How to Write a Literature Review for a Final-Year Project

To write a literature review, define the question, search for relevant sources, read and compare them, then organise a synthesis around the ideas that answer your question. Explain agreements, differences and limitations, and support any research-gap claim with the evidence you checked. A review develops an argument from sources rather than listing their summaries.

In a final-year project, this may form Chapter 2 or a related-work section. Your supervisor and confirmed departmental brief determine the headings, scope and referencing requirements. Use the search log and review matrix below to keep the evidence behind your discussion traceable.

If you still need publications, UniDraft’s citation finder can help locate sources to check. Read the full works and verify that they support your statements; a search result or generated reference is not a substitute for that review.

Your source notes should become a connected discussion you can defend. UniDraft helps organise the publications, comparisons and project instructions you confirm into an editable document. Start your project document, then check each citation and claim against the works you read. See the final-year project workflow; a polished paragraph does not verify a source.

The short answer

• A literature review is a discussion organised around ideas, not a list of summaries. Bring sources into conversation when they address the same question; do not force a fixed number into every paragraph.
• One possible outline is: conceptual review, theoretical framework, empirical review, summary and gap. Confirm the exact headings with your department and supervisor first.
• Group studies by theme (usually your variables or research questions), not by author or year.
• Explain the supported gap or unresolved question, including the limits of your search, and connect it to your research questions.
• Cite only sources you have actually read and can find again. Never cite from memory.

What a literature review is actually for

It is not a list of summaries. A literature review's job is to show that you have read widely enough to place your own study accurately: what is already known, what methods have been used, where researchers disagree, and what is still missing. That last part is what justifies your project existing at all.

This is also why a literature review is not the same as an annotated bibliography. An annotated bibliography treats each source separately, one paragraph per source. A literature review integrates several sources into a single discussion, organised around a theme or concept rather than around individual authors. University writing centres draw the same line: the UNC Writing Center's guide to literature reviews describes a review as usually organised around ideas, not around the sources themselves.

A useful test: cover the author names in a paragraph. If the paragraph still makes an argument, it is synthesis. If it collapses into nothing, it was a list.

The standard structure (and why yours might differ)

A possible four-part outline, where your assignment calls for it, is:

1. Conceptual review: defining your key terms and variables, drawing on how different scholars have used them. If your topic involves "financial inclusion" or "academic performance," this is where you establish what those terms mean in your study specifically.
2. Theoretical framework: the established theory or theories your study is grounded in (for example the Technology Acceptance Model or Diffusion of Innovation Theory). This explains why you would expect the relationships you are studying to exist.
3. Empirical review: what previous studies actually found when they investigated similar questions, organised around themes rather than one study at a time.
4. Summary and gap: a short closing section that pulls the chapter together and states, explicitly, what your study adds that the reviewed literature does not already cover.

This is a common pattern, not a rule set by NUC, ITF or any single regulator. These conventions are general academic practice and they vary by department. Some departments use "Related Works" instead of "Literature Review" (common in Computer Science). Some want conceptual, theoretical and empirical review as separate numbered subsections; others want them woven together. Check your department's project guideline and your supervisor's preference before you commit to a structure. Restructuring a finished Chapter 2 costs far more time than confirming the expected headings first. Your university's format guide is a good place to start; the free format guides cover available Nigerian institutions.

How the four parts differ in practice

Students often blur these parts together. This table shows what each one is really answering.

Part	Question it answers	What goes in	Typical mistake
Conceptual review	What do the key terms mean?	Definitions from several scholars, and the definition you will use	Copying one dictionary definition
Theoretical framework	Why should these variables be related?	One or two theories, with what each predicts	Naming a theory and never using it again
Empirical review	What have others found?	Findings, methods and settings, grouped by theme	Author-by-author summaries
Summary and gap	What is still missing?	A short synthesis and one gap sentence	"More research is needed"

If the framework sections are still unclear, read conceptual vs theoretical framework before drafting them.

Before you write: searching and reading efficiently

A weak literature review often starts as a weak search. Spend the first stretch of Chapter 2 work searching deliberately, not reading at random.

Build a search log

1. List your keywords. Take every variable and setting in your title and write two or three alternative terms for each ("mobile banking", "mobile money", "fintech adoption").
2. Search in layers. Combine keywords with a place or population ("students", "polytechnic", "Nigeria"), then remove the place to widen the results.
3. Record every search: the database or site, the exact words, the date, and how many useful hits you got. Those notes help explain how the sources were found, and you will not repeat searches you forgot.
4. Follow the references. When a strong recent paper cites something important, find that original paper too.

The free citation finder helps with the checkable-source part: it locates real academic sources and formats them in APA, IEEE or Harvard style. Your university library's e-resources, if it subscribes to any, are worth asking about at the library desk.

For a worked concept-to-keyword search, use how to find research papers. Once a source looks relevant, the research-paper reading worksheet helps separate its actual finding from your interpretation. Keep the full reference, a passage or table locator and a limit beside every important reading note. Search notes and extraction notes serve different purposes: the first explains how you found the work; the second explains why it supports your argument.

Read in three passes

You do not need to read every paper start to finish. Read the abstract to decide if it is relevant, then the methods and findings for the parts you will actually use, and only then the introduction or discussion if you need the argument. Then capture what you learned immediately, in a matrix.

Use a synthesis matrix

A synthesis matrix is a simple table (a Word table or spreadsheet) with one row per source. The columns are the things you will compare: setting, method, sample, main finding, and the theme it belongs to. The matrix makes those comparisons visible.

The matrix does two jobs. Reading down a column shows you which studies belong in the same paragraph. Spotting what your collected sources do not cover suggests a possible gap. Search further before claiming an absence: an incomplete matrix can reflect an incomplete search.

Thematic vs. chronological organisation

Within the empirical review especially, you have to decide how to arrange the studies you have found. Two common approaches:

	Thematic	Chronological
Organised by	Concept, variable, or sub-topic	Publication date / order of development
Best for	Comparing findings, spotting patterns and contradictions across studies, identifying a gap	Showing how thinking on a topic evolved over time
Typical use	Most undergraduate project literature reviews	Fields where the research clearly builds in stages (a specific technology's development, a policy's history)
Risk if done badly	Themes that do not map to your actual research questions	Turns into "Author A did X. Then Author B did Y." -- a list, not an analysis

For most Nigerian undergraduate projects, thematic organisation works better. It forces you to group studies around the variables your own project is investigating, which makes the empirical review connect naturally to your methodology and results chapters later. The UNC guide above also lists a third option, methodological organisation (grouping by how studies were done), which is useful when your gap is a method gap. You can still note publication years and order studies within a theme roughly chronologically.

Choosing your themes

Start from your research questions. If you have three objectives, you will probably have three to five themes: one for each variable or relationship, plus perhaps one on context or measurement. If a theme does not connect to any objective, either your objectives are incomplete or that theme belongs in a sentence, not a subsection. If you have not fixed your objectives, see how to write strong research questions and objectives.

A weak paragraph vs. a strong one

These two paragraphs use fictional placeholder citations to demonstrate structure. They are not evidence and must not be copied into a project as real references. For a comparison using identifiable published studies, see the empirical review worked example.

Weak (summary-by-summary, no synthesis):

Many studies have looked at mobile banking adoption. Okafor (2021) studied mobile banking in Lagos. Bello (2022) also studied mobile banking adoption among students. Both found that mobile banking is increasing in Nigeria.

This paragraph tells a reader almost nothing specific. It does not say what each study actually found, whether they agree, or why it matters to your project.

Strong (synthesised, specific, pointing toward a gap):

Okafor (2023) found that trust in fintech platforms was the strongest predictor of mobile banking adoption among urban university students, while Bello (2024), studying a comparable population in a semi-urban context, found that network reliability outweighed trust as a barrier. This divergence suggests that adoption drivers may vary by infrastructure context rather than being uniform across student populations -- a distinction few studies have tested directly within a single Nigerian state, which is the gap this study addresses.

The strong version names what each study found, puts them in conversation with each other, and uses the disagreement to justify the current project. (The citations above are illustrative placeholders. Always cite real sources you have actually read, never invented ones.)

A repeatable paragraph pattern

Strong empirical paragraphs tend to follow four moves. Use this as a drafting scaffold, then vary it so the chapter does not sound mechanical.

1. Claim: one sentence stating the point of the paragraph ("Evidence on what drives adoption is split between trust and infrastructure").
2. Evidence: two or three studies, with what each found and in what setting.
3. Comparison: how they agree, differ, or use different methods that might explain the difference.
4. Link: one sentence connecting the point to your study or the next theme.

Useful sentence starters

Synthesis needs verbs beyond "said". Compare "Okafor (2023) reported... whereas Bello (2024) found...", "Both studies converge on...", "This contrasts with...", "A possible explanation for the difference is...", and "None of these studies tested...". Reporting verbs carry meaning: "argued" signals a claim, "found" signals evidence, and "suggested" signals caution. Pick the one the source has earned.

Conceptual review: a short worked example

For a project on "digital literacy", a weak conceptual paragraph copies one definition. A stronger one compares two, then commits:

Some scholars define digital literacy narrowly as the technical skill of operating devices and software, while others extend it to include the ability to judge and create online information. This study adopts the broader definition because its research questions concern how students evaluate online sources, not only whether they can use them.

The last sentence serves a useful purpose: it tells the reader which definition governs the rest of your project, so the terms stay consistent through Chapters 3 and 4. (The example is illustrative and no real author is being cited.)

Theoretical framework: use it, do not just name it

Explain how the chosen theory or concepts guide the question, evidence and interpretation. Read the original model before attributing a prediction to it. A qualitative framework may guide interpretation rather than a statistical test; use the framework guide and connect the framework to the discussion where appropriate.

Finding your own study's gap

Name the unresolved issue rather than jumping directly from source summaries to the decision to conduct your study.

A real gap is specific and falls into one of four patterns:

1. Population/context gap: the studies you reviewed covered a different group or setting than yours (for example, all studied private universities; none covered polytechnics or your specific state).
2. Method gap: existing studies used one method (all qualitative, say) and a different method might reveal something they could not.
3. Variable/relationship gap: a specific relationship between two things has not been tested, even though each thing has been studied separately.
4. Contradiction gap: studies disagree, and nobody has tried to reconcile why.

Watch for these signal phrases in the papers you are reading, because they usually mark exactly where a gap is hiding: "further research is needed on...", "this study did not examine...", "limited to [a specific context]...". Those sentences, written by other researchers about their own work, are often the clearest pointer to your gap. Check the limitations and suggestions-for-future-research sections of your best sources first.

Write your gap as one direct sentence naming the specific prior work and what it left out, not "more research is needed in this area," which says nothing. Compare: "more research is needed on fintech adoption" (says nothing) versus "no reviewed study has tested whether network reliability, rather than trust, is the dominant adoption barrier for students outside major urban centres -- a gap this study addresses" (specific, and it sets up your research questions).

Testing whether your gap is real

Before you commit to a gap, run three checks:

• Is it actually unfilled? You can only say "no reviewed study" honestly. Say "in the studies reviewed" rather than "nobody has ever", because you cannot prove a negative across all literature.
• Does it matter? A gap nobody needs filled is just a difference. Say who benefits: a bank, a school, a ministry, a future student.
• Can you fill it? A national gap cannot be answered by a survey of 60 students. Match the gap to what your sample, time and budget can support. If it is too large, narrow your claim; the problem statement guide shows how the same gap should appear in Chapter 1.

If you are still shopping for a topic where a gap is easy to see, the free topic generator and how to choose a final year project topic can help before you invest weeks in reading.

How the review varies by discipline

The four-part layout above suits social science, management and education projects best. Other departments bend it.

Field	What Chapter 2 often emphasises	Watch out for
Social sciences, education, management	Concepts, theory, empirical studies by theme	Missing theoretical framework
Computer Science / IT	"Related Works": comparing existing systems, algorithms or tools	Listing tools without comparing features or limits
Engineering	Prior designs, standards, and technical specifications	Textbook definitions instead of comparable designs
Health and nursing	Clinical or public health evidence, guidelines, prevalence data	Old statistics and unverified figures
Sciences	Earlier experiments and results, methods used	Ignoring conflicting results

In a Computer Science project, a good related-works section ends with a comparison table (system, what it does, what it lacks) and then one sentence on what yours adds. If your department expects a specific layout, the guides under /format or your department's handbook take priority over this table.

Writing the summary and gap section

Keep this section short, usually one to three paragraphs. Restate the main themes in a sentence each, say where the evidence agrees and where it is thin, then state the gap. End by pointing forward: the gap leads to your research questions, and your research design in Chapter 3 is your answer to "how do I address this gap?" If your gap sentence and research questions do not point at the same thing, one of them needs to change.

How long should Chapter 2 be, and how many sources?

There is no national rule. Length and reference expectations are set by departments and supervisors, and they differ, so ask for the department's guideline or a recent approved project from your department to see what is normal there. Do not pad to hit a count. Twenty relevant sources organised into four clear themes will read better than sixty that were grabbed to look thorough. Use the sources necessary to examine each theme accurately, including current evidence where relevant; there is no fixed per-theme count. Whatever you cite must appear in your reference list in the required style; see how to write the references list and the APA 7 in-text citation guide.

Paraphrasing, quoting and secondary sources

• Paraphrase, do not patch. Changing a few words of a source's sentence is still copying. Read, close the paper, write the point in your own words, then cite it.
• Quote rarely. Use direct quotes only for a definition or a striking phrase, with quotation marks and a page number where your style requires it.
• Avoid secondary citing. If you only saw Author A described inside Author B's paper, you have not read A. Find the original if you can; if you truly cannot, follow your referencing style's rule for secondary sources and be open about it.
• Similarity software. Many institutions run similarity checks. The way to pass one honestly is genuine synthesis and correct citation, not rewording tricks. See why reports read like AI wrote them for the real issue.

Where to look for real sources

Use peer-reviewed journal articles and academic books first, then reports from credible bodies (regulators such as the NUC, ministries, the World Bank, WHO), then well-sourced theses in your department's library. Be cautious with blogs and anonymous "project material" pages: they often reproduce old work without verification and cannot be traced back. Write down enough detail as you read (authors, year, title, journal or publisher, DOI or link) so you never have to hunt for a reference at 2 a.m. the night before submission.

Common mistakes and how supervisors flag them

What the supervisor writes	What it usually means	Fix
"This is descriptive"	You summarised without comparing	Rewrite each paragraph with the claim, evidence, comparison, link pattern
"Where is your gap?"	Gap missing or vague	Add a named-work gap sentence in the summary
"Outdated"	Most sources are old	Add recent studies; keep older ones for classic theory only
"Not relevant"	Sources do not match your variables	Cut them or explain why they still matter
"Check your references"	In-text citations and list do not match	Cross-check every citation against the list
"Too Western"	No Nigerian or African evidence	Search for local studies, or state honestly that they are scarce

What to do when things go wrong

• You find almost nothing on your topic. Widen in steps as described in the FAQ above, and say so in the chapter. Do not fill space with unrelated sources.
• You find too much. Narrow by date, setting or the variable you care about most, and let the matrix decide what is essential.
• Your supervisor rejects your framework. Ask which theory they would accept and why; then rewrite that section only, since your empirical themes may still stand.
• You are short of time. Do the matrix, then draft the empirical review by theme, and write the conceptual review last. See the step-by-step writing order.

A realistic timeline

One optional way to spread the work is described below. It assumes roughly eight weeks; compress or stretch it to your deadline, but keep the order.

A 5-point checklist before you call Chapter 2 done

Does your literature review actually:

1. Group sources by theme or concept, not one paragraph per author?
2. Name what each study found, specifically, rather than vague claims like "many studies show"?
3. Put studies in conversation with each other: agreement, disagreement, or a pattern across several?
4. State your theoretical or conceptual framework explicitly, not just imply it?
5. End with a specific, nameable gap that your research questions directly address?

And a few final checks:

6. Every in-text citation appears in the reference list, and the reverse.
7. The source dates are appropriate to the question, including current evidence and relevant older work.
8. Terms are used the same way you defined them.
9. Headings and spacing follow your department's rules (the free formatting checker tests a .docx's fonts, spacing and margins against your university's requirements), and the general layout advice in how to format a final year project applies.

If any of these is missing, that is usually the fastest fix available before a supervisor flags it.

How this connects to the rest of your project

A gap statement you cannot defend usually traces back to research questions that were never specific enough to begin with. Once your literature review and gap are settled, Chapter 3 should follow logically from the gap you just identified. If you have not finished Chapter 1, how to write the introduction covers where Chapter 2 fits. In your defence, panels often ask why you chose your theory and how you found your sources; the common defence questions post shows how to answer.

Check a research gap before claiming it

A blank cell in your matrix is a reason to investigate, not proof that nobody has studied the subject. You may have missed a synonym, database or related discipline. Keep a short gap-verification record next to the matrix.

The research-gap workflow works through alternative searches, a fictional claim revision and the checks needed before proposing a contribution. Use it for the gap decision; this page remains the guide to organising the full literature review.

Question	Evidence to record
What appears missing?	A specific population, comparison, method or unresolved result
Where did I search?	Databases, search phrases, dates and alternative terms
What relevant work did I find?	Identifiable studies and the limits of each
Why does the difference matter?	A reason tied to your question, not just a different town name
What can I reasonably claim?	A bounded statement about the work reviewed
What can I not claim?	An exhaustive absence or novelty you have not demonstrated

Prefer "The studies reviewed here did not examine X in Y setting" to "No researcher has ever studied X." Then explain why examining that setting would answer a meaningful question. Your supervisor may still decide the proposed contribution is too small; that is a reason to revise the question, not exaggerate novelty.

For an example grounded in identifiable publications, use the empirical review worked guide. If your own approved project uses qualitative interviews, the thematic analysis guide explains one analysis approach; it is not a method you should add simply to make a gap look different. Take the gap, objectives and proposed method together into your proposal defence.

If your approved assignment is a systematic review requiring PRISMA reporting, use the PRISMA count-checking guide. It distinguishes records, reports and studies and shows a fictional accounting exercise. A normal search log or narrative review does not become systematic merely by adding a flow diagram.

If you are planning a Chemistry seminar review, use the Chemistry topic and evidence worksheet to define the process comparison, measures and boundaries before organising the literature. It helps distinguish a broad sustainability claim from a question your sources can answer.

For a Physics seminar about measurement evidence, the Physics topic and source worksheet helps check what each publication measured and what its reporting supports before combining sources into an argument.

For a Mass Communication seminar about news evidence, the topic and evidence worksheet helps define the question and material before choosing a method. A textual comparison should not quietly become a claim about what an audience believed.

Where UniDraft fits in

If your assignment is about interpreting a creative work, use the Literature seminar question and passage worksheet to plan that analysis. A textual interpretation and a review of academic research have different evidence tasks, although your report may need both.

Check the current price and included correction allowance in the document workflow before payment.

Start your document →

Sources

• UNC Writing Center: Literature Reviews (university guide on organising a review around ideas, and on thematic, chronological and methodological organisation)
• National Universities Commission (NUC) (the regulator's site, for accreditation results and guideline documents; the Chapter 2 conventions above are general academic practice, not an NUC rule)