What Is a Literature Review? Purpose, Types, and How to Write One

A literature review is a structured, critical evaluation of existing research on a defined topic. It goes beyond listing what has been published. A good literature review maps the current state of knowledge, identifies where findings agree or conflict, and pinpoints the gaps that justify further investigation. Whether it sits inside a dissertation chapter, a journal article, or a grant proposal, the review is the foundation that connects a new study to the broader scholarly conversation.

Every research discipline relies on literature reviews. In medicine, they inform clinical guidelines and treatment protocols. In the social sciences, they frame policy debates and theoretical models. In engineering and computer science, they benchmark existing solutions before proposing new ones. The format varies, but the purpose is consistent: to evaluate what is known so that new research asks the right questions and avoids repeating work that has already been done.

The volume of published research has grown rapidly. One analysis of PROSPERO-registered reviews found that systematic reviews alone take an average of 67 weeks to complete, with screening and synthesis consuming the largest share of that time [3]. As the evidence base expands, so does the importance of structured, well executed literature reviews that help researchers and practitioners make sense of what is known and what still needs investigation [9]. AI-powered search and screening tools are beginning to reduce the manual workload, but the intellectual core of a literature review still depends on human judgment: deciding which sources matter, evaluating their quality, and weaving them into a coherent argument that advances understanding.

This guide covers what a literature review is, why it matters, the main types (narrative, systematic, scoping, and integrative), how to structure and write one, common mistakes to avoid, and how AI tools are supporting the workflow.

Why Literature Reviews Matter in Research

Literature reviews do more than satisfy a formatting requirement. They serve four distinct purposes that make them essential to credible research.

They establish context. Before readers can evaluate new findings, they need to understand what has already been discovered about a topic. The literature review provides that context by summarizing existing knowledge, tracing how understanding has evolved, and identifying the theories and frameworks that shape current thinking. A study without this grounding reads as disconnected from its own field.

They reveal gaps. By mapping what has been studied, a literature review also exposes what has not. These gaps are the raw material for new research questions. A well constructed review makes a clear case that the researcher's question matters because prior work has not adequately addressed it. Funding bodies and dissertation committees look for exactly this kind of justified positioning.

They prevent duplication. Researchers who skip the literature review risk repeating experiments that have already been run, missing findings that would strengthen their methodology, or arriving at conclusions that contradict established evidence without realizing it. In clinical research, this can mean wasting limited resources on trials that answer questions the field has already settled [3].

They demonstrate expertise. In dissertations and grant proposals, the literature review signals that the researcher understands the field well enough to contribute to it. Examiners, peer reviewers, and funding panels use it to assess whether the applicant has engaged seriously with prior work.

Types of Literature Reviews

The term "literature review" covers several distinct approaches. Each has its own methodology, purpose, and conventions. Choosing the right type depends on the research question, the available evidence, and the level of rigor required.

Traditional (Narrative) Literature Review

A narrative review is an expert-driven synthesis. The author selects and interprets sources based on their knowledge of the field, organizes findings by themes or theoretical frameworks, and draws interpretive conclusions about the state of knowledge.

High-impact journals regularly commission narrative reviews from recognized experts. Publications like Nature Reviews, Annual Review of Psychology, and Lancet specialty journals use them to survey entire fields. The strength of the format is interpretive depth: an experienced researcher can connect findings across disciplines, identify emerging patterns, and propose new conceptual frameworks that structured review methods would not surface.

The trade-off is reproducibility. Another expert might select different studies and reach different conclusions. The SANRA (Scale for the Assessment of Narrative Review Articles) provides a quality framework with six criteria, but its use remains optional [2]. For topics where a thorough, reproducible evidence base is required, a systematic review is the stronger choice.

Systematic Review

A systematic review follows a rigorous, pre-defined protocol to identify, screen, appraise, and synthesize all available evidence on a focused research question. Every step is documented and reproducible.

Key characteristics include:

• A registered protocol (on PROSPERO or OSF) published before the review begins
• Comprehensive searching across multiple databases (PubMed, Scopus, Cochrane, Web of Science)
• Dual-reviewer screening with documented eligibility criteria
• Formal quality assessment using validated tools such as RoB 2, ROBINS-I, or the Newcastle-Ottawa Scale [8]
• Reporting according to the PRISMA 2020 standard [5]

When the included studies report compatible quantitative data, the systematic review may incorporate a meta analysis to statistically pool results across studies.

Systematic reviews sit at the top of the evidence hierarchy in evidence-based medicine. They inform clinical guidelines from Cochrane, WHO, and NICE, and they satisfy regulatory requirements for medical device submissions under EU MDR and FDA pathways [4].

Scoping Review

A scoping review maps the breadth and nature of available evidence on a topic. Developed by Arksey and O'Malley and later refined by JBI, scoping reviews follow a structured methodology: a defined research question using the PCC framework (Population, Concept, Context), comprehensive searching, screening with documented eligibility criteria, and data charting [1] [7].

Unlike systematic reviews, scoping reviews do not formally assess study quality or pool results statistically. Their purpose is descriptive: cataloging what evidence exists, identifying the research methods that have been used, and mapping where gaps remain. They follow the PRISMA-ScR reporting extension and are particularly useful for emerging fields where the evidence base needs to be mapped before more focused synthesis can begin.

Other Review Types

Integrative reviews combine qualitative and quantitative evidence to build a complete picture of a topic. They are common in nursing, health sciences, and education research, where the evidence base spans both randomized trials and qualitative fieldwork.

Critical reviews go beyond summarizing. They evaluate theoretical frameworks, methodological assumptions, or conceptual models within a body of literature. The goal is to challenge existing thinking rather than merely catalog it.

Rapid reviews use abbreviated systematic review methods to produce evidence summaries within compressed timeframes. They serve urgent policy or clinical decisions where a full systematic review would take too long. WHO and national health agencies increasingly commission rapid reviews during public health emergencies.

Umbrella reviews (reviews of reviews) synthesize findings from multiple systematic reviews on related topics, providing a high-level evidence map across an entire domain.

Structure of a Literature Review

The exact structure depends on the discipline, the publication venue, and whether the review is a standalone article or part of a larger work. Most literature reviews follow a three-part framework.

Introduction

The introduction defines the topic, states the review's purpose, and sets the scope. In a dissertation chapter, this section explains how the review connects to the overall research question. In a standalone article, it establishes why the topic warrants a thorough review at this moment. A strong introduction also briefly describes the search and selection strategy so readers understand how the sources were identified.

Body

The body is where the actual synthesis happens. It organizes existing research into a coherent analysis rather than a list of summaries. Three organizational approaches are common.

Thematic organization groups studies by topic or concept. This is the most widely used approach and works well when the research spans multiple subtopics that each contribute to the larger question. For example, a review of remote learning might organize findings by student engagement, instructor adaptation, and assessment outcomes.

Chronological organization traces how understanding has evolved over time. This approach works when the historical development of ideas is itself important to the argument, such as reviewing how diagnostic criteria for a condition have changed across decades of research.

Methodological organization groups studies by research design, comparing what different approaches (surveys, experiments, observational studies) have found about the same question. This is useful when methodological differences might explain contradictory results.

The critical distinction between a good literature review and a poor one lies here. Effective reviews synthesize across studies, identifying where findings converge, where they conflict, and what might explain the differences. Weak reviews summarize each study in turn without drawing connections [6].

Conclusion

The conclusion pulls together the key findings, identifies the gaps that remain, and connects those gaps to the researcher's own work. In a dissertation, this section directly sets up the research questions or hypotheses that follow. In a standalone article, it proposes specific directions for future research and explains why those directions matter.

How to Write a Literature Review

Writing a literature review is a structured process with six core steps. The process is iterative: most researchers cycle back to earlier steps as their understanding deepens.

Step 1: Define Your Research Question

Start with a focused question. The question determines which databases to search, what keywords to use, and which studies to include or exclude. A question that is too broad produces an unmanageable volume of sources. A question that is too narrow may not yield enough evidence to support meaningful synthesis.

For systematic and scoping reviews, structured frameworks help sharpen the question. PICO (Population, Intervention, Comparison, Outcome) works for clinical questions. PCC (Population, Concept, Context) works for scoping reviews.

Step 2: Search the Literature

Search academic databases relevant to your discipline. PubMed covers biomedical research. Scopus and Web of Science provide broad multidisciplinary coverage. Discipline-specific databases like PsycINFO (psychology), ERIC (education), and IEEE Xplore (engineering) add depth in their fields.

Use a combination of keywords, subject headings (MeSH terms in PubMed, for example), and Boolean operators (AND, OR, NOT) to build your search strategy. Document every search string so it can be reported in your methodology. AI-powered academic search tools can supplement traditional database searching by surfacing relevant papers across 200 million+ indexed sources and ranking results by quality metrics such as SJR and SNIP, helping researchers identify high-impact work faster.

Consider including grey literature (conference proceedings, theses, technical reports) if full coverage is important to your review type.

Step 3: Screen and Select Sources

Work through results in two passes. First, scan titles and abstracts to filter out clearly irrelevant hits. Then read the full text of promising papers to confirm they meet your inclusion criteria.

Keep a record of why you excluded sources. For formal review types, this screening log becomes part of the PRISMA flow diagram. Even for a traditional literature review, documenting your selection decisions strengthens the review's credibility and helps you explain your scope to readers.

Step 4: Read and Analyze

Read each selected source carefully, extracting the information that matters for your review: research question, methodology, sample characteristics, key findings, limitations, and relevance to your topic. Many researchers build a data extraction table to organize this information systematically across studies. Structured extraction makes it far easier to spot patterns later.

As you read, look for convergences and contradictions. Which findings are consistent across multiple studies? Where do results conflict? What methodological differences might explain the disagreements? What populations, contexts, or variables remain underexplored?

Step 5: Organize and Synthesize

Arrange your findings into a coherent framework. Choose thematic, chronological, or methodological organization based on what best serves your research question. Write synthesis paragraphs that connect multiple studies rather than summarizing them one at a time.

Literature Review AI can accelerate this stage by screening up to 200 papers and synthesizing the top 50 into a structured review with citations, giving researchers a solid starting point to build on.

Step 6: Write and Revise

Draft the review section by section, ensuring each paragraph advances your argument. Use topic sentences that state the paragraph's point, support every claim with citations, and maintain a critical analytical voice. The goal is to evaluate the evidence, not merely describe it.

Revise for clarity, logical flow, and completeness. Check that every major finding in your extraction table appears in the review. Confirm that contradictory evidence is acknowledged and discussed. Read the review aloud to catch awkward phrasing and overly long sentences.

Common Mistakes in Literature Reviews

Five problems recur across disciplines. Recognizing them early saves significant revision time.

Summarizing instead of synthesizing. This is the most common weakness. A review that treats each study as a standalone paragraph ("Author A found X. Author B found Y.") reads as an annotated bibliography, not a literature review. The fix is to organize by themes and write paragraphs that draw connections across multiple sources.

Insufficient coverage. Relying on a narrow set of sources, particularly only recent publications or only one database, creates blind spots. Foundational works that shaped the field deserve inclusion alongside recent developments. A strong review shows the trajectory of knowledge, not just the latest snapshot.

Lack of critical analysis. Describing what studies found without evaluating their significance, limitations, or methodological quality produces a passive review. Strong reviews ask: How strong is this evidence? What are the limitations? How does this finding relate to conflicting results from other studies?

Poor organization. Listing studies one after another without a thematic or argumentative structure makes it difficult for readers to follow the reasoning. Before drafting, create an outline that groups sources by theme, chronology, or method so the structure is clear before writing begins.

Ignoring contradictory evidence. Selectively citing only studies that support a preferred conclusion undermines credibility. Transparent reviews acknowledge conflicting findings, discuss possible explanations for the contradictions (differences in sample, methodology, or context), and explain what the weight of evidence suggests overall.

Literature Review vs Other Review Types

Understanding where the traditional literature review sits relative to other review types helps researchers choose the right approach for their question and context.

A traditional literature review offers flexibility and interpretive depth. It serves academic purposes well: framing new research, demonstrating knowledge of a field, and building theoretical arguments. But it lacks the structured methodology that makes findings reproducible.

A systematic review trades that flexibility for rigor. Its pre-registered protocol, comprehensive searching, and standardized reporting through PRISMA 2020 make it the gold standard for evidence-based decision-making. Clinical guidelines, regulatory submissions, and health technology assessments all require systematic reviews.

A scoping review occupies the middle ground. It uses structured methodology (documented search strategy, screening criteria, data charting) but skips formal quality assessment. Scoping reviews are the right choice when the goal is to map what evidence exists on a broad topic rather than to answer a specific clinical or policy question.

A narrative review privileges expert interpretation over methodological transparency. The best narrative reviews, commissioned from leading authorities in a field, provide insights that structured approaches cannot replicate. The trade-off is that another expert reviewing the same field might reach different conclusions.

How AI Tools Support Literature Reviews

AI tools are reducing the manual workload at every stage of the literature review process, from discovery through synthesis.

AI-powered academic search helps researchers find relevant papers faster by searching across 200 million+ peer-reviewed sources and ranking results by research quality indicators like SJR quartile and SNIP. Rather than running separate searches across PubMed, Scopus, and discipline-specific databases, researchers can run a single query and get citation-grounded answers synthesized from the top results.

For the analysis phase, Chat with PDF lets researchers interrogate individual papers conversationally, extracting specific findings, methods, or data points without reading every section. Structured data extraction organizes findings across multiple sources into comparison tables, making it easier to spot patterns, contradictions, and gaps across a body of literature.

Literature Review AI can automate the most time-consuming part of the process by screening up to 200 papers and synthesizing the top 50 into a structured review with citations. The output is a starting point, not a finished product, but it can cut weeks from the early stages of a review.

Built-in reference management keeps citations organized from the first search through to the final manuscript, supporting over 1,000 citation styles and import from Zotero, Mendeley, RIS, and BibTeX files.

Conclusion

A literature review is more than a summary of published research. It is a critical analysis that establishes context, identifies gaps, and positions new work within the existing body of knowledge. The format ranges from the flexible traditional literature review in a dissertation chapter to the rigorous systematic review that informs clinical guidelines and regulatory decisions.

Choosing the right review type depends on the research question, the academic context, and the level of evidence required. What remains constant is the core skill: the ability to evaluate, organize, and synthesize existing research into a coherent analysis that advances understanding and justifies new inquiry.

Frequently Asked Questions

What is a literature review in simple terms?

A literature review is a written summary and analysis of existing research on a specific topic. It surveys what other researchers have found, identifies patterns and contradictions in the evidence, and highlights gaps where further research is needed. Literature reviews appear in dissertations, journal articles, and as standalone publications.

What is the difference between a literature review and a systematic review?

A traditional literature review uses flexible methodology to synthesize research and frame a research question, typically as part of a dissertation or journal article. A systematic review follows a registered protocol with comprehensive searching, dual-reviewer screening, quality assessment, and PRISMA reporting. Systematic reviews produce the highest level of evidence and inform clinical guidelines, while traditional literature reviews serve primarily academic purposes like framing research and demonstrating field knowledge.

How long should a literature review be?

Length depends on context. A dissertation literature review chapter typically runs 5,000 to 15,000 words depending on the discipline and institution. A journal article's literature review section is usually 500 to 1,500 words. A standalone review article ranges from 3,000 to 8,000 words. The length should be determined by the scope of the topic and the depth of analysis required, not by an arbitrary word count.

How many sources should a literature review include?

There is no fixed number. Dissertation literature reviews typically cite 50 to 200 sources depending on the breadth of the topic. Journal article literature review sections cite 20 to 50 sources. Standalone narrative reviews often cite 100 or more. The goal is thorough coverage of the relevant research, not reaching a specific count.

Can a literature review be published as a standalone paper?

Yes, though the venue and format depend on the review type. Systematic reviews with PRISMA reporting and scoping reviews with PRISMA-ScR reporting are regularly published as standalone research articles in peer-reviewed journals. Narrative reviews are typically published as invited articles in high-impact journals. A traditional dissertation literature review would need to be reformatted and strengthened methodologically before it could be published as a standalone article.

What is the difference between a literature review and a meta analysis?

A literature review synthesizes research qualitatively, organizing findings by themes and drawing interpretive conclusions. A meta analysis uses statistical methods to pool quantitative data from multiple studies into a single effect estimate with a confidence interval. Meta analysis is conducted within a systematic review framework and produces precise, reproducible estimates of treatment effects or associations.

What are the main types of literature reviews?

The main types include traditional (narrative) literature reviews, systematic reviews, scoping reviews, integrative reviews, critical reviews, and rapid reviews. Each follows different conventions for searching, selecting, and synthesizing sources. Traditional literature reviews are the most common in academic contexts, while systematic reviews are required for clinical and policy evidence.

How do I organize a literature review?

The three main organizational approaches are thematic (grouping studies by topic or concept), chronological (tracing how understanding has evolved), and methodological (comparing findings across different research methods). Thematic organization is the most common and works well for most research questions. The key is to synthesize across studies rather than summarizing them one by one.

Do I need to assess study quality in a literature review?

Formal quality assessment using validated tools like RoB 2 or the Newcastle-Ottawa Scale is required for systematic reviews but not for traditional literature reviews. Even in a traditional review, critically evaluating the strengths and limitations of key studies strengthens the analysis and demonstrates methodological awareness to readers and reviewers.

What software helps with literature reviews?

Literature review tools include reference managers (Zotero, Mendeley, Paperpile) for organizing sources, academic search engines for discovering papers, and AI-powered research platforms for accelerating screening and synthesis. Many researchers combine several tools across their workflow, though integrated platforms that handle search, screening, extraction, and citation management in one workspace are increasingly common for researchers who want to reduce context-switching between tools.

References

1. Arksey, H., & O'Malley, L. (2005). Scoping studies: towards a methodological framework. International Journal of Social Research Methodology, 8(1), 19-32. https://doi.org/10.1080/1364557032000119616
2. Baethge, C., Goldbeck-Wood, S., & Mertens, S. (2019). SANRA: a scale for the quality assessment of narrative review articles. Research Integrity and Peer Review, 4, 5. https://doi.org/10.1186/s41073-019-0064-8
3. Borah, R., Brown, A. W., Capers, P. L., & Kaiser, K. A. (2017). Analysis of the time and workers needed to conduct systematic reviews of medical interventions using data from the PROSPERO registry. BMJ Open, 7(2), e012545. https://doi.org/10.1136/bmjopen-2016-012545
4. Higgins, J. P. T., Thomas, J., Chandler, J., Cumpston, M., Li, T., Page, M. J., & Welch, V. A. (Eds.). (2019). Cochrane Handbook for Systematic Reviews of Interventions (2nd ed.). Wiley. https://training.cochrane.org/handbook
5. Page, M. J., McKenzie, J. E., Bossuyt, P. M., et al. (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ, 372, n71. https://doi.org/10.1136/bmj.n71
6. Pautasso, M. (2013). Ten simple rules for writing a literature review. PLoS Computational Biology, 9(7), e1003149. https://doi.org/10.1371/journal.pcbi.1003149
7. Peters, M. D. J., Godfrey, C., McInerney, P., Munn, Z., Tricco, A. C., & Khalil, H. (2020). Chapter 11: Scoping Reviews. In E. Aromataris & Z. Munn (Eds.), JBI Manual for Evidence Synthesis. JBI. https://synthesismanual.jbi.global
8. Sterne, J. A. C., Savovic, J., Page, M. J., et al. (2019). RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ, 366, l4898. https://doi.org/10.1136/bmj.l4898
9. Snyder, H. (2019). Literature review as a research methodology: An overview and guidelines. Journal of Business Research, 104, 333-339. https://doi.org/10.1016/j.jbusres.2019.07.039