so you can open the passage and check it before it goes into your notes.
Use “Quiz me” to test whether you really remember which study found what, export a session to keep a record of what you asked, and talk it through with the tutor by voice or video call when that helps. The synthesis, and the judgment about which sources matter, stay yours. How much each plan includes is listed on the pricing page.
Run your draft against this list before you hand it in.
If you are unsure whether the gap you name at the end is real, check the newest work that cites your key studies; the guide to using Google Scholar's Cited by links and date filters shows how to do that before you commit to it.
What an AI research assistant can do for students, what it can't, how to test one before you rely on it, and what to keep out of the files you upload to it.
APA citation in the 7th edition: in-text rules for one, two and three or more authors, plus reference formats for articles, books, chapters and webpages.
Learn how to avoid plagiarism with clear rules for quoting, paraphrasing and citing, examples in APA and MLA, and a checklist to run before you submit.
Where to find credible sources for research: library databases, scholarly search engines, official data and citation trails, plus quick checks before citing.
How to read a research paper in three passes: what to read first, how to decode figures and methods, which questions to ask, and how to take notes that last.
How to use AI for research without being misled: where AI tools help, the five ways they mislead, a verification workflow and how to disclose AI use honestly.
It depends on where it appears. A review section in a journal article may run a few pages, a dissertation chapter is usually much longer, and a standalone assignment normally comes with a word count. Follow your brief first; otherwise match recent reviews in your field, and let the number of themes, not the number of sources, decide the length.
There is no universal number, and your instructor's guidance overrides anything general. The better test is coverage: include the foundational works, the most important recent studies, and anything that directly challenges your argument. When fresh searches keep returning studies you already have, you have probably reached a reasonable stopping point for the scope you defined.
An annotated bibliography lists sources one at a time, each followed by a paragraph that summarizes and often evaluates it. A literature review is continuous prose organized by theme or debate, comparing sources against each other to make an argument about the field. Many students write the bibliography first and then mine it as raw material for the review.
Often, yes, in moderation. APA style accepts first person for describing what you did, such as “I searched three databases,” and many humanities fields allow it too. Some instructors still prefer impersonal phrasing, so check your assignment guidelines, and keep the attention on the literature rather than on yourself.
It should include your judgment, not your personal preference. Deciding that one group of studies is more convincing because of its methods, or that a whole debate rests on a shaky assumption, is exactly what a strong review does. Saying you like or dislike a theory without evidence is not. Tie every evaluative claim to something specific in the sources.
Begin with recent review articles and handbook chapters on your topic. They map the main debates, name the foundational studies, and give you reference lists to work through. Read two or three of them before searching for individual studies, note the vocabulary they use, and build your database search strings from those terms rather than from guesses.
Every Cavua course comes with a 24/7 tutor you can video call, speak to or chat with — on the material you are actually studying.
