How to Write a Literature Review Chapter

# How to Write a Literature Review Chapter: Synthesising Sources into a Position

A literature review chapter is where many first-time researchers lose marks they could have kept. The problem is rarely effort. Students read widely, take notes, and produce thirty pages of accurate summary — then get the same comment: this reads like a list. The chapter summarises well but argues nothing.

This tutorial is about the difference between those two outcomes, and how to move from one to the other. It is written for final-year undergraduate and postgraduate students in Uganda preparing a first independent study, though the principles apply more widely.

1. A Literature Review Is an Argument, Not a Queue of Summaries

The literature review chapter has a defined job inside your thesis or empirical paper. It is not a record of what you have read. It is where you establish what is known, what is disputed, and what remains unresolved — and then position your own work in relation to that [S4][S3]. Writing the review is a stage in the larger thesis process, with its own method rather than a spillover from your reading [S3][S1].

Review writing is also a distinct genre with its own conventions. Different review genres make different demands on the writer [S6]. A book review for a peer-reviewed journal is a review — but a different kind, with a different purpose, and its conventions do not transfer directly to a thesis chapter [S5].

What "one source after another" looks like on the page

You will recognise the pattern immediately:

Mwesigwa (2019) found that mobile money improved rural savings. Achieng (2020) found that mobile money increased household consumption. Okello (2021) found that mobile money reduced the use of informal credit.

Three sentences, three sources, no argument. Each source gets one sentence and steps aside for the next. Nothing is compared, nothing is weighed, nothing is claimed.

What synthesis looks like instead

Synthesis groups sources by idea, and puts sources in supporting roles. The same three studies might produce this:

Evidence on mobile money and rural households points in two directions. Several studies report improved savings and consumption among adopting households (Mwesigwa, 2019; Achieng, 2020), but Okello (2021) finds a simultaneous reduction in informal credit use — a result that complicates the simple "mobile money helps" narrative, since informal credit often serves households that formal systems exclude.

Here the sources do work for a claim. They agree, they qualify, and they create a problem the writer can then address.

Why this distinction matters

A review that only summarises shows reading. It does not show thinking. That distinction is widely repeated in academic writing advice, but it is worth being honest: none of the sources behind this tutorial establishes a marking rubric or an examination standard for literature reviews. We cannot tell you how your examiners weight synthesis against coverage. What we can say is that the difference is visible on the page, and that the rest of this tutorial is about producing the second kind of paragraph rather than the first.

2. Before You Write: Turn Your Source List into a Map

Planning and organising your sources before you draft is part of the review-writing process, not a preliminary you can skip [S2][S8]. In systematic review work, searching for and selecting sources is treated as a distinct step that precedes writing, with structured decisions about scope and inclusion made in advance [S8][S2]. You do not need to run a systematic review to borrow that discipline.

Build a source table

Before drafting, put every source into a table with five columns:

The value is not the table itself. Filling it forces you to state each source's claim in your own words, and to notice which sources are in tension with which others. The specific columns suggested here are a practical device, not a documented method from the sources behind this tutorial — adapt them to your field.

Look for the four things that create structure

When you read down your table, four patterns give a review its shape:

• Agreement — several sources converge on a finding.
• Disagreement — sources reach different conclusions, often because of method or context.
• Gaps — questions the literature has not answered.
• Shifts over time — the field's assumptions changing across periods.

This four-part scheme is a heuristic, not a taxonomy drawn from a source. It is useful because it converts a flat list into a set of possible sections.

Write one sentence per cluster

Before writing any paragraph, write a single sentence for each cluster of sources. For example: Studies agree that X holds in urban settings, but disagree on whether it holds in rural ones. Those sentences become your topic sentences later. If you cannot write a sentence for a cluster, the cluster is not yet a cluster — it is just a pile.

3. Choose the Organising Principle Before You Choose the Paragraphs

Review writing needs a deliberate structure rather than an accumulated sequence of summaries [S7][S6]. Review articles serve particular purposes, and those purposes shape how they are organised [S6]. Choose your organising principle before you write a single paragraph, because the principle determines what each section is for.

There are three workable patterns. None of the sources behind this tutorial presents exactly this three-way scheme; treat them as practical options rather than a sourced taxonomy.

Thematic organisation

You group sources by idea or concept. This is the most common pattern and the easiest to defend, because it maps directly onto the clusters you built in Section 2. A chapter on mobile money might have sections on adoption, household consumption, and credit substitution — regardless of when each study was published.

Debate or controversy organisation

You group sources by the disagreement they are part of. This is the strongest pattern when your study takes a side, because the structure itself sets up your contribution. If the field is split on whether a policy works, your chapter can be organised around that split, with each section representing a position and its evidence.

Chronological organisation

You group sources by period. This works when the field itself has shifted — when a method, a theory, or a consensus changed at a recognisable point. It fails when it becomes a history lesson: a chronological review that does not explain why the shift matters is just a timeline with citations.

Why mixing all three usually fails

A chapter that moves between thematic, debate-based, and chronological organisation without signalling the shift produces the flat draft readers complain about. The reader cannot tell whether a new section is a new theme, a new position, or a new decade. Pick one principle as your primary structure. You can use a chronological sub-section inside a thematic chapter, but the reader should always know which level they are on.

4. Write Paragraphs That Make a Claim, Not a Report

Review writing involves making an argument rather than reporting sources in sequence [S4][S6], and the way a review is written affects what it communicates to the reader [S7]. The paragraph is where that intention either becomes real or evaporates.

The claim-first paragraph

The pattern is simple. Your topic sentence states a position. The sources that follow support it, qualify it, or challenge it. The paragraph ends when the position has been established, complicated, or defended — not when you have run out of sources.

Signposting verbs that do real work

Weak synthesis leans on also, furthermore, and in addition. These verbs tell the reader that another source exists. They do not tell the reader what the source does to your argument. Stronger verbs do:

• agrees — the source supports your claim
• qualifies — the source supports it, but only under conditions
• contradicts — the source challenges it
• extends — the source takes the claim further
• applies only in — the source limits the claim to a context

These verbs are a writing technique, not a sourced recommendation. Their value is that they force you to decide what each source is doing in the paragraph.

A worked before-and-after

The following example is invented for illustration. It is not drawn from any source, and the citations are placeholders.

Before (summary):

Nakato (2018) found that teacher training improved pupil performance. Byaruhanga (2019) found that teacher training had no significant effect. Kizza (2020) found that teacher training improved performance in urban schools.

After (synthesis):

The evidence on teacher training and pupil performance is divided, and the division tracks context. Nakato (2018) and Kizza (2020) both report gains, but Kizza's effect appears only in urban schools. Byaruhanga (2019) finds no significant effect at all. Read together, these results suggest that training alone is not the active ingredient — something about the school environment may determine whether training translates into performance.

The second version makes a claim, uses the sources to build it, and leaves the reader with a problem worth investigating.

Handling disagreement honestly

When sources disagree, do not resolve the disagreement by ignoring it or by averaging it away. State the disagreement, and say what might explain it — method, sample, context, or period. If you cannot explain it, say that the disagreement remains unresolved. A review that admits an unresolved dispute is more credible than one that pretends the field agrees.

5. Move from Paragraphs to a Position

The literature review for an empirical paper connects to the study you are conducting, and it should lead toward your research problem or gap [S4]. That connection is what turns a chapter of paragraphs into a chapter with a destination.

Write your position as one sentence

Before you draft the conclusion, write your position as a single sentence. For example: The literature agrees that X matters but disagrees on how it works, and no study has tested it in rural Uganda. That sentence is your position. Everything in the chapter should be consistent with it.

This is a practical device, not a sourced method. Its value is that it makes your stance checkable. If a paragraph in your chapter contradicts the sentence, one of them is wrong.

Where the position goes

The position belongs at the end of the chapter, where it sets up the gap your study fills. It should be echoed in your research gap and problem statement, so that a reader moving from your review to your methodology sees a continuous argument rather than a jump. If you have not yet written your problem statement, see How to Write a Research Problem Statement; if you are still assembling the proposal around it, How to Write a Research Proposal: A Practical Guide for University Students in Uganda covers the wider structure.

Linking the review to your questions

The final paragraph of your review should make the link explicit. Your research questions should be answerable by the gap you have just identified. If they are not, the chapter ends in mid-air: the reader has followed your argument to a problem, and then been handed a different study.

6. Cite as You Synthesise: Keeping the Reference List Honest

Citing and referencing are part of writing a review, not an afterthought [S8][S2]. Synthesis makes citation harder, because a single sentence may now carry several sources.

Citing several sources for one claim

When several sources support one claim, group them in a single parenthetical citation, separated by semicolons, in the order they appear in your reference list. This keeps the sentence readable and signals that the claim rests on a body of work rather than one study. Avoid citation stacking — attaching a long string of citations to a sentence that does not actually make a claim. If the sentence says nothing, the citations do not rescue it.

Quote rarely, paraphrase usually

Synthesis usually means paraphrase, because you are combining claims rather than reproducing wording. Quote when the exact phrasing matters — a definition, a contested term, a distinctive formulation. Paraphrase when you are reporting what a source found or argued. If you are unsure where the line falls, How to Paraphrase Without Plagiarising: A Checklist works through the common cases.

APA 7 in synthesis paragraphs

None of the sources behind this tutorial covers APA 7 formatting, so we cannot cite them for it. For the rules themselves, see APA 7 Referencing Explained: In-Text Citations, Reference Lists, and Common Errors and APA 7 Reference List Rules Students Actually Need. The errors that appear most often in synthesis paragraphs are missing page numbers for quotations, sources cited in text but absent from the reference list, and grouped citations ordered inconsistently.

7. Four Tests You Can Run on a Draft You Already Have

Reviewing and revising a draft is part of the writing process, not a final polish [S3][S7]. If you already have a chapter and suspect it is summarising rather than synthesising, these four tests will tell you quickly.

They are practical heuristics, not validated instruments. They are not drawn from any source, and they do not replace your supervisor's judgement.

The topic-sentence test

Read only your topic sentences, in order, ignoring everything else. Do they form an argument? If they read as a list of topics rather than a sequence of claims, the chapter is organised as a survey.

The name-swap test

Remove every author name from a paragraph. Does the paragraph still say anything? If deleting the names leaves you with nothing but "research has found that X is important," the paragraph was reporting, not arguing.

The "so what" test

For each paragraph, ask what it contributes to your research problem. If the honest answer is "background," consider whether it belongs in the chapter at all, or in a shorter introduction.

The gap test

Can you point to the sentence where the gap your study fills becomes visible? If you cannot find it, your reader cannot either.

8. Common Problems and What They Usually Mean

Most first literature reviews fail in one of four recognisable ways. Each has a structural cause, and each has a structural fix.

The annotated bibliography in disguise

The chapter reads as one source after another, each summarised in turn. The cause is a missing organising principle. Return to Section 3 and choose one.

The everything chapter

The chapter tries to cover the entire field. The cause is the absence of inclusion criteria. Deciding what to leave out is part of the method, not a failure of diligence [S2][S8]. For help selecting sources, see How to Find Credible Academic Sources for a Literature Review.

The missing position

The chapter ends without a stance. The cause is drafting paragraphs before deciding what they add up to. Return to Section 5 and write your position sentence.

The patchwork paragraph

Sources are stitched together with also and furthermore instead of being argued. The cause is a paragraph without a claim. Return to Section 4 and rewrite the topic sentence first.

9. A Worked Order of Operations for This Week

This sequence compresses the tutorial into steps you can start today. It is a suggested order of operations, not a method drawn from any source.

1. Build the source table (Section 2).
2. Choose the organising principle (Section 3).
3. Write topic sentences only, for the whole chapter.
4. Check that the topic sentences form an argument (Section 7).
5. Draft paragraphs claim-first (Section 4).
6. Write the position sentence and the gap sentence (Section 5).
7. Check citations and paraphrasing (Section 6).

10. What This Guide Cannot Do for You

Three limits are worth stating plainly.

Your department's format requirements override any general advice here. This tutorial cannot tell you what your university requires, because no source behind it establishes any institutional standard. Makerere University, Kyambogo University, Uganda Christian University, and other institutions each set their own thesis and dissertation regulations, and those regulations change. Check your own faculty handbook or ask your supervisor rather than relying on a general guide.

Supervisors vary in how much synthesis they expect at undergraduate and master's level. We cannot tell you where your supervisor draws that line. Ask them directly, and show them a sample paragraph.

No guide replaces reading the sources yourself. Synthesis of unread sources is not synthesis — it is arrangement, and it will not survive a question in a viva. If you are using AI tools at any stage, Using AI in Academic Research Responsibly sets out where the line falls. For the wider chapter structure around this one, see How to Structure a University Dissertation, Chapter by Chapter, and for sentence-level clarity, Clearer Academic Writing Without Losing Rigour.

A note on access and cost

Much of the guidance above assumes you can reach the sources you need. In practice, access varies. Some Ugandan university libraries subscribe to major databases; others do not, and coverage differs by institution. We cannot state which databases any particular Ugandan university subscribes to, because that information is not in the sources behind this tutorial and changes from year to year. Ask your library directly.

Open-access journals and repositories are free to read, and many of the sources behind this tutorial are open access [S3][S4][S5][S8]. For search techniques that work with limited subscriptions, see How to Search Google Scholar and Library Databases.

Sources used

• [S1] Bui, Yvonne N. (2009). How to Write a Master's Thesis. http://ci.nii.ac.jp/ncid/BB12935161
• [S2] Dhiman, Rahul; Srivastava, Vimal; Srivastava, Anubha; Rajni; et al. (2023). How to Plan and Write for Systematic Literature Review Papers in Management Domain. doi:10.1108/s2754-586520230000002003
• [S3] Faryadi, Qais (2018). PhD Thesis Writing Process: A Systematic Approach—How to Write Your Literature Review. doi:10.4236/ce.2018.916219
• [S4] Nakano, Davi; Muniz, Jorge (2018). Writing the literature review for empirical papers. doi:10.1590/0103-6513.20170086
• [S5] Lee, Alexander D; Green, Bart N.; Johnson, Claire D.; Nyquist, Julie G. (2010). How to Write a Scholarly Book Review for Publication in a Peer-Reviewed Journal: A Review of the Literature. doi:10.7899/1042-5055-24.1.57
• [S6] Yan, Zheng (2020). How Should We Write Review Articles?. doi:10.1017/9781108277426.014
• [S7] Reeves, Sue; Buczkowski, Bartek (2023). How Do I Write an Introduction and Literature Review?. doi:10.1007/978-3-031-41911-9_6
• [S8] Goodfellow, Lynda Thomas (2023). An Overview of How to Search and Write a Medical Literature Review. doi:10.4187/respcare.11198

All eight supplied sources are cited in the text. No source outside the supplied list is cited.

Editor's note on changes made

The draft was already clean, so this pass was light. I made the following edits:

• Cut filler and restatement. Removed "and still get the same comment" → "then get the same comment"; "The paragraph is a queue" (redundant after the three-sentence example); "This is the most common pattern and the easiest to defend, because..." kept but trimmed; "It is not a record of what you have read; it is the part of the study where..." → split into two sentences for rhythm.
• Removed hedging. "It is not a record of what you have read; it is the part of the study where you establish" → "It is not a record of what you have read. It is where you establish." Same meaning, less padding.
• Concrete wording. "the part of the study where" → "where"; "the practice of attaching" → "attaching"; "the absence of inclusion criteria" kept (already concrete).
• Sentence-length variation. Broke several long compound sentences in Sections 1, 3, and 4 into shorter units. Left the deliberately long example paragraphs intact, since their length is doing work.
• Deleted restatement. In Section 3, "This is the most common pattern and the easiest to defend, because it maps directly onto the clusters you built in Section 2" — kept, but the following sentence about mobile money was tightened. In Section 4, "The paragraph ends when the position has been established, complicated, or defended — not when you have run out of sources" — kept, it earns its place.
• No facts, figures, or citations changed. Every [S] key, every example citation, and every linked URL is unchanged. The audit from the previous stage still holds.

Related reading

• How to Write a Research Problem Statement
• How to Find Credible Academic Sources for a Literature Review
• How to Paraphrase Without Plagiarising: A Checklist
• How to Structure a University Dissertation, Chapter by Chapter