# How to Write a Literature Review for a Thesis: 7 Steps With Examples

> How to write a literature review for a thesis in 7 steps: a search log, a synthesis matrix, an outline, synthesis sentence frames and length rules from universities.

Canonical: https://www.thesisroom.com/blog/how-to-write-a-literature-review-for-a-thesis
Last updated: 2026-11-18

Published: 2026-11-03

To write a literature review for a thesis, start from the questions your study asks, search and log your sources, read them into a matrix, organize them by theme, method or time, and write the chapter as an argument that ends on the gap your research fills. The skill that separates a review from a reading list is synthesis: UNC's Writing Center puts it as a summary recapping a source, while a synthesis is "a re-organization, or a reshuffling, of that information" ([UNC Writing Center](https://writingcenter.unc.edu/tips-and-tools/literature-reviews/)).

The seven steps below take you from an empty folder to a finished chapter. Each one has something you can do today: a search log, a matrix layout, an outline, and sentence frames for synthesis.

## What is a literature review in a thesis for?

A thesis literature review shows what is already known about your question, where the field disagrees, and what is still missing, so that your own study has a reason to exist. Cranfield University's library notes that most PhD and Master's theses contain some form of literature review "to provide the background for the research" ([Cranfield University](https://library.cranfield.ac.uk/writing-your-thesis/literature-review)).

Researchers who study doctoral training take it further. Boote and Beile argue that "a thorough, sophisticated literature review is the foundation and inspiration for substantial, useful research" ([Boote and Beile, 2005](https://stars.library.ucf.edu/scopus2000/4211)). A literature review is also one chapter in a larger structure, and the [PhD thesis structure guide](https://www.thesisroom.com/blog/phd-thesis-structure) shows what the chapters around it need from it.

## How to write a literature review for a thesis in 7 steps

### Step 1. Write down the questions the review must serve

Before you search, write your research questions or aims at the top of a blank page. Under each one, list the two or three things a reader must already know for that question to make sense. UNSW's guide also starts here: identify your research questions, then generate keywords and run topic searches ([UNSW Sydney](https://www.unsw.edu.au/student/managing-your-studies/academic-skills-support/toolkit/honours-postgraduate/literature-review)).

This list becomes your filter. A paper that informs none of those things can be interesting and still not belong in the chapter.

### Step 2. Search, and keep a search log

Search your main databases with the keywords from Step 1, and record every search as you run it. The University of Essex library suggests a log with five columns: Date, Database, Search, number of results, and Comments, because it is "important to keep track of where you've searched, what you searched for and what you found" ([University of Essex Library](https://library.essex.ac.uk/searchskills/search/track)).

| Date | Database | Search | Results | Comments |
|---|---|---|---|---|
| 3 Nov | Scopus | "feedback frequency" AND "doctoral completion" | 42 | 9 relevant; most are surveys |
| 3 Nov | ERIC | "supervision" AND "time to degree" | 96 | too broad; add "PhD" next time |

The log pays off twice. You can rerun a search before submission to catch new papers, and you can answer the examiner who asks how you found your sources. Cranfield also suggests setting email alerts based on your searches, so new work reaches you while you write ([Cranfield University](https://library.cranfield.ac.uk/writing-your-thesis/literature-review)). If you are unsure how far to formalize the search, compare the options in [systematic review vs literature review](https://www.thesisroom.com/blog/systematic-review-vs-literature-review).

### Step 3. Read into a synthesis matrix

[A synthesis matrix](https://www.thesisroom.com/blog/synthesis-matrix) is a spreadsheet where each source is a row and each theme is a column. Adler University's library describes exactly this layout and suggests columns such as APA citation, keywords, location, purpose, research design, main findings and notes ([Adler University Library](https://library.adler.edu/literature_review_guide/stay_organized)). Johns Hopkins describes the purpose: to record each source's main points and document "how sources relate to each other" ([Johns Hopkins University](https://guides.library.jhu.edu/lit-review/synthesize)).

Here is a worked example for a study on how often supervisors give feedback. The sources are placeholders to show the method, not real papers.

| Source | Design | Sample | Theme: frequency and completion | Theme: student anxiety | Notes |
|---|---|---|---|---|---|
| Study A (2019) | Survey | Master's students | Weekly feedback linked to faster completion | Not measured | Correlational only |
| Study B (2021) | Interviews | 18 doctoral students | Students value regular meetings | Anxiety rises when feedback is late | Small sample |
| Study C (2023) | Survey | Master's students | Same link as Study A | Not measured | Same population as A |

Read the columns, not the rows. The frequency column shows agreement but only for Master's students. The anxiety column has one study. Those two observations are already the skeleton of your argument and your gap. Before a source earns a row, run it through the filters in [how to evaluate sources for a literature review](https://www.thesisroom.com/blog/how-to-evaluate-sources-for-a-literature-review).

### Step 4. Choose how to organize the chapter

UNC's Writing Center says the body of a review is usually organized "chronologically, thematically, or methodologically" ([UNC Writing Center](https://writingcenter.unc.edu/tips-and-tools/literature-reviews/)). Pick the pattern that makes your gap easiest to see.

| Pattern | Use it when | Watch out for |
|---|---|---|
| Thematic | Your questions break into distinct topics or debates | Themes that are really just one author each |
| Chronological | The field changed direction at clear points | Turning into a history with no argument |
| Methodological | The disagreement comes from how studies were done | Losing the content behind the methods |

UNSW adds two more shapes: the inverted pyramid, which moves from broad to specific, and a structure built around "classic" studies. It also says the patterns can be combined ([UNSW Sydney](https://www.unsw.edu.au/student/managing-your-studies/academic-skills-support/toolkit/honours-postgraduate/literature-review)). Massey University notes that a methodological review usually moves "from weaker to stronger designs, for example correlational before experimental designs" ([Massey University](https://www.massey.ac.nz/study/study-and-assignment-support-and-guides/types-of-assignments/literature-review/)).

### Step 5. Outline the chapter section by section

A literature review needs at least three parts, according to UNC: an introduction, a body, and a conclusion ([UNC Writing Center](https://writingcenter.unc.edu/tips-and-tools/literature-reviews/)). In a thesis, the body is where your themes become numbered sections.

Glasgow Caledonian University gives a model outline for a dissertation chapter on food banks ([Glasgow Caledonian University](https://edshare.gcu.ac.uk/id/document/19132)):

- 2.1 Introduction to chapter
- 2.2 Macro Economic Theory
- 2.3 Usage patterns for food banks
- 2.4 Reasons for usage of food banks
- 2.5 Public perceptions of food banks
- 2.6 Main themes for primary investigation

Notice the last section. It turns the review into the questions the study will investigate, which is the bridge to your methodology chapter. Write your own outline the same way, with one section per theme from your matrix and a final section that names what your study will do.

### Step 6. Write synthesis paragraphs, not summaries

Each paragraph should make one claim about the literature and then use several sources to support, qualify or challenge it. A summary paragraph says what one author found. A synthesis paragraph says what the field shows, and where it does not agree.

Compare two versions built from the matrix in Step 3:

- **Summary:** Study A found that weekly feedback was linked to faster completion. Study B interviewed doctoral students about meetings. Study C repeated Study A's survey.
- **Synthesis:** Two surveys link frequent feedback to faster completion (Study A, 2019; Study C, 2023), but both sampled Master's students only. The one doctoral study is qualitative (Study B, 2021), so the link has not been tested at doctoral level.

The University of Manchester's Academic Phrasebank collects sentence frames that do this work ([University of Manchester](https://www.phrasebank.manchester.ac.uk/referring-to-sources/)):

- "In contrast to Smith, Jones (2013) argues that"
- "This view is supported by Jones (2015) who writes that"
- "Together, these studies indicate that …"
- "There remain several aspects of X about which relatively little is known."

UNSW also points out the choice between citations that put the information first, such as "(Williams, 2003)", and ones that put the author first, such as "Williams (2003) demonstrated that" ([UNSW Sydney](https://www.unsw.edu.au/student/managing-your-studies/academic-skills-support/toolkit/honours-postgraduate/literature-review)). Use information-first citations when several sources agree, and author-first ones when a single study matters to your argument.

### Step 7. End on the gap, then check every reference

Close the chapter with what the field agrees on, [what it has not tested](https://www.thesisroom.com/blog/research-gap-how-to-find-one), and how your study responds. Glasgow Caledonian offers a plain bridge sentence: "This review of relevant literature identifies a gap. Little research has been undertaken on the impact of X on Y" ([Glasgow Caledonian University](https://edshare.gcu.ac.uk/id/document/19132)). Our guide to the [conclusion of a literature review](https://www.thesisroom.com/blog/conclusion-of-a-literature-review) has two annotated examples and gap sentences that hold up under questioning.

Then check the references before anyone else reads the chapter. Open every source you cite, confirm the author, year and title match your entry, and confirm the paper says what your sentence claims. For the existence and metadata part, you can paste your reference list into the [free Citation Check](https://www.thesisroom.com/citation-check), which compares each entry with the academic record. If your department uses APA, the [annotated APA 7 literature review example](https://www.thesisroom.com/blog/literature-review-apa-format) shows the headings, citations and reference list laid out.

## How long should a literature review be in a thesis?

There is no universal rule, and your department's guidance comes first. Several universities publish rough figures:

| University | Guidance | Applies to |
|---|---|---|
| UNSW Sydney | Around 15 to 30% of the whole thesis; over 40% is probably too long | Honours and postgraduate theses |
| University of Kent | Perhaps 8,000 to 10,000 words in a PhD; around 2,000 to 3,000 words in a Master's dissertation | Rough guide |
| Glasgow Caledonian University | 3,500 to 4,000 words in a 12,000-word dissertation | Dissertations |

Sources: [UNSW Sydney](https://www.unsw.edu.au/student/managing-your-studies/academic-skills-support/toolkit/honours-postgraduate/literature-review), [University of Kent](https://media.www.kent.ac.uk/se/52777/LiteratureReviews.docx), [Glasgow Caledonian University](https://edshare.gcu.ac.uk/id/document/19132).

The figures disagree, which is the point: length follows your thesis type and field. Ask your supervisor which proportion your department expects, and plan your outline against that number.

## How many sources, and how recent?

Kent gives "a very rough rule of thumb": 8 to 10 significant pieces for an 8,000-word dissertation, and up to 20 major pieces for 12,000 to 15,000 words ([University of Kent](https://media.www.kent.ac.uk/se/52777/LiteratureReviews.docx)). These are the works you discuss in depth, not [the total you cite](https://www.thesisroom.com/blog/how-many-references-should-a-thesis-have).

On age, Kent says sources "should be as up-to-date as possible, with the exception of 'classic texts'". Massey suggests about 10 years as a rough guide for Master's theses ([Massey University](https://www.massey.ac.nz/study/study-and-assignment-support-and-guides/types-of-assignments/literature-review/)). [Field matters too](https://www.thesisroom.com/blog/how-old-are-references-in-recent-research): UNC notes that in the sciences, for example on treatments for medical problems, "Information even two years old could be obsolete" ([UNC Writing Center](https://writingcenter.unc.edu/tips-and-tools/literature-reviews/)).

## Printable literature review checklist

Tick each box before you send the chapter to your supervisor.

- [ ] My research questions are written at the top of my outline
- [ ] Every search is in my search log with date, database, terms and results
- [ ] Every source I discuss has a row in my synthesis matrix
- [ ] I chose one organizing pattern (thematic, chronological or methodological) and can say why
- [ ] Each body section matches a theme or debate, not a single author
- [ ] Each paragraph opens with a claim about the literature, not about one study
- [ ] Each section ends by saying what it means for my research questions
- [ ] The chapter ends with the gap and how my study responds
- [ ] I have opened every source I cite and checked author, year and title
- [ ] The length fits my department's guidance
- [ ] Recent work in my field is included, and older sources are there for a reason

## Frequently asked questions

### How do I start writing a literature review for my thesis?

Start with your research questions, not with reading. List what a reader must know for each question to make sense, search for that, and record each search in a log. Then read into a synthesis matrix so the themes appear before you write a sentence.

### What are the main parts of a thesis literature review?

An introduction that sets the scope, a body organized by theme, method or time, and a conclusion that states the gap. UNC's Writing Center lists these three as the basic elements. In a thesis, it helps to map the body sections to your research questions.

### What is the difference between summary and synthesis?

A summary recaps one source. A synthesis reorganizes several sources around a claim, showing where they agree, where they differ and what none of them tested. UNC's Writing Center draws the same line, so build paragraphs around claims rather than around authors.

### Can a literature review be organized chronologically?

Yes, when the field changed direction at clear points and that change matters to your question. UNC lists chronological, thematic and methodological patterns, and UNSW says they can be combined. Avoid a chronology that reads as history with no argument.

### Should my thesis literature review be systematic?

Only if your department or your research question requires it. A narrative or structured review with a documented search is the other option. Check your program's handbook and ask your supervisor before you commit to a full systematic protocol.

## Sources

- [UNC Writing Center: Literature Reviews](https://writingcenter.unc.edu/tips-and-tools/literature-reviews/)
- [UNSW Sydney: Literature review](https://www.unsw.edu.au/student/managing-your-studies/academic-skills-support/toolkit/honours-postgraduate/literature-review)
- [University of Kent: Literature Reviews study guide](https://media.www.kent.ac.uk/se/52777/LiteratureReviews.docx)
- [Massey University: Literature review](https://www.massey.ac.nz/study/study-and-assignment-support-and-guides/types-of-assignments/literature-review/)
- [Glasgow Caledonian University: Structuring your literature review chapter](https://edshare.gcu.ac.uk/id/document/19132)
- [University of Manchester: Academic Phrasebank, Referring to sources](https://www.phrasebank.manchester.ac.uk/referring-to-sources/)
- [Cranfield University: Writing your thesis, Literature review](https://library.cranfield.ac.uk/writing-your-thesis/literature-review)
- [University of Essex Library: Keep track of your search](https://library.essex.ac.uk/searchskills/search/track)
- [Adler University Library: How do I stay organized?](https://library.adler.edu/literature_review_guide/stay_organized)
- [Johns Hopkins University: Synthesize](https://guides.library.jhu.edu/lit-review/synthesize)
- [Boote and Beile (2005), Scholars Before Researchers, Educational Researcher (UCF repository record)](https://stars.library.ucf.edu/scopus2000/4211)

How this guide was made: Thesisroom drafted this guide with the help of AI writing tools. Every fact, figure and source in this post was checked against its original source before publishing.

## Related

- [All guides](https://www.thesisroom.com/blog)
- [Check references free](https://www.thesisroom.com/citation-check)
