What is a literature review? Definition, structure and example

A literature review is a structured survey of the published research on a topic. It doesn't just list sources: it groups them, compares them, judges their strengths and weaknesses, and shows what is known, what is debated and what is still missing. A good literature review reads as an argument about the state of the research, ending with the gap your own work addresses. It can be a section of a research paper or thesis, or a stand-alone assignment.

This guide explains what a literature review does, how it is structured, and how to write one, with an original example excerpt showing the difference between summarizing and synthesizing.

What a literature review is for

A literature review does four jobs:

1. Maps the field. It shows the reader the main studies, theories and debates on the topic.
2. Evaluates the research. It judges which findings are strong and which rest on small samples, old data or weak methods.
3. Finds the gap. It identifies a question that hasn't been answered, or answered well.
4. Justifies your work. In a research paper or thesis, it explains why your study is needed and how it builds on earlier work.

A stand-alone literature review does the first three and ends with what future research should do.

What a literature review is not

• Not an annotated bibliography. An annotated bibliography summarizes sources one at a time. A literature review is organized by ideas, and discusses several sources in the same paragraph.
• Not a list of summaries. "Smith found X. Jones found Y. Lee found Z." is summary. A review says how X, Y and Z relate.
• Not your opinion of the topic. Your voice appears in how you organize and evaluate the research, not in personal views about the subject.

Types of literature review

If your assignment just says "literature review", it almost always means a narrative review.

How a literature review is structured

Like an essay, a literature review has an introduction, a body and a conclusion.

Introduction

• The topic and why it matters.
• The scope: which years, populations, fields or types of study you included, and why.
• How the review is organized.

Body

The body groups the research. Choose one of these organizing patterns:

• Thematic: by theme or finding. The most common and usually the most useful.
• Chronological: by period, showing how thinking changed over time. Works when the history of the debate matters.
• Methodological: by research method, such as surveys versus experiments versus interviews. Works when different methods produce different results.
• Theoretical: by the theories or models researchers use.

Within each group, discuss several sources together, compare them, and evaluate them.

Conclusion

• The main patterns across the research.
• The key gap, limitation or unresolved debate.
• What it means for your study, or for future research.

Summary versus synthesis: an example

The difference between a weak and a strong literature review is synthesis: connecting sources rather than reporting them one by one. The example below reviews research on peer tutoring in university mathematics. The studies and authors are invented for this guide, so it shows the technique without borrowing anyone's findings.

A summary (weak)

Patel (2018) studied peer tutoring at one university and found that students who attended tutoring sessions had higher exam scores. Okoro and Lind (2020) surveyed 400 students and found that tutored students felt more confident. Brandt (2021) found no difference in grades between tutored and untutored students. Nguyen (2022) found that tutoring helped students who attended at least five sessions.

Every sentence is accurate, but the reader has to work out what it all means. The paragraph doesn't say whether tutoring works, why studies disagree, or which evidence is strongest.

A synthesis (strong)

Research on peer tutoring in university mathematics broadly suggests a benefit, but the size of that benefit seems to depend on how much tutoring students receive. Studies tracking attendance found higher exam scores among tutored students (Patel, 2018), particularly those attending five or more sessions (Nguyen, 2022). Brandt (2021), however, found no difference in grades, and unlike the other two studies, counted any student who attended a single session as tutored. This difference in definitions may explain the conflicting results. A second limitation runs through this work: students choose whether to attend tutoring, so more motivated students may be overrepresented among the tutored group. Okoro and Lind's (2020) finding that tutored students reported more confidence is consistent with both readings: tutoring may build confidence, or confident students may be the ones who attend. No study in this group assigned students to tutoring at random, leaving open whether tutoring itself raises scores.

What changed:

• The paragraph opens with a claim about the research, not a source.
• Sources are grouped by what they show, not listed by date.
• The writer explains a disagreement (different definitions of "tutored").
• The writer evaluates the evidence (self-selection is a weakness).
• It ends by naming a gap, which is what the rest of a paper could address.

A short literature review example

Here is how the pieces fit together in a short stand-alone review. It uses the same studies, with the strong paragraph above as its first body section. Every study and author in it is invented for this guide, so don't cite them; the example shows the shape of a review, not real findings.

Peer tutoring in university mathematics: a short review

Introduction. Many universities offer peer tutoring for introductory mathematics, usually as drop-in or scheduled sessions led by trained students. This review asks whether tutoring improves students' results in these courses. It covers studies of undergraduate mathematics tutoring published since 2015 and is organized by two themes, effects on grades and effects beyond grades, followed by the limits of the evidence.

Effects on grades. Research on peer tutoring in university mathematics broadly suggests a benefit, but the size of that benefit seems to depend on how much tutoring students receive. Studies tracking attendance found higher exam scores among tutored students (Patel, 2018), particularly those attending five or more sessions (Nguyen, 2022). Brandt (2021), however, found no difference in grades, and unlike the other two studies, counted any student who attended a single session as tutored. This difference in definitions may explain the conflicting results. A second limitation runs through this work: students choose whether to attend tutoring, so more motivated students may be overrepresented among the tutored group. No study in this group assigned students to tutoring at random, leaving open whether tutoring itself raises scores.

Effects beyond grades. Studies of outcomes other than grades are fewer but point the same way. Tutored students report more confidence in mathematics (Okoro & Lind, 2020), and frequent attenders are more likely to take a second mathematics course (Nguyen, 2022). Both findings rest on self-reports or enrollment records, however, and neither study could separate the effect of tutoring from the motivation of the students who chose it.

Conclusion. Taken together, these studies link peer tutoring to better grades and greater confidence, especially for students who attend regularly. Because students choose whether to attend, and studies define "tutored" differently, the evidence cannot yet show that tutoring causes these gains. A study that assigns students to tutoring at random, or compares similar students with and without access, would close this gap.

Notice that each section opens with a claim about the research, the introduction states the scope, and the conclusion names the gap. A real review would end with a reference list giving the full details of every source it cites.

How to write a literature review, step by step

1. Define your question and scope

Write the question your review answers, such as "What does research show about the effect of peer tutoring on university mathematics outcomes?" Then set limits: years, populations, countries, kinds of study. Limits keep the review manageable and should be stated in your introduction.

2. Search systematically

Use your library's databases for your field, plus Google Scholar. Keep a list of search terms and combinations so you can repeat and extend the search. Two habits find sources that keyword searches miss:

• Backward searching: check the reference lists of the best sources you find.
• Forward searching: use "cited by" links to find newer work that cites them.

3. Read and take structured notes

For each source, record the same things: the question, the method, the sample, the main findings, limitations, and how it relates to other sources. A table or spreadsheet with one row per source makes patterns easy to spot.

It also helps to save the exact passage a note came from, with its page number, so you can check your reading when you write. In ScribeSprout's Reader, you can save a passage of a source as a note that links back to where it came from, and get a summary of a source in which every point cites the passage it came from.

4. Find the patterns

Look across your notes for:

• Agreements: findings several studies share.
• Disagreements: findings that conflict, and possible reasons (different samples, methods, definitions or dates).
• Trends: how research has changed over time.
• Gaps: questions no one has studied, populations left out, methods not yet tried.

These patterns become the themes of your body.

5. Outline by theme

Make each theme a section heading, and list the sources you'll discuss under it. For the tutoring example:

1. Evidence of a benefit to grades
2. Why studies disagree: definitions and dosage
3. Effects beyond grades: confidence and persistence
4. Methodological limits: self-selection and short follow-up
5. Gap: no randomized studies

6. Draft, synthesizing in every paragraph

Start each paragraph with a claim about the research, then bring in the sources that support, qualify or challenge it. Useful phrases for connecting sources:

• "Consistent with this, ..."
• "In contrast, ..."
• "This finding is qualified by ..."
• "A limitation shared by these studies is ..."
• "Few studies have examined ..."

7. Write the introduction and conclusion, then revise

With the body done, write an introduction that sets out the scope and organization, and a conclusion that pulls the patterns together and states the gap. Then revise: check that every paragraph synthesizes, that every source is cited correctly, and that the gap you name follows from the evidence you've reviewed.

Common mistakes

• Organizing by source instead of by theme.
• Describing studies without evaluating them.
• Including only sources that agree with you. A review that ignores conflicting evidence is not credible.
• Quoting heavily. Paraphrase findings in your own words and cite them; save quotations for wording that matters. Our guide on what paraphrasing is shows how.
• Leaving the gap unstated. The reader should finish knowing exactly what is missing.

Frequently asked questions

What does a literature review look like?

It looks like an essay with headings, not a list. There is an introduction setting out the topic and scope, a body divided into themed sections that discuss several sources together, and a conclusion naming the main patterns and the gap. In-text citations appear throughout, and a full reference list follows.

What is a literature review in a research paper?

It is the section, often after the introduction, that surveys earlier research on your topic and shows where your study fits. In a short paper it may be a few paragraphs inside the introduction; in a longer paper or thesis it is usually its own section or chapter.

What is the purpose of a literature review?

To show what is already known about a topic, evaluate the quality of that research, and identify what is missing. In a research paper, it also justifies your study by showing the gap it fills.

How long should a literature review be?

It depends on the assignment. As a rough guide, a literature review section in a research paper is often about 15 to 25 percent of the paper, while a stand-alone review for a course may be 2,000 to 5,000 words, and a thesis chapter longer still. Always follow your assignment instructions.

How many sources should be in a literature review?

Enough to represent the main research on your question fairly. A short undergraduate review might use 10 to 20 sources and a thesis chapter many more, but your instructor's requirement comes first. Quality and relevance matter more than the count.