How to write a literature review: a practical step-by-step method

The literature review is often the section that takes the most time in a thesis, and it is also one of the most poorly written when you do not know how to approach it. This practical guide gives you the concrete method, step by step, from the literature search through to analytical writing.

To understand the different forms a review can take and its fundamental role in a thesis, see our conceptual article: The literature review: role, forms and approaches.

Step 1: Define your topic precisely

Before opening a single database, explicitly define the boundaries of your literature search:

  • Keywords: identify the key terms of your topic in French AND English. Most scientific literature is in English: if you search only in French, you will miss the essentials.
  • Time period: typically the last 10-15 years for recent literature, plus the foundational texts (regardless of date). Do not set too strict a limit: an article from 1985 may still be central in your field.
  • Disciplines involved: does your topic touch on several disciplines? In education, you will often need cognitive psychology and sociology. In public health, medicine AND social sciences.
  • Geographic scope: are you looking for international studies, or only studies in a specific context (Quebec, francophone, American)?

Record these decisions in writing: you will justify them in your methodology chapter or in the presentation of your documentary approach.

Step 2: Choose the right databases for your discipline

Not all databases are equivalent. Here are the main ones by discipline:

General databases

  • Google Scholar: free access, all fields, very broad coverage. Useful for a first exploration, but full-text access is limited.
  • Scopus: an international commercial database, sciences and social sciences. Available through your university library.
  • Web of Science: the reference for natural sciences, engineering, health and certain social sciences. Journal impact factors available.

Francophone databases

  • Érudit: Quebec and Canadian francophone journals in the humanities and social sciences. Free access for member Quebec universities.
  • CAIRN: francophone humanities and social sciences (France, Belgium, Switzerland, Quebec). Very rich for theses in sociology, education and psychology.
  • Persée: archives of French journals in the humanities.

Specialized databases

  • ERIC: education and learning sciences
  • PsycINFO / APA PsycArticles: psychology and behavioural sciences
  • MEDLINE / PubMed: medicine and health sciences (free access through PubMed)
  • JSTOR: archives of academic journals, very useful for older texts in the humanities
  • IEEE Xplore: computing, electrical engineering, telecommunications
  • Business Source Complete (EBSCO): management, finance, marketing

Step 3: Build an effective search strategy

A good literature search strategy follows a precise logic:

1. Start with review articles and meta-analyses. These texts give you an overview of the field and already contain the key references. They save you weeks of reading.

2. Use Boolean operators. In most databases:

  • AND: both terms must be present (narrows the results)
  • OR: one term or the other (broadens the results)
  • NOT: excludes a term
  • Quotation marks for an exact phrase: "blended learning"

3. Trace the bibliographies. When you find a relevant foundational article, look at who cited it (the “cited by” function on Google Scholar and Scopus) and go through its own references. This is the technique known as “snowball searching”.

4. Document your process. Note the databases consulted, the keywords used, the number of results obtained, and the selection criteria applied. This documentation can be included in your methodology or in an appendix.

Step 4: Evaluate and select your sources

Not all texts are of equal value. To evaluate the quality of a source, use the CRAAP method:

  • Currency: is the date relevant to your topic?
  • Relevance: does the text address your research question?
  • Authority: who is the author? In which journal is it published? Is there a peer review process?
  • Accuracy: is the method described and justified? Are the claims supported?
  • Purpose: does the text inform, argue, or sell something?

Prioritize: articles in peer-reviewed journals, works published by university presses, recent theses and dissertations on closely related topics.

Avoid: non-academic websites, blog posts, reports without peer review (except for official data such as government statistics).

Step 5: Organize your sources before writing

Do not start writing before you have organized your sources. It is a step that takes time but makes the writing much smoother.

  • Create collections by theme
  • Annotate each article with your own comments (not just highlights)
  • Use tags to categorize by method, key findings, level of relevance
  • Export your references in APA or MLA with one click

With a spreadsheet

For a synthetic overview, create a table with the following columns:

| Author(s) | Year | Question/Objective | Method | Key findings | Limitations | Relevance to my thesis |

This table will give you a map of your literature before you write.

With Notion or Obsidian

These tools let you link notes together. Very useful for the thematic synthesis phase.

Step 6: Write analytically and synthetically

The golden rule: do not summarize, analyze and synthesize.

The concrete difference:

  • Summary: “Dupont (2018) studied the impact of X and found Y. Martin (2020) examined Z and concluded W. Jones (2022) analyzed…”
  • Synthesis: “Several studies converge on the idea that X improves Y in contexts of [conditions] (Dupont, 2018; Martin, 2020; Jones, 2022). However, Tremblay (2021) qualifies this result by showing that in rural settings, the effect reverses.”

To synthesize effectively:

  • Group together the authors who share similar positions, rather than presenting them one by one
  • Set in tension the contradictory perspectives: who says what, and why do the results differ?
  • Comment on the strengths and limitations of the studies (method, context, sample size)
  • Show the progression of knowledge: what does research understand better today than 20 years ago?

A useful formula for citing several authors together: “Several studies have shown that X (Dupont, 2018; Martin, 2020; Jones, 2022), although Tremblay (2021) qualifies this result by noting that in rural contexts…”

Step 7: Conclude on the gaps and your positioning

The final section of your review must do three things:

  1. Identify the gap: what the literature does not yet cover, or not in your specific context
  2. State how your research addresses this gap: how your study brings something new or different
  3. Announce your positioning: how you will use the existing work as support or as a starting point for your own analysis

This section makes the direct link between your review and your research question. Do not rush it: it is one of the passages your committee reads most carefully.

Example of a complete structure (education sciences)

Topic: The use of digital technologies in teaching mathematics at the secondary level

  1. Definitions and conceptual framing: digital technology in education, pedagogical integration (the operational definitions adopted)
  2. Overview of empirical research: studies on effectiveness, mixed results across national contexts
  3. Determining factors: teacher training, instructional design, access to resources
  4. Limitations of the literature: few studies in the Quebec francophone context, almost nothing on the secondary level in Quebec
  5. Positioning of our research: this study aims to fill this specific gap

Length and formatting

The literature review generally represents 20 to 30% of the body of the thesis. For a master’s thesis, expect 20 to 40 pages; for a doctorate, 40 to 80 pages.

It must follow your university’s standards: hierarchical heading styles, citations in APA or another style, line spacing, margins. Uniformat automatically applies these standards to your Word or LaTeX document, saving you hours of reformatting at the end of writing.

Try Uniformat for free →

Practical tips for moving quickly without sacrificing quality

  • Start early: the review often takes 3 to 6 months of reading and organizing
  • Skim first: abstract, introduction, conclusion, before reading everything in detail
  • Annotate as you read: note your own reflections, not just highlights
  • Plan for iterations: your review will be revised several times over the course of your research
  • Check your bibliography with Zotero before final deposit: reference errors cost points

Conclusion

A successful literature review is much more than a catalogue of summaries. It is the demonstration that you have understood the issues in your field, that you are able to analyze positions that are sometimes contradictory, and that your research is anchored in a real dialogue with the scientific community. Follow the method step by step, use the right tools, and invest the time it takes to build it with care.

For the formatting of your literature review and your entire thesis, visit uniformat.ca.