Reading Notes: How to Write a Literature Review

Knopf (2006)

Annotated Bibliography
Methodology
Author

Tales Mançano

Published

April 19, 2026

Knopf, J. W. (2006). How to write a literature review. PS: Political Science and Politics, 39(1), 127–132. https://doi.org/10.1017/S1049096506060264

@article{Knopf2006,
  author  = {Knopf, Jeffrey W.},
  title   = {How to Write a Literature Review},
  year    = {2006},
  journal = {PS: Political Science and Politics},
  volume  = {39},
  number  = {1},
  pages   = {127--132},
  doi     = {10.1017/S1049096506060264},
}

Last updated: 2026-04-19 Model: Perplexity, powered by Sonnet 4.6 Thinking Generated at: 2026-04-19T16:54:00-03:00


1 Introduction and Purposes of the Literature Review (pp. 127–128)

1.1 Defining the Literature Review and Its Three Contexts [§1–§4]

Jeffrey Knopf opens the article by acknowledging the relative scarcity of guidance on how to conduct literature reviews in political science and related social science methodology training. He notes the existence of some general research methods texts and a handful of dedicated guides on the topic, while observing that these are rarely cited or required in graduate training. The article positions itself as a practitioner-oriented overview, drawing in part on points from a lecture by Paul Pitman and a handout by John Odell for students at the University of Southern California.

In §2–§3, Knopf identifies three distinct contexts in which a literature review is produced and used. The first is a standalone review essay, whose purpose is to summarize and evaluate the existing state of knowledge on a subject for the scholarly community. The second is the literature review section of a research proposal, which serves to justify the proposed research by demonstrating where a gap in knowledge exists or where a debate remains unresolved. The third is a literature review in a finished research report, such as a thesis or dissertation, where it situates the author’s final conclusions within the broader scholarly conversation. In §4, Knopf emphasizes that this third form generally builds on and revises work completed at the proposal stage, serving to show how the final findings relate to prior wisdom about the subject.

1.2 Ways to Frame the Contribution to Knowledge [§5–§9]

In §5–§6, Knopf introduces a framework for thinking about what it means to “contribute to knowledge.” He is careful to note that knowledge in the social sciences does not mean “Truth” with a capital T, but rather refers to beliefs — specifically, beliefs that individuals hold with varying degrees of confidence as a result of study or experience. Because many social science hypotheses cannot be proven conclusively, literature reviews typically engage with the “claims” or “arguments” advanced by studies or schools of thought, and assess the strength of the support offered for those claims.

Note

Core conceptual framework: Knopf proposes that knowledge has two elements — what we believe and how strongly we believe it. Further research can affect either or both elements, positively or negatively, and any such result constitutes a contribution to knowledge.

In §7–§9, Knopf explicitly draws an analogy to Bayesian analysis: just as a Bayesian statistician revises the estimated probability of a proposition being true upon acquiring new data, a researcher can attempt an analogous qualitative assessment. This framework suggests that a contribution to knowledge may involve (a) creating an entirely new belief where none existed before (new factual information, a new theoretical proposition, or a new policy proposal); (b) strengthening an existing belief through corroborating evidence; or (c) challenging an existing belief through contradictory evidence or analysis. Knopf anchors this Bayesian analogy not as a call for quantification, but as a conceptual heuristic for identifying where new research can make a difference by locating existing beliefs and their levels of confidence.

1.3 Casting the Net Widely: Beyond Traditional Academic Sources [§10–§13]

In §10–§12, Knopf argues that while the traditional literature review focuses on peer-reviewed books and journal articles, the growth of alternative research producers and Internet-based dissemination makes it advisable to consider a wider range of sources. He suggests that “review of existing knowledge” is a more accurate term for this enterprise than “literature review” per se. The entities that might produce relevant research include government agencies, international governmental organizations, non-governmental organizations, think tanks, and independent researchers. Conference papers, working papers, and monographs posted online also represent scholars’ most current thinking and may not yet be published in traditional outlets.

In §13, Knopf issues a careful caveat: the Internet must be used with great caution, since — unlike peer-reviewed publications — anyone with access to the necessary equipment can post content without submitting to scholarly scrutiny. He recommends screening Internet postings for author credentials, source documentation, and the credibility of their evidence base. If the research interest involves existing policy proposals or practices, academic credibility may matter less than whether the source is in a position of authority or has inside knowledge — but even then, researchers must screen out postings that lack a valid basis for their assertions.


2 Pointers for an Effective Literature Review (pp. 128–129)

2.1 Seven Practical Guidelines for Conducting a Review [§14–§20]

In §14, Knopf opens his practical guidance with the suggestion that researchers — especially those producing their first literature review — should begin by reading existing review essays to observe how other researchers have carried out the task, imitating what seems effective and avoiding what seems unnecessary or unhelpful. Graduate-level course assignments and advisor recommendations are identified as natural first sources for such models.

The second pointer (§15) concerns the ability to succinctly summarize each study’s main claim. The researcher should be able to describe in one or two sentences the central argument of each item read, even if not every such description ultimately appears in the written review. This discipline ensures conceptual clarity and facilitates comparison across works.

Note

Seven pointers for effective literature reviews:

  1. Read existing review essays as models.
  2. Summarize each study’s main claim in one or two sentences.
  3. Be selective — include only items directly relevant to your focus.
  4. Do not summarize item by item; impose intellectual order on the material.
  5. Group studies into “camps” or “schools of thought” by theory, method, or policy preference.
  6. Read key sources yourself; do not rely solely on others’ summaries.
  7. Associate individual authors with major camps and theoretical labels.

In §16–§17, Knopf introduces the third and fourth pointers: selectivity and intellectual ordering. The written review should not discuss every item read, but only those with a direct bearing on the central focus. More crucially, the review should not be structured as a sequential summary — “paragraph 1 notes that book A says X; paragraph 2 notes that article B says Y” — but should instead impose conceptual organization on the material.

In §18–§19, the fifth through seventh pointers are developed. Grouping studies into “camps” or “schools of thought” (by theoretical perspective, methodological approach, or policy orientation) allows the review to efficiently convey that multiple authors share a position without devoting separate paragraphs to each. Knopf notes that for well-developed subjects, other scholars will often have already classified the research into contrasting schools of thought — encyclopedias, handbooks, review articles in academic journals, and dissertation literature review chapters are flagged as useful resources for orienting oneself to an established field. He issues a caution (§19) against relying exclusively on others’ summaries: reading the most critical sources oneself and drawing independent conclusions is essential. The final pointer (§20) is to form the habit of associating individual author names with theoretical camps, since academic writing routinely uses author surnames as shorthand for theoretical traditions (e.g., “Waltzian” for neo-realism).


3 The Nuts and Bolts: What the Review Must Answer (pp. 129–131)

3.1 Four Tasks of the Literature Review [§21–§26]

In §21–§22, Knopf identifies four core tasks that most literature reviews must address. The first two are analytical prerequisites: determining what each study examined (its dependent variable, conceptualization, operationalization) and what each study concluded (its thesis, its level of confidence in its findings, and its stated qualifications). These two steps are essential before collective results can be meaningfully compared, since different studies may reach different conclusions simply because they asked different questions or defined the phenomenon of interest in different ways.

Note

Four tasks of the literature review: 1. Determine what each study examined (question, dependent variable, conceptualization). 2. Identify the main argument and its supporting evidence in each work. 3. Summarize collective results: areas of consensus, areas of disagreement, and gaps. 4. Evaluate the overall quality of the literature: valid findings vs. areas needing more research.

In §23–§26, Knopf elaborates on the third task — summarizing existing studies — through a three-category typology. (1) Areas of consensus or near-consensus represent the “conventional wisdom” about a subject; these involve shared beliefs (positive or negative) about what is true or what works. (2) Areas of disagreement or debate produce alternative “camps” or “schools of thought.” (3) Gaps refer to aspects of a topic not yet examined — unanswered questions, unexplored perspectives, or unanalyzed bodies of information. Knopf stresses that once these three categories are identified, the literature review can precisely describe the proposed research’s contribution and why it will matter to the intended audience. Contributions may address any of the three: challenging conventional wisdom, weighing in on an existing debate, or filling an empirical or theoretical gap.

3.2 Evaluating the Quality of the Literature [§27–§31]

In §27–§29, Knopf addresses the fourth task: assessing the quality of the literature. This involves examining four dimensions of how scholars have reached their conclusions. The first is assumptions — whether disagreements among studies can be traced to different foundational premises, and whether those premises are plausible or so problematic as to undermine the analysis that rests on them. The second is logic — whether the reasoning offered by each study is internally consistent and logically persuasive, or whether it proceeds by assertion and contains internal contradictions or “giant leaps” at critical analytical junctures.

In §30–§31, Knopf addresses evidence and methodology. Regarding evidence, the key questions concern whether claims are empirically documented, whether the evidence is factually accurate and on-point, whether all relevant cases and data have been considered, and whether selection bias might have affected the results. Regarding methodology, the review should evaluate whether the methods used were appropriate for the question being asked and whether they were applied correctly. By systematically identifying and comparing the assumptions, theories, data, and methods of studies, the reviewer can pinpoint the underlying disagreements responsible for debates in the literature and target their own research accordingly.


4 Managing Sources: Too Few and Too Many (pp. 130–131)

4.1 The Two-Tier Framework for Thin Literature [§32–§36]

In §32–§33, Knopf addresses the common situation where students believe no literature is available on their topic. He recommends conducting a serious search before concluding that the topic is virgin territory, noting that a thorough search that uncovers nothing is itself valuable evidence for a research proposal’s claim to originality. When sources are genuinely scarce, Knopf proposes thinking in terms of two tiers (or circles) of literature.

In §34–§36, the first tier (inner circle) covers publications that directly address the proposed research question. The second tier (outer circle) encompasses work that overlaps with some aspect of the question — for instance, studies employing a theory that could be applied to the topic even if it has not been used in that specific context, or research on analogous situations. Knopf illustrates this with the example of agro-terrorism research: if no literature exists on protecting crops from agro-terrorism, a researcher might consult work on protecting crops from natural disease outbreaks and assess whether those techniques could be adapted. The analogy-based expansion of scope allows the review to situate a novel topic within existing intellectual frameworks.

4.2 Strategies When There Are Too Many Sources [§37–§40]

In §37–§38, Knopf identifies the opposite problem — an abundance of potentially relevant literature — that often arises once the researcher expands to the second tier and encounters well-developed theoretical fields. He warns against selecting sources at random (e.g., whatever happens to be on the library shelf or appears in the first Google results), since random selection risks producing a review that does not accurately represent the current state of knowledge and debate.

In §39–§40, Knopf offers three rules of thumb for restricting focus when sources are abundant: (1) prioritize leading authorities — authors and studies that are frequently cited in the literature; (2) emphasize recent studies from high-prestige outlets — books from highly ranked university presses and articles in leading journals, as well as high-visibility sources; and (3) focus on studies most directly relevant and helpful for the research question. He concludes that when there is a great deal of literature, comprehensiveness is not required; the review should focus on those parts of the literature that relate to and advance the researcher’s specific interests.


5 The Bottom Line (pp. 131–132)

5.1 Summary Criteria and Guiding Questions [§41–§44]

In §41–§42, Knopf provides a concluding summary of what a literature review should accomplish: it should concisely summarize, from a set of relevant sources, the collective conclusions most pertinent to the researcher’s own interests, and evaluate the state of knowledge in terms of what is right, what is wrong, what remains an unresolved area of uncertainty, and what is missing because no one has carefully considered it.

Tip

Knopf’s operative synthesis is that proceeding systematically and aiming at a considered judgment about the state of knowledge on a subject transforms the literature review from a passive summary exercise into an independent contribution to knowledge in its own right.

In §43–§44, Knopf closes with a checklist of guiding questions for constructing a rigorous review: What questions have existing publications addressed, and what has been neglected? What are the main conclusions, and where do studies converge or diverge? What theories, policies, or evidence have been examined, and what potentially relevant alternatives have not? How solid are the existing conclusions? What are the most important remaining problems and gaps? These questions apply regardless of whether the review is a standalone essay, part of a research proposal, or a component of a thesis or dissertation.


6 Synthetic Argument

Note

Central thesis: A literature review is not a passive summary of prior work but a structured intellectual exercise in evaluating the state of knowledge on a subject — identifying where consensus exists, where debates remain unresolved, and where gaps persist — in order to justify and situate new research.

Nature of the argument: The argument is primarily normative-pedagogical: Knopf prescribes how literature reviews should be conducted rather than describing how they typically are produced or testing a hypothesis about their effects. The text is structured as a set of action-guiding prescriptions backed by practical rationale.

What the text demonstrates vs. what remains agenda: The article successfully demonstrates, through clear conceptual exposition and practical examples, that literature reviews serve distinct functions depending on their context (standalone, proposal, or final report), and that they can be structured analytically rather than descriptively. However, the text does not provide empirical evidence that its recommendations produce better outcomes (e.g., higher-quality proposals), nor does it engage systematically with the variation in disciplinary conventions across subfields. The question of how to evaluate contested or ideologically charged literatures, where the notion of “consensus” is itself politically contested, remains undeveloped.

Contribution to broader debate: The article belongs to a pedagogical tradition in political science methods training (alongside King, Keohane, and Verba 1994, and the Cresswell 2003 mixed-methods tradition) that seeks to make implicit scholarly practices explicit and teachable. Its specific contribution is to operationalize the literature review as a discrete, teachable genre with identifiable tasks, structures, and quality criteria — a contribution of practical significance for graduate training even if its originality as a theoretical claim is limited.


Critical Analytical Sheet

Note

This section follows the IA Spreadsheeting Texts v13.1 format. The analysis is adversarial without being destructive: it identifies real vulnerabilities in the argument and acknowledges genuine contributions with equal precision. The analytical reasoning underlying each dimension is recorded in the Analytical Reasoning field of the table.

Dimension Analytical Reasoning Content
Research Question The article does not pose a formal research question; reconstructing it requires inferring a normative question from its prescriptive structure. The alternative interpretation — that this is merely a how-to manual without a genuine intellectual puzzle — is plausible but ignores that Knopf is implicitly responding to a real deficit in methodology pedagogy. The most vulnerable point is the absence of any stated research design or empirical basis for the prescriptions offered. Reconstructed question: How should graduate students and early-career scholars in political science conduct and write a literature review so as to make a genuine contribution to knowledge? Nature: normative-evaluative.
Secondary Questions The Bayesian framing raises an implicit secondary question: What constitutes a contribution to knowledge in the social sciences, and how can one assess marginal epistemic additions? A third secondary question concerns the reliability of non-traditional sources. (1) What constitutes a meaningful contribution to knowledge in social science? (2) How should researchers handle sources from non-academic outlets? Both are subordinate to the primary normative question about review construction.
Puzzle-Type The puzzle is pedagogical rather than theoretical: there is a gap in explicit methodological guidance on literature review construction in political science training. This is a genuine practical gap, but it is not generalizable in the way that an explanatory puzzle in comparative politics would be. The puzzle is narrow and discipline-specific. Descriptive-normative gap: Explicit, codified guidance on how to conduct literature reviews is underdeveloped in political science methods pedagogy compared to the attention given to other research skills (e.g., case study design, causal inference). The puzzle is genuine but limited in scholarly generalizability.
Conclusion / Central Argument The central claim is normative: a good literature review is selective, analytically organized, evaluative, and calibrated to identify consensus, debate, and gaps. The claim of discovery is modest — that framing the review as an exercise in Bayesian knowledge-updating and in identifying three categories (consensus, debate, gaps) provides a useful and actionable framework. What supports this claim is primarily the internal plausibility of the framework and its practical usefulness, not any independent empirical test. A well-constructed literature review must: (a) be organized by schools of thought rather than sequentially; (b) evaluate, not merely summarize; (c) frame the contribution of proposed research in terms of consensus/debate/gaps; and (d) treat the exercise as analogous to Bayesian knowledge-updating. The argument is normative; its support is appeal to pedagogical experience and internal logic.
Methods This is a reflective-prescriptive essay, not an empirical study. The argumentative strategy is deductive: the author moves from a general theory of knowledge (Bayesian framework) to specific prescriptions, bolstered by illustrative examples (agro-terrorism, neo-realism/Waltz). No systematic evidence is provided that the recommended practices produce better reviews. The article incorporates some intertextual borrowing from Paul Pitman and John Odell but does not develop these attributions into a cumulative theoretical argument. Type: normative-pedagogical essay. Strategy: deductive reasoning from a conceptual framework (Bayesian epistemology) to procedural prescriptions. Supporting technique: illustrative examples. No empirical data; no comparative evaluation of review quality.
Data Generation Process (DGP) There is no DGP in the conventional sense. The prescriptions are grounded in the author’s experience and disciplinary common sense, not in systematic observation. The unit of analysis, if one can be inferred, is the individual literature review as a genre artifact — but this unit is never explicitly defined, sampled, or analyzed. No formal DGP. Prescriptions are generated from the author’s experiential knowledge and disciplinary convention, supplemented by informal pedagogical materials (Pitman lecture, Odell handout). No sampling, data collection, operationalization, or formal inference. Inference type: practical reasoning from authority and experience.
Findings and Contributions The article’s strongest contribution is the three-category heuristic (consensus/debate/gaps) and the Bayesian framing of knowledge contribution, both of which provide operational vocabulary for a task graduate students typically approach without clear criteria. The two-tier source model is a practical innovation of limited but real utility. Empirical results: None. Contributions: (1) Operational three-category taxonomy for evaluating the state of knowledge; (2) Bayesian framework for conceptualizing what constitutes a contribution; (3) two-tier source model for managing thin literature; (4) seven practical pointers for structuring the review. The contributions are pedagogically concrete and practically useful, though not theoretically novel.
Critical Analysis of Findings The core vulnerability is circular: the prescriptions are justified primarily by reference to the framework they themselves constitute. There is no external validation. Additionally, the assumption that “consensus” is an identifiable and relatively stable epistemic state is questionable in fields where methodological pluralism is high or where power dynamics shape which claims achieve consensus status. The article’s examples are drawn exclusively from international relations and security studies, raising questions about cross-subfield applicability. The author adequately addresses the practical question but does not adequately address: (a) what happens when a field lacks identifiable consensus because of ideological or disciplinary fragmentation; (b) how to weight incommensurable theoretical frameworks that do not map onto a simple debate/consensus structure; (c) how disciplinary context (e.g., interpretive vs. positivist traditions) affects what counts as a “contribution to knowledge.” Generalizability beyond positivist comparative and IR subfields is unexamined.
Limitations The asymmetry between the empirical richness expected of the reviews being prescribed and the evidence offered for the prescriptions themselves is notable. Acknowledged by authors: None explicitly acknowledged; the article is presented as practical guidance without formal caveats.
Unacknowledged or underestimated: (1) The Bayesian knowledge-updating framework assumes a cumulative, linear model of knowledge progress that may not apply to interpretivist, critical, or poststructuralist traditions; (2) the article does not address how power asymmetries in citation practices shape what becomes “consensus”; (3) the guidance is implicitly positivist and may require substantial adaptation for qualitative, normative, or theory-building research programs; (4) no consideration of how AI-assisted search tools alter the “too many sources” problem.
Theoretical Perspective The implicit ontology is social-scientific positivism: knowledge accumulates incrementally, disagreements can be adjudicated through evidence and logic, and the purpose of research is to advance a shared epistemic project. This is coherent but narrow. Theoretical tradition: Social-scientific positivism; Bayesian epistemology (applied informally). Framework is adequate for the article’s explicit purpose (training students in political science), but its coherence is contingent on the positivist assumptions of the discipline. It would require significant reworking to apply to interpretivist or critical theory traditions, where the very notion of “knowledge contribution” is contested.
Main References The engagement with the methodology literature is thin relative to what the article prescribes for students. The article largely rests on practical experience rather than engagement with epistemological or science studies scholarship that directly bears on knowledge-making. Central works cited: King, Keohane & Verba (1994) on selection bias; McKeown (1999) on Bayesian reasoning in qualitative research; Booth, Colomb & Williams (2003) on research craft; Cresswell (2003), Fink (2005), Galvan (2005), Pan (2004), Patten (2005) on literature review methods. The article does not engage with science and technology studies literature on knowledge production, nor with feminist or postcolonial critiques of epistemic authority. The reference list is practitioner-focused rather than theoretically grounded.
Observations The article will remain useful as a graduate methods primer, but its implicit positivism and its lack of engagement with disciplinary power dynamics limit its critical purchase. For a researcher working on educational policy and inequality — where critical and interpretive traditions intersect with quantitative causal inference — the article is most useful as a scaffolding tool for the technical structure of a review, but insufficient as a guide for navigating ideologically contested literatures or for deciding how to engage with non-academic or policy-facing knowledge production. (1) The absence of critical reflection on whose “consensus” counts is a significant omission in a post-KKV methodological landscape. (2) The two-tier source model is practically useful but does not address the epistemological status of grey literature in policy-facing fields. (3) The article’s prescriptions are most actionable for research proposals in positivist social science; students working in interpretive, normative, or mixed-methods traditions will need additional guidance. (4) The article’s age (2006) means it predates the transformation of the literature search landscape by systematic review methodologies, AI-assisted tools, and preprint ecosystems — significant omissions for a contemporary methods course.