How to Write a Literature Review: A Step-by-Step Guide

A literature review surveys the existing research on a topic and explains how it fits together. It appears as a chapter in theses and dissertations, as a section in research papers, and sometimes as a standalone assignment. A strong review does more than summarise: it identifies themes, debates and gaps, and shows why your own question matters.

Step 1: Define your scope

Start with a focused question. "Social media and mental health" is too broad for most assignments. "Social media use and sleep quality among university students" is manageable. Decide on boundaries: time period, types of study, populations, languages and disciplines.

Step 2: Search systematically

• Use academic databases relevant to your field, available through your library, as well as Google Scholar.
• Build a search string from key terms and synonyms, combined with AND, OR and quotation marks.
• Follow citation trails: check the reference lists of good papers, and look at who has cited them since.
• Record your searches: database, search terms, date and number of results. This makes your method transparent and repeatable.

Step 3: Manage your sources

Use a reference manager such as Zotero (free and open source) or the tool your institution supports. Save PDFs, tag papers by theme and generate citations automatically. Setting this up at the start saves hours at the end.

Step 4: Read efficiently

You do not need to read every paper word for word. Use a staged approach:

1. Screen: title and abstract. Is it relevant?
2. Skim: introduction, headings, figures and conclusion. What did they find?
3. Read closely only the papers central to your question, paying attention to methods and limitations.

For each important paper, record: research question, method, sample, key findings, limitations, and how it relates to other papers.

Step 5: Synthesise, do not summarise

This is the step that separates a strong review from a weak one. Instead of describing papers one by one, organise them by idea. A synthesis matrix helps: a table with papers as rows and themes or questions as columns. Patterns become visible:

• Where do studies agree?
• Where do they disagree, and why? Different methods, populations or definitions?
• How has thinking changed over time?
• What has not been studied, or not studied well?

Step 6: Choose a structure

• Thematic: sections organised by theme. The most common and usually the strongest.
• Chronological: how the field developed over time. Useful when that history matters.
• Methodological: grouped by research approach, useful when methods drive different conclusions.

Whatever the structure, include an introduction that explains scope and approach, and a conclusion that summarises the state of knowledge and identifies the gap your work addresses.

Step 7: Write with your own voice

Use topic sentences that make claims about the literature: "Studies using objective sleep measures consistently report weaker associations than those relying on self-report." Then support the claim with citations. Your analysis should lead each paragraph, with sources as evidence.

Common mistakes

• A chain of summaries: "Smith (2019) found... Jones (2020) found..."
• Relying on a handful of sources or only very old ones
• Ignoring studies that contradict your expectations
• Quoting too much instead of paraphrasing and analysing
• Inconsistent citation style

Make the process manageable

Break the work into focused sessions using the Pomodoro technique: one block for searching, several for reading, then dedicated blocks for building the synthesis matrix before you start drafting. If you are unsure what your institution expects, ask your supervisor for an example of a strong review in your department.