How to Write a Literature Review: A UK Student's Complete Guide

How to write a literature review that demonstrates critical thinking

A critical literature review does not simply describe what each source says. It evaluates the strengths and weaknesses of existing research, identifies patterns and contradictions, and positions your own study within the broader academic conversation. The University of Sheffield defines it as an account of the current thinking in a specific area of study, whose purpose is to introduce the reader to what has gone before and provide a foundation for your own research.

The difference between a descriptive review and a critical one is the difference between reporting and arguing. Descriptive writing tells the reader what Smith (2022) found and what Jones (2023) concluded. Critical writing evaluates whether Smith's methodology was robust, whether Jones's findings contradict earlier work, and what the combined evidence actually tells us about the research problem.

This distinction matters across every subject. Whether you are writing about leadership theory in a business dissertation, patient outcomes in a nursing project or policy effectiveness in criminology, the expectation is the same: analyse, do not describe.

Choosing your literature review structure

There is no single correct way to organise a literature review, but three approaches dominate in UK universities. The right choice depends on your discipline, your research questions and the nature of your sources.

Thematic structure. The most widely used structure in UK social science, business and humanities dissertations. You organise the review around key themes, concepts or debates rather than listing sources one by one, drawing on multiple sources within each theme to build an argument. Thematic organisation forces synthesis, which is exactly what UK markers reward.

Chronological structure. Traces how understanding of a topic has developed over time. It works well when your research sits in a field with clear historical development, such as education policy reform or clinical practice in health and social care. The risk is that it becomes a descriptive timeline, so explain why each shift happened and what it means for your research.

Methodological structure. Organises sources by the research methods they used: qualitative, quantitative or mixed-methods. This is useful when your own methodology is a key contribution. Students working on a research proposal often draft a short methodological review to justify their chosen approach.

Combining approaches. Many strong reviews blend these structures, organised thematically but discussing the chronological development of ideas within each theme. The key is that your structure serves your argument, not the other way around.

Creating a literature review outline

Before you write, map your chapter. A literature review outline is a skeleton that shows your planned sections, sub-sections and which sources sit under each heading. Without one, literature reviews tend to sprawl. A practical outline for a 3,000-word chapter might look like this.

Notice that the outline ends with the gap. Your literature review should funnel towards the justification for your own study; the gap you identify here should align directly with the aim stated in your dissertation introduction.

• Introduction (about 200 words): define the scope, state your search strategy, preview the themes.
• Theme 1 (about 800 words): key sources, areas of agreement, points of debate.
• Theme 2 (about 800 words): key sources, methodological strengths and weaknesses, contradictions.
• Theme 3 (about 700 words): emerging perspectives and under-researched areas.
• Research gap and summary (about 500 words): synthesise the themes, state the gap your study addresses, link to your aim and objectives.

Using a synthesis matrix

A synthesis matrix is a table that organises your sources by theme rather than by author. It is one of the most effective tools for moving from summary to synthesis, yet surprisingly few UK students use one.

Build it as a spreadsheet with one row per source and one column per theme, plus columns for methods and key limitations. Reading across a row tells you what one source contributes; reading down a column tells you what the literature collectively says about one theme. The column view is where synthesis happens, because you can see agreements, contradictions and gaps at a glance.

Populate the matrix as you read. By the time you have processed 15 to 20 sources, the themes and gaps will be visible without forcing them. Students working on statistical analysis projects may also find the matrix useful for comparing methodological approaches across quantitative studies.

Literature review example: descriptive vs critical

The same source can be handled two ways.

Descriptive (weak): 'Brown (2023) conducted a study on student engagement in online learning. He found that students who attended live sessions performed better than those who watched recordings. He used a sample of 200 undergraduate students at a UK university.'

Critical (strong): 'Brown (2023) found a positive association between live-session attendance and academic performance in a sample of 200 UK undergraduates, supporting earlier findings by Garcia (2021). However, the study did not control for prior attainment or motivation, which limits the causal claim. Patel (2024), using a larger mixed-methods design, identified motivation as a significant mediating factor, suggesting attendance alone is an insufficient predictor of engagement.'

The critical version evaluates methodology, connects the source to other literature, and identifies what the evidence actually supports. This is the standard expected at UK universities, whether you are studying law, psychology or engineering.

How many sources should a literature review include?

There is no universal rule, but these benchmarks are widely used in UK institutions.

Quality matters more than quantity: ten well-analysed, synthesised sources are worth more than 30 that are merely listed. Aim for at least 70 percent of your sources from the past five to ten years, with the exception of seminal works. Use peer-reviewed journal articles and academic books as your primary sources; government and professional-body publications (CIPD, NMC or BPS) suit applied subjects such as CIPD assignments and HND projects. Avoid blogs and unverified websites where specialist sources exist.

• Undergraduate dissertation (8,000 to 10,000 words): 20 to 30 sources.
• Master's dissertation (12,000 to 20,000 words): 30 to 50 sources.
• PhD thesis: 50 or more, depending on the breadth of the field.

Common mistakes to avoid

A strong literature review is as much about avoiding these traps as hitting the word count.

• Source-by-source structure: one paragraph per source with no connection between them. Organise by themes, not authors.
• Insufficient critical analysis: describing what authors found without evaluating how they found it.
• No clear gap: if the review does not identify a specific gap your study addresses, it has not done its job.
• Over-quoting: excessive direct quotation signals a lack of understanding. Paraphrase and synthesise.
• Ignoring contradictions: presenting only supportive sources weakens your credibility.
• No search strategy: markers expect a brief account of databases, search terms and inclusion criteria.
• Repetition from the introduction: the introduction gives brief context; the review provides depth.

When to write your literature review

Draft your outline after your initial reading but before finalising your methodology, so the review can inform your methodological choices rather than the other way around. If previous qualitative studies on your topic produced thin findings, that might justify a quantitative or mixed-methods approach.

Revisit and refine the chapter throughout the writing process. New sources will emerge, and your understanding of the themes will sharpen as you analyse your own data. For essay writing, a case study or a reflective report, the same principle applies: read, synthesise, then position your own argument.