How to Write a Literature Review: Structure and Sources

Most students meet the literature review as a chapter that has to be written before the “real” research begins. Many treat it as a list: one paragraph per paper, in the order they happened to read them. Examiners notice this at once.

A good literature review does something different. It tells the reader what is already known about a question, where researchers agree and disagree, and what is still missing. That missing piece is the gap your own study will try to fill. Done well, the review is the argument for why your research is worth doing.

This guide explains how to plan, organise and write a literature review for a thesis, dissertation or research paper. The examples lean towards agriculture and the sciences, but the method works in any discipline.

What a literature review is (and what it is not)

A literature review is a critical, organised discussion of published research on a defined topic. University writing centres describe it as a work of synthesis: you bring separate studies together to build a picture that no single study gives on its own.

It is not:

an annotated bibliography, where each source is summarised separately;
a chronological history of everything written on the subject;
a collection of quotations strung together.

The difference comes down to one question. After reading your review, does the reader understand the state of knowledge, or only what each author said?

Types of literature review

The word “review” covers several kinds of work. Knowing which one you are writing shapes how you search and how much you must report about your method.

Type	What it does	Typical use
Narrative (traditional) review	Discusses the literature selectively to frame a question	Thesis chapters, introductions to research papers
Systematic review	Follows a pre-defined, reproducible search and selection protocol	Answering a focused question, e.g. the effect of a treatment
Meta-analysis	Statistically combines results from comparable studies	Usually part of a systematic review
Scoping review	Maps the extent and type of evidence on a broad topic	New or scattered research areas

Most MSc and PhD chapters are narrative reviews. If you are writing a systematic review for publication, follow the PRISMA 2020 reporting guideline. It provides a 27-item checklist and flow diagrams for reporting how studies were identified, screened and included.

Step 1: Define the scope before you search

A review without boundaries never ends. Before opening a database, write down:

your research question or objectives, in one or two sentences;
the main concepts and their synonyms (for example, “zinc biofortification”, “zinc enrichment”, “micronutrient fertilisation”);
the time period, regions, crops, populations or study types you will include;
what you will leave out, and why.

These decisions can change as you read, but writing them down early stops you from drifting into side topics.

Step 2: Find reliable sources

Start with databases rather than a general web search. Search engines rank pages by popularity, not by scholarly quality.

Where to search

Source	What you will find
Google Scholar	Broad coverage across disciplines; useful for following citations forwards and backwards
Scopus and Web of Science	Curated, indexed journals with citation data (usually through institutional access)
PubMed	Biomedical and life sciences literature
AGRIS (FAO)	International agricultural research, including reports and grey literature
CeRA (ICAR)	Agricultural journals available to students and staff of Indian agricultural universities and ICAR institutes
Shodhganga (INFLIBNET)	Full-text PhD theses from Indian universities
Krishikosh	Theses, books and reports from the Indian agricultural research and education system
DOAJ	A directory of peer-reviewed open-access journals

Theses on Shodhganga and Krishikosh are especially useful. Their own literature review chapters can point you to sources you would otherwise miss, and they show what has already been studied in Indian conditions.

How to search effectively

Combine terms with AND, OR and NOT, and use quotation marks for exact phrases.
Use the reference lists of good papers to find older work, and the “cited by” link to find newer work.
Look for recent review articles first. They give you an overview and a ready list of key studies.
Keep a simple search log: database, search terms, date and number of results. You will need it if your examiner or a journal asks how you found your sources.

Judging quality

Not everything published is reliable. Prefer peer-reviewed journals, official reports and theses from recognised institutions. Be wary of journals that promise very fast publication, send unsolicited emails asking for papers, or have vague information about their editorial board and peer review. These are common signs of predatory journals. Tools such as the Think. Check. Submit. checklist help you assess a journal before you trust it, or submit to it.

Indian researchers should note that the UGC discontinued its CARE list of journals in February 2025. Universities now set their own criteria for judging journals, so check your institution’s current guidance.

Step 3: Read actively and take structured notes

You do not need to read every paper from start to finish. For each source, read the abstract, introduction and conclusion first. Then decide whether it deserves a full reading.

For the papers you keep, record the same details every time:

full citation;
research question or objective;
location, crop, population or sample;
methods and design;
main findings, with the key numbers;
limitations stated by the authors, and any you notice yourself;
how it relates to your own question.

Use a reference manager such as Zotero, Mendeley or EndNote from the first day. It stores your PDFs and notes, and it formats citations automatically, which saves hours at the end.

Step 4: Build a synthesis matrix

A synthesis matrix is a table that places sources in rows and themes in columns. Many university writing centres recommend it because it shows patterns that are hard to see in separate notes.

A simple example, with placeholder entries:

Source	Theme 1: Effect on yield	Theme 2: Effect on grain zinc	Theme 3: Method of application	Gaps or limitations
Author A (Year)	Finding	Finding	Soil application	One season only
Author B (Year)	Finding	Not studied	Foliar spray	Pot experiment
Author C (Year)	Not studied	Finding	Both	No economic analysis

Once the matrix fills up, read down each column. Where do studies agree? Where do they conflict, and could differences in method, soil or climate explain it? Which cells are empty? Empty cells are often where your research gap lies.

Step 5: Choose a structure

Like most academic writing, a literature review has an introduction, a body and a conclusion. The body is where you have real choices.

Introduction

State the topic, explain why it matters, define the scope of the review and tell the reader how the chapter is organised.

Body: four common ways to organise

Thematic. Group studies by theme or sub-question. This is the most common and usually the strongest choice, because it forces you to compare studies.
Methodological. Group studies by approach, for example field trials, laboratory studies and modelling. Useful when methods strongly affect results.
Chronological. Trace how understanding developed over time. Use it only when the history itself matters, as it easily becomes a list.
Theoretical. Compare the theories or models used to explain a problem. Common in social sciences and extension research.

You can combine these, for example a thematic structure with a short chronological overview in the opening section.

Conclusion

Summarise what is known, state clearly what remains unknown or contested, and explain how your study addresses that gap. This paragraph should lead naturally into your objectives or methods chapter.

Step 6: Write in synthesis, not summary

The most useful habit is to start paragraphs with a point, not with an author’s name.

Summary style (weak):

Author A (Year) studied zinc application in wheat and found an increase in yield. Author B (Year) studied foliar zinc and found higher grain zinc. Author C (Year) compared methods.

Synthesis style (stronger):

Studies agree that zinc application can raise grain zinc concentration, but its effect on yield is less consistent (Author A, Year; Author C, Year). Foliar application appears more effective for grain enrichment than soil application (Author B, Year; Author C, Year), although most trials ran for a single season, which limits conclusions about long-term soil effects.

The second version makes a claim, groups evidence behind it, notes a disagreement and identifies a limitation. Each paragraph in your review should do some of those things.

Other practical points:

Paraphrase in your own words and cite every idea you take from a source. Quote directly only when the exact wording matters.
Be fair to studies that disagree with your expectations. Ignoring them weakens your argument.
Use reporting verbs carefully. “Found”, “reported” and “suggested” imply different levels of certainty.
Follow the citation style your university or target journal requires, and use it consistently.

Common mistakes and how to avoid them

Mistake	How to fix it
Describing studies one by one	Organise by theme and compare studies within each paragraph
No clear research gap	End each section, and the chapter, with what remains unknown
Relying on old sources only	Include recent work alongside foundational studies, and explain any gap in dates
Too broad a scope	Return to your written scope and cut material that does not serve your question
Citing papers you have not read	Read and verify every source you cite; do not copy citations from other reviews
Uncritical acceptance of findings	Comment on sample size, design, location and limitations
Using unreliable or predatory sources	Check journal credentials before citing
Too many direct quotations	Paraphrase and synthesise; keep quotations short and rare
Plagiarism, including patchwriting	Write from your notes, not from the open paper, and run a similarity check before submission
Leaving formatting to the end	Use a reference manager from the start

Using AI tools responsibly

AI tools can help you brainstorm search terms or check grammar. They should not write your review or supply your references. Such tools can produce citations that look real but do not exist, and they cannot judge the quality of a study in the way you can. Check your university’s policy on AI use, verify every source yourself, and disclose AI assistance where your institution or journal requires it.

A short checklist before you submit

The introduction states the scope and how the review is organised.
Each section is built around a theme or question, not an author.
Agreements, disagreements and limitations are discussed.
The research gap is stated clearly and links to your objectives.
Every source cited has been read and appears in the reference list.
Citations follow one style consistently.
The text has been checked for similarity and proofread.

Conclusion

A literature review is the place where you show that you understand your field well enough to add to it. Define your scope, search reliable databases, take consistent notes, and organise your sources by theme with a synthesis matrix. Then write paragraphs that compare and evaluate studies rather than summarise them, and finish with a clear statement of the gap your research will address. That chapter will be far easier to write, and far more convincing to read.