Critical Literature Review

A critical literature review evaluates the research it covers instead of simply reporting it. It asks how good the evidence is, what assumptions sit behind it, where studies agree and disagree, and what all of this means for the question you are investigating. In most dissertations this is not an optional extra: when a supervisor or marking scheme asks for a “critical” literature review, it is asking you to show judgement about the literature, not just familiarity with it.

For a dissertation, the practical question is not simply “what is a critical literature review?” It is:

How do I turn a review that summarises sources one by one into a review that judges them, and how do I show that judgement in my writing?

This page explains what being critical means in a literature review, how a critical review differs from a descriptive, narrative and argumentative review, and gives you an appraisal grid, a before-and-after example and a step-by-step process you can apply to your own chapter. It sits alongside the overview of types of literature review. If you have been told to “defend a position”, see the argumentative literature review: an argumentative review starts from a position and builds the case for it, while a critical review starts by judging the quality and assumptions of the literature, and any position it reaches comes out of that judgement.

On this page:

Critical Literature Review Explained Simply

Imagine asking two friends to recommend a laptop. The first reads out the specifications of six models from a website. The second tells you that two of the reviews you found were written by the manufacturer, that the cheapest model’s battery claims were measured with the screen dimmed, that most independent tests agree on one model for students, and that the right choice depends on whether you need to run statistical software. Both friends used the same sources. Only the second one was critical.

A critical literature review does what the second friend did. It still reports what the literature says, but it also asks who produced each piece of evidence, how, under what conditions, and how much weight it can bear. It looks for patterns, contradictions and gaps across studies, and it ends with a clear view of what is known, what is uncertain and what your own study needs to do. Saunders and Rojon (2011) put it in similar terms: a critical review discusses and evaluates research rather than simply describing it, considers work that both supports and opposes the writer’s ideas, and uses the result to justify the aims of the new study.

Being critical does not mean being negative. Saying that a study is well designed, that its findings have been replicated, or that a theory explains a wide range of cases is also a critical judgement, as long as you give the reasons for it.

Dudovskiy Research Assistant

Not sure what kind of literature review your dissertation needs?

Enter your topic to receive a free, tailored methodology preview.

Free · No account required
Learn more about Dudovskiy Research Assistant

Two Meanings: a Review Type and a Quality of Every Review

The phrase “critical literature review” is used in two related ways, and it helps to know which one your marking scheme means.

1. A quality expected of every dissertation literature review. This is the most common meaning for students. In widely used business research textbooks, the literature review chapter is described as a critical review by default: Saunders, Lewis and Thornhill (2019) title their chapter on the topic “Critically reviewing the literature”. In this sense, a narrative, thematic or even systematic review can and should be critical. Boote and Beile (2005) argued that a thorough, sophisticated review is the foundation of useful research, and proposed criteria for judging the quality of dissertation literature reviews after examining dissertations at three universities.

2. A distinct type of review. In the review-methods literature, the critical review is one of several named review types. In their widely cited typology of 14 review types, Grant and Booth (2009) describe it as a review that shows the writer has researched the literature extensively and evaluated its quality, going beyond description to include analysis and conceptual innovation. It seeks the most significant items rather than every item, has no formal quality-assessment procedure, usually uses narrative synthesis, and typically produces a hypothesis or model rather than a final answer. Grant and Booth also list its main weakness: because it judges studies by their conceptual contribution rather than by a formal quality checklist, its conclusions are more subjective than those of a systematic literature review.

Wright and Michailova (2023) treat the critical literature review as a distinct review type in management and organisation studies. Examining reviews that their authors had labelled “critical”, they found that many did not explain what “critical” meant and were largely uncritical in practice, and they identified 19 exemplary reviews that showed genuine judgement about the literature they covered. The lesson for a dissertation is direct: calling your review critical does not make it so. The criticality has to be visible on the page.

Critical vs Descriptive Literature Review

Most feedback that a literature review is “too descriptive” refers to the same handful of habits. The table contrasts them with what a critical review does instead, drawing on the attributes set out by Saunders and Rojon (2011).

Feature Descriptive review Critical review
Organisation Source by source, like an annotated bibliography By theme, debate or question, with several sources in each paragraph
Sources included Everything found on the topic The most relevant and significant work, including key theories and studies that contradict each other
Treatment of studies Reports what each author found Assesses how the findings were produced and how much weight they can bear
Disagreement Ignored, or mentioned without comment Explained: different samples, methods, definitions, contexts or assumptions
Writer’s voice Absent; the paragraph could have been written by anyone Present; evaluative statements are backed by evidence and separated from established facts
Ending A summary of what has been covered A clear gap, problem or question that justifies the study’s aims

How It Differs from Narrative and Argumentative Reviews

The Portal covers several review types. The table shows where the critical review sits among the ones students most often confuse.

Review type Main purpose Starting point Typical output
Narrative Give a readable overview of a topic A topic A summary of the state of knowledge
Argumentative Support or challenge a specific position A position A case for or against that position
Critical Judge the quality, assumptions and contribution of the literature A question about how far the literature can be trusted An assessment of what is known and weakly known, often a model or a research gap
Systematic Answer a focused question from all eligible studies A protocol A transparent synthesis with formal quality assessment

These types overlap. A narrative review written critically and a critical review written narratively can look very similar. What matters for your mark is less the label than whether each paragraph shows judgement.

What Being Critical Actually Means

Mingers (2000), writing about teaching management students to think critically, distinguished four aspects of criticality, which Saunders and Rojon (2011) apply to literature reviews. They translate into four questions you can ask of any source or body of literature:

  • Rhetoric: Is the argument well made? Do the conclusions follow from the evidence, or does the language claim more than the data show?
  • Tradition: What does this work take for granted because “everyone in the field” assumes it? Would researchers from a different tradition frame the problem differently?
  • Authority: Is a claim accepted because of who made it or where it was published, rather than because of the evidence? Are some voices or contexts missing from the literature?
  • Objectivity: What values, interests or funding might shape the research questions, methods or interpretation?

In practice, most dissertation-level criticism concentrates on a smaller set of concrete issues: the design and sample of each study, how key concepts were defined and measured, the context in which the data were collected, how strongly the conclusions are worded, and how the study fits with others. The appraisal grid below turns these into a routine.

The Critical Appraisal Grid

Use a grid like this while you read, one row per key study. It forces you to record judgements, not just findings, and it makes the comparison across studies visible before you start writing. The example row shows how a completed entry might look for an invented study.

Question What to record Example entry (invented study)
Claim The main finding in one sentence Remote work increased productivity by 12%
Evidence and design Method, sample, setting, period Survey of 180 staff at one software firm, self-rated productivity, 2021
Concepts and measures How key concepts were defined and measured “Productivity” measured by self-report, not output
Limits What the design cannot show Single firm, lockdown conditions, possible self-report bias, no comparison group
Assumptions What the study takes for granted Assumes productivity is individual rather than team-based
Fit with other studies Agrees with, contradicts or extends whom, and why Agrees with other self-report studies; contradicts output-based studies
Weight in my review How much your argument should rely on it Moderate: useful for perceptions, weak for actual output
Relevance to my question What it means for your study Suggests my study should use an output measure

The last two rows are where the critical work happens. A review built from completed grids rarely reads as descriptive, because every study arrives in your draft with a judgement already attached.

From descriptive to critical: five levels of engagement with the literature, from describe and compare to evaluate, synthesise and position, with an example phrase for each level

The diagram shows the same idea as a progression. Describing and comparing studies are necessary, but they are only the first two steps. Evaluation, synthesis and a clear position for your own study are what make a review critical.

Worked Example: From Descriptive to Critical

The paragraphs below discuss the same three invented studies on remote work and productivity. Authors A, B and C are placeholders, not real sources.

Descriptive version. Author A (2021) studied remote work at a software company and found that productivity increased by 12%. Author B (2022) surveyed employees in banking and found no change in productivity. Author C (2023) found that productivity fell among new employees who worked remotely. These studies show that remote work affects productivity.

Critical version. Evidence on remote work and productivity is mixed, and much of the disagreement appears to reflect how productivity is measured. Author A (2021) reports a 12% increase, but the figure comes from self-ratings at a single software firm during lockdown, when there was no office-based comparison group. Author B (2022), using supervisor ratings in banking, finds no change, while Author C (2023), using output records, finds a decline among new employees only. Read together, the studies suggest that positive results rely on self-reports and that effects may depend on tenure, a factor A and B did not examine. The literature therefore offers weak evidence for a general productivity gain and points to tenure as an untested condition. This study addresses that gap by comparing output-based productivity for new and established employees.

The critical version is about the same length. It does not add more sources. It changes what the paragraph does: it is organised around a debate rather than a list of authors, explains why the studies disagree, assigns different weight to different evidence, separates what the studies found from the writer’s interpretation (“suggest”, “appears”), and ends by connecting the gap to the writer’s own study.

How to Write a Critical Literature Review: Step by Step

  1. Start from your research question. Criticism needs a direction. Decide what your review must establish for your study, such as which concepts to use, which relationships are already supported and where evidence is weak.
  2. Select the significant work, not everything. Identify the key theories, the most cited and most rigorous studies, and work that contradicts the main view. Keep searching as you write, so that recent studies are not missed.
  3. Appraise as you read. Complete an appraisal grid for each key study. For empirical studies, a structured appraisal checklist for the relevant design can help you ask consistent questions.
  4. Group by theme or debate. Sort your grids into themes, debates or variables. Each theme becomes a section, and each paragraph brings several studies together.
  5. Explain agreement and disagreement. For each theme, ask why the studies differ. Different samples, contexts, measures, time periods or theoretical assumptions usually explain more than “the evidence is mixed”.
  6. Write evaluative topic sentences. Open each paragraph with your judgement, then support it with evidence. A paragraph that begins “Author A found…” usually ends up descriptive.
  7. End each section, and the chapter, with the implication. State what the evidence allows you to conclude, what it does not, and how that shapes your research questions, hypotheses or design.
  8. Check the draft. Saunders and Rojon (2011) end their paper with a checklist for testing whether a review is critical. Ask, for every paragraph: could someone who had only read the abstracts have written this? If yes, it is still descriptive.

Language for Critical Writing

Critical judgement shows partly through word choice. The phrases below are starting points; the evidence that follows them is what makes them critical.

Purpose Example phrases
Weighing evidence “provides strong support for”, “rests on a small and specific sample”, “has been replicated in”, “remains largely untested”
Explaining disagreement “the difference may reflect”, “these findings are difficult to compare because”, “once measurement is taken into account”
Questioning assumptions “assumes rather than demonstrates”, “takes for granted that”, “is framed almost entirely from the perspective of”
Synthesising “taken together”, “a consistent pattern across”, “the weight of evidence suggests”
Positioning your study “this leaves open the question of”, “no study has yet examined”, “this study therefore”

Use reporting verbs carefully. “Shows” and “proves” treat a finding as settled; “reports”, “argues”, “suggests” and “claims” signal that you are presenting someone else’s position, which you can then evaluate.

Critical Literature Review in Business Research

Business and management dissertations rely heavily on sources that need careful appraisal: practitioner reports, consultancy surveys, company publications and popular management books sit alongside peer-reviewed research. A critical review in this field asks who produced the evidence and why. A consultancy survey reporting that most firms plan to adopt a technology may be commercially motivated and based on a convenience sample; a company’s own sustainability report is a statement of intent, not independent evidence of performance.

Management research is also full of influential frameworks, such as Porter’s Five Forces, the resource-based view or Maslow’s hierarchy, that students tend to apply without question. A critical review recognises why such frameworks are useful, but also asks where they came from, what assumptions they make, whether they have been tested in the student’s context, and what later research has added or challenged. Showing awareness of these limitations, and choosing a framework for reasons connected to the research question, is one of the clearest ways to demonstrate criticality in a business dissertation.

Common Mistakes

  • One paragraph per source. This produces an annotated bibliography, however good the summaries are.
  • Criticism without reasons. “This study is weak” or “this study is flawed” is an opinion until you say what the weakness is and why it matters for your question.
  • Criticising only studies you disagree with. Apply the same appraisal to evidence that supports your expected answer.
  • Confusing “critical” with “negative”. Strong evidence should be recognised as strong. A review that dismisses everything is as uncritical as one that accepts everything.
  • Generic limitations. “The sample size was small” applies to many studies. Say what the limitation means for this finding and this question.
  • No conclusion. A review that ends with “more research is needed” without saying what research, and why, has not done its job. The review should lead directly to your aims.
  • Citing sources you have not read. Criticism of a study known only from its abstract, or from someone else’s summary, is easy for examiners to spot.

Critical Literature Review in the Age of AI and Digital Research

AI tools can now summarise papers, suggest sources and draft review paragraphs in seconds. Used carelessly, they make the descriptive problem worse: a summary of a paper’s abstract is, by definition, a description, and an AI-generated review paragraph usually reads like the descriptive version in the worked example above. The judgement that makes a review critical cannot be delegated, because it depends on your research question and on reading the evidence itself.

AI tools also introduce a new risk to the reliability of the literature itself. Walters and Wilder (2023) asked ChatGPT to write short literature reviews on 42 topics and checked all 636 references it produced: 55% of the GPT-3.5 citations and 18% of the GPT-4 citations were fabricated, and many of the real ones contained substantive errors. A critical reviewer therefore treats any AI-suggested source as unverified until it has been found in a database or on the publisher’s site, and treats any AI summary as a claim to be checked against the original text.

Used well, AI can support criticality rather than replace it. You can ask a tool to suggest counter-arguments to a position, identify studies that report contradictory findings, or list the methodological limitations typically raised for a design, and then check each suggestion against real sources. Digital tools such as reference managers, citation-tracking databases and literature-mapping software can also show how a debate has developed and which studies are central to it. In each case the tool widens your search; deciding what the evidence is worth remains your task, and it is the part examiners assess.

Exam Tip

If you are asked what makes a literature review critical, do not answer with a definition alone. Name the shift from describing sources to evaluating them, give two or three concrete things a critical reviewer examines (study design and sample, how concepts were measured, the assumptions behind the work, and why studies disagree), and show the difference with a one-sentence example, such as contrasting “Author A found X” with “Author A’s finding rests on self-reported data from one firm, so it is weaker evidence than B’s output-based study”. Mention that critical does not mean negative, and that the purpose of the critique is to justify the research question. That combination demonstrates the skill rather than describing it.

Once your critical review has identified the gap your study will address, the next step is choosing a methodology that can address it. Dudovskiy Research Assistant can review your research questions and help you choose and justify a research philosophy, approach, design and methods for your methodology chapter. It does not write literature reviews; it supports your own judgement, and the final decisions remain yours and your supervisor’s.

John Dudovskiy

References

Boote, D.N. and Beile, P. (2005). Scholars before researchers: on the centrality of the dissertation literature review in research preparation. Educational Researcher, 34(6), 3–15. https://doi.org/10.3102/0013189X034006003

Grant, M.J. and Booth, A. (2009). A typology of reviews: an analysis of 14 review types and associated methodologies. Health Information and Libraries Journal, 26(2), 91–108. https://doi.org/10.1111/j.1471-1842.2009.00848.x

Mingers, J. (2000). What is it to be critical? Teaching a critical approach to management undergraduates. Management Learning, 31(2), 219–237.

Saunders, M.N.K. and Rojon, C. (2011). On the attributes of a critical literature review. Coaching: An International Journal of Theory, Research and Practice, 4(2), 156–162. https://doi.org/10.1080/17521882.2011.596485

Saunders, M.N.K., Lewis, P. and Thornhill, A. (2019). Research Methods for Business Students. 8th edn. Harlow: Pearson.

Walters, W.H. and Wilder, E.I. (2023). Fabrication and errors in the bibliographic citations generated by ChatGPT. Scientific Reports, 13, 14045. https://doi.org/10.1038/s41598-023-41032-5

Wright, A. and Michailova, S. (2023). Critical literature reviews: a critique and actionable advice. Management Learning, 54(2), 177–197.

Last reviewed: November 2026.

[]