How to write a literature review in education

Education research borrows from psychology, sociology, economics, and policy studies, which makes its literature review harder than it looks. You are not mapping one conversation; you are showing how several overlapping conversations bear on your question, and holding them together with a conceptual framework of your own.

Examiners in education pay particular attention to that framework. A review that summarises studies without building toward a set of defined constructs and expected relationships reads as unfinished, however wide the reading behind it.

This guide covers the education databases and source types, the role of policy documents and grey literature, thematic structure, conceptual frameworks, when scoping and systematic reviews apply, a worked outline, and how to keep AI assistance grounded in the papers you hold.

What an education literature review needs to do

The review has three jobs: situate your question in theory, in empirical evidence, and in the policy and practice context it lives in. Unlike clinical fields, education has no single hierarchy of evidence, so a fourth job follows: argue explicitly for the weight you give each study, whether it is a randomised trial, an ethnography, or a national evaluation.

It also has a destination. By the final section a reader should be able to see your conceptual framework assembled from the parts the review has established: the constructs, the relationships between them, and the reason your study is needed to test or explore them.

Scale is a further complication. Education studies range from a single classroom to international assessment programmes, and findings do not transfer automatically across levels. Part of the review's work is saying at which level each claim holds, and at which level your own study will sit.

Where to search: ERIC and the core education databases

Education is indexed unevenly across databases, so a serious search uses several:

The Literature Finder in CiteDash searches an internal corpus plus live OpenAlex, PubMed, Semantic Scholar and arXiv in a single query, which is a fast way to map the peer-reviewed terrain and follow citation trails before you work through ERIC's descriptors in your library's interface.

• ERIC: the central index for education research, covering journals and a deep pool of reports and grey literature
• Education Source and British Education Index: complementary coverage across school systems and regions
• PsycINFO: essential when your question touches learning, motivation, or development
• Scopus or Web of Science: cross-disciplinary reach and citation chasing in both directions
• ProQuest Dissertations and Theses: unpublished studies and examples of instruments
• Policy sources: education departments, inspectorates, and international bodies for the policy context

Source types in education research

Education draws on a wider range of source types than most fields, and each has a distinct role:

The distinction that matters is between evidence of effect and evidence of context. A ministry report tells you what a policy requires; it does not tell you whether the policy works. Reviews that blur that line get flagged.

Large-scale international studies and national datasets deserve their own care: they are strong on patterns and weak on mechanisms, so use them to establish the landscape and rely on closer-grained studies to explain it.

• Peer-reviewed empirical studies: quantitative, qualitative, and mixed methods
• Systematic reviews and meta-analyses where they exist for your topic or intervention
• Policy documents and government reports that define the context you are studying
• Books and edited chapters, which carry more weight in education than in the sciences
• Dissertations and conference papers, used as grey literature with appropriate care
• Practitioner publications for problem framing and context, not as evidence of effect

Build a conceptual framework, not just a summary

A conceptual framework names your constructs, defines them, and states the relationships you expect between them. The literature review is where it gets built: each thematic section should establish a construct or a relationship, so the framework arrives as a conclusion rather than an announcement.

Argue the theory choice. If you frame motivation through self-determination theory, say why that lens fits your question better than the alternatives you passed over. Examiners probe theory choice at the viva precisely because weak reviews treat it as given.

A simple diagram helps more than another page of prose. Boxes for constructs, arrows for expected relationships, and a caption that separates what is hypothesised from what is exploratory give examiners a single figure to test your logic against, and give you a map for the methods chapter that follows.

Structure the review thematically around your research questions

Map your themes to your research questions so that every section is visibly doing work. Within each theme, funnel: the broad territory, the studies closest to your question, the state of the evidence, and what remains open. End each section with evaluation, not a running summary.

Chronological organisation is only right when development over time is itself the point, such as tracing how a policy or a construct evolved. Otherwise it hides your argument behind a timeline.

Signpost heavily. Open each theme with a sentence that says what the section establishes and how it connects to the previous one, and close with what it hands forward. Education reviews cross disciplinary boundaries, and readers lose the thread precisely at those crossings unless you carry them over explicitly.

Weighing quantitative, qualitative, and mixed methods evidence

Because education has no single evidence hierarchy, each study should be judged against the quality criteria of its own tradition: design and power for quantitative work, depth and transparency for qualitative work, integration for mixed methods. An effectiveness question leans on trials and quasi-experiments; a meaning or experience question leans on qualitative studies.

Make the weighing visible in the prose. A sentence that says why a study is credible, or why its context limits generalisation, is the appraisal examiners are looking for, and it is what separates synthesis from summary.

Sample and context reporting matter as much as design. Note where each study was conducted, with whom, and under what conditions, because education effects are notoriously sensitive to context, and your review's credibility rests on not stretching a finding beyond the setting that produced it.

Where reviews or meta-analyses exist, use them as scaffolding rather than substitutes: cite them for the shape of the field, then read the primary studies that matter most to your question, because examiners ask about primaries, not summaries.

Scoping and systematic reviews in education: when PRISMA applies

If you claim a systematic or scoping review, the reporting standards follow: documented databases and search strings, inclusion and exclusion criteria fixed before screening, counts at every stage, and a PRISMA flow diagram, with the scoping extension applying to scoping work.

Most education thesis reviews are narrative, and that is legitimate; the mistake is claiming the systematic label without the protocol behind it. If you are weighing how much AI help to use in either mode, see how to write a literature review with AI for where assistance is safe and where it is not.

Scoping reviews are increasingly popular in education for mapping emerging topics, and they carry their own reporting expectations: a clear charting framework, explicit inclusion logic, and a statement of what the map does not cover.

Document your search, even in a narrative review

Even without a formal protocol, record which databases you searched, the main search terms and their combinations, the date limits, and the date of the final search. In ERIC, note the descriptors you used alongside free-text terms, because the controlled vocabulary is what makes an education search reproducible.

Screening discipline transfers too: a first pass on titles and abstracts, a second on full text, and a short note on why borderline studies were excluded. When an examiner asks why a well-known study is absent, a documented decision is a strong answer and an undocumented absence is a weak one.

A worked outline for an education literature review

Here is an outline for a thesis on formative assessment and student motivation in secondary schools:

Notice that the conceptual framework has its own section near the end, built from the themes that precede it. That ordering is the strongest signal an education examiner gets that the review did its job.

Length allocation is a signal too: the themes closest to your research questions deserve the most space, and a review that spends more words on background than on the studies nearest its own question has its proportions backwards.

• Introduction: scope, search approach, and how the review is organised
• Policy and practice context: how formative assessment entered curriculum and inspection frameworks
• Theme 1: formative assessment, definitions, and evidence on classroom practice
• Theme 2: student motivation, the theoretical lenses and how each is measured
• Theme 3: studies connecting assessment practice to motivation, contradictions included
• Methodological critique: designs, contexts, and the limits of generalisation
• Conceptual framework: the constructs and relationships your study will examine
• Synthesis and gap: the specific contribution your study makes

How CiteDash grounds an education literature review

AI assistance fails badly when it invents references or attributes findings to the wrong study, and education's mix of similar-sounding reports and papers makes that easy. In CiteDash every citation is a database object that resolves to a real paper, and free-text references do not exist.

Save papers to your library, upload PDFs or fetch details from a DOI, and use Ask and extract to pull samples, contexts, and findings with sentence-level provenance, so each extracted claim binds to the passage behind it. Build an evidence matrix in the Synthesis Lab with context, design, and findings across your included studies; the same matrix supports PRISMA-friendly work if your review is systematic.

When you draft, the Fact Checker verifies each claim against the full-text PDFs you hold before the text reaches you, flagging anything it cannot verify rather than letting it through.

Reading statuses in the library keep a long screening pile manageable, and the Reference Manager formats the final bibliography in your required style, with APA and Harvard among the 12 it supports.

Mistakes examiners flag in education reviews

The recurring problems in education reviews cluster around the framework and the evidence-context line:

All of them are avoidable with the same habits: search deliberately, read what you cite, weigh evidence against its own tradition, and build the framework in the open. The review that results is the one an education examiner can pass without hesitation.

• A conceptual framework that appears at the end without being built through the review
• Policy documents treated as evidence of effectiveness
• One-paragraph-per-paper structure with no synthesis
• Qualitative literature ignored on a question that is partly about classroom experience
• The systematic label claimed without a reproducible search
• Seminal theorists cited second-hand rather than from the original texts