First Decide What Kind of Review You Are Writing
A coursework review, thesis chapter, scoping review, and systematic review do not have identical methods.
Start With a Review Question, Not a Search Box
The question decides what counts as relevant evidence.
Local checklist
Define the review before collecting sources
Progress is stored only in this browser.
Keep a Search and Selection Record
Transparency begins before writing.
| Field | What to record |
|---|---|
| Database or source | The database, catalogue, repository, citation trail, or official source searched. |
| Date | The date the search was run. |
| Search expression | Keywords, phrases, Boolean operators, and filters. |
| Results | Approximate number retrieved and the number screened. |
| Decision rule | Why sources were included or excluded. |
| Version | Whether the item is a preprint, accepted manuscript, or final published version. |
Separate Source Language From Source Meaning
Plagiarism prevention begins during note-making.
A safer note-making sequence
- 1
Read for meaning
Identify the claim, method, evidence, and limitation before copying any sentence.
- 2
Close or hide the source
Write a short account from understanding, not from the source's visible syntax.
- 3
Reopen and verify
Check that your note preserves the original meaning and strength of claim.
- 4
Label quotations visibly
If exact wording is essential, store it in quotation marks with a page number and keep it separate from paraphrase notes.
- 5
Attach the citation immediately
Never allow an extracted idea to become separated from its source identity.
Build a Source Matrix Before Drafting Paragraphs
Comparison is easier when every source is described through the same fields.
| Field | Question |
|---|---|
| Research problem | What does the study try to explain or resolve? |
| Context or sample | Where, with whom, or on what material was the study conducted? |
| Method | How was evidence generated or the argument developed? |
| Main finding or claim | What is the strongest conclusion supported? |
| Limitation | What boundary, uncertainty, or missing comparison matters? |
| Theme | Which part of my review does this source inform? |
| Relationship | Does it support, extend, qualify, or contradict another source? |
Synthesis Means Explaining Relationships
A paragraph should be organised by an issue, not by the order in which papers were downloaded.
Summary and synthesis
| Topic | Source-by-source summary | Synthesis |
|---|---|---|
| Paragraph structure | Study A says… Study B says… Study C says… | The paragraph begins with a claim about a theme, then compares the evidence. |
| Writer's work | Reports each source separately. | Explains agreement, disagreement, method differences, and boundaries. |
| Research value | Shows that sources were read. | Shows what the body of evidence means for the review question. |
| Risk | Patchwork prose and borrowed structure. | Overgeneralisation if differences among studies are ignored. |
A Worked Example of Synthesis
The studies below are hypothetical; the example demonstrates paragraph structure, not a real empirical conclusion.
Do Not Manufacture a Research Gap
A gap is a conclusion from evidence, not a sentence added before the objectives.
| Gap type | What the review must show |
|---|---|
| Population or context | A relevant group or setting has not been adequately represented. |
| Method | Available studies rely on methods that cannot answer an important part of the question. |
| Theory | Findings exist, but the explanatory framework is weak, disputed, or untested. |
| Evidence | Results conflict, remain imprecise, or depend on low-quality data. |
| Time or policy | Earlier findings may not reflect a material change, and current evidence is insufficient. |
| Synthesis | Many isolated studies exist, but their collective pattern has not been examined responsibly. |
Use Reporting Guidelines When They Fit
A checklist supports transparency; it does not replace methodological judgment.
From matrix to final review
- 1
Step 1
Group sources by theme, method, debate, chronology, or another structure that answers the review question.
- 2
Step 2
Write a one-sentence claim for each planned paragraph before adding citations.
- 3
Step 3
Select sources because they support, qualify, or challenge that claim.
- 4
Step 4
Explain important differences in method, context, and evidence strength.
- 5
Step 5
End each section with what the evidence establishes, what remains uncertain, and why it matters.
- 6
Step 6
Audit every citation to confirm that the source supports the precise sentence attached to it.
- 7
Step 7
Review the full chapter for missing perspectives, outdated searches, overconfident language, and unattributed borrowed structure.
Local checklist
Final integrity check
Progress is stored only in this browser.
Questions about literature reviews
Can I use direct quotations in a literature review?
Yes, when exact wording is analytically important. Use quotation marks, cite accurately, and explain why the wording matters. Most research reviews should rely more on synthesis than on long quotations.
How many papers are enough?
There is no universal number. Sufficiency depends on the question, field, review type, date range, and whether additional searching still changes the main conclusions.
Can AI write a literature review from my sources?
AI may help organise notes or test an outline, but it can misrepresent papers, invent citations, and hide weak synthesis. The researcher must read the relevant sources, verify every claim, and take responsibility for the final argument.