How to Write the Literature Review Section of Your Thesis or Paper

You have read the papers, taken the notes, and worked out how they fit together. Then you open a blank document to write the literature review section, and it will not come. What lands on the page is a list: Smith found this, then Jones found that, then Lee found something else. It is not wrong, exactly. It is just not a literature review. It is a summary of your reading wearing the title of one.

The reason is that writing the review is a different skill from doing it. Reading is about understanding the field. Writing is about making an argument out of it, on the page, in a shape a reader can follow. As a guide in PLOS Computational Biology⁠[1] puts it, "Reviewing the literature is not stamp collecting. A good review does not just summarize the literature, but discusses it critically, identifies methodological problems, and points out research gaps." Here is how to make that turn.

Writing it up is its own skill

The most common failure is structural, and it happens before you write a word. If you organize the section source by source, one paragraph per paper, you have guaranteed a list no matter how well each paragraph is written. The structure itself is doing the summarizing. Fixing the writing means fixing the structure first, so the section is built to carry an argument instead of a reading log.

The problem is common even at the doctoral level. When two researchers scored the literature reviews in a small sample of education dissertations from three universities, average scores were surprisingly low⁠[2], and the worst reviews were, in their words, "mere disjointed summaries of a haphazard collection of literature."

Organize by theme, not by paper

The alternative is to organize around ideas. George Mason University's writing center puts the rule simply: in a literature review, each section is focused on a topic, and the relevant sources are discussed within that topic⁠[3], rather than the other way around. Your themes become the sections. Within each one, several papers appear together because they speak to the same point, not because they sit next to each other in your notes.

This is the same move that synthesis is built on, carried into the writing. If you already grouped your reading by theme⁠[4], the outline of the section is done: one theme, one section. If you did not, grouping now is the first writing task, before any sentences get drafted.

Make every theme make a claim

Thematic sections can still read like lists if each one is just a bucket labeled with a topic. The fix is a small, powerful move the same George Mason guide recommends: turn each theme's label from a noun phrase into an assertion. "Sleep and memory" is a topic. "Sleep deprivation impairs memory consolidation, though studies disagree on the mechanism" is a claim, and a claim gives the section a job. The reader learns something from the heading alone, and every paragraph underneath has a clear point to support or complicate.

Carry the same principle down to your topic sentences. Open each paragraph with a statement about the literature, not about a single study. "Several studies link X to Y, but they measure Y in incompatible ways" is a topic sentence that promises synthesis. "Smith (2020) studied X" is a topic sentence that promises a summary. The first tells the reader what to make of the paragraph before they have read it.

Synthesize inside the paragraph

With a claim at the top of each paragraph, the sources become evidence for it. This is where synthesis actually shows up in the prose: within a single paragraph you bring several papers together, note where they agree, mark where they conflict, and use them to support the point you opened with. A synthesized paragraph might cite four studies in five sentences, not to march through them but to build one idea from all of them. The sources serve your sentence, rather than each getting a sentence of its own.

Doing this fluidly means moving constantly between your own argument and the citations that support it, so it helps to write where that is easy. In Researcher+'s notes you can drop a citation to the exact source and page at the cursor as you write, then render the whole reference list in APA, MLA, or Chicago at the end, which keeps the mechanics out of your way while you build the argument.

Two bookends finish the section. Open with a short passage that tells the reader what the review will argue and in what order, and close with one that states where the field stands and the gap your own work will fill⁠[5]. Everything between them should be doing the same job: not listing what you read, but using it to build a case only someone who did the reading could make. That is the difference between a literature review and a very long summary, and it is entirely a matter of how you write it.