What makes a literature review publishable?

Collated and edited by Ross Woods, 2026

In many fields, the most influential literature reviews are not remembered because they summarized hundreds of papers. They are remembered because they changed how researchers think about the field by offering a new conceptual lens, an explanatory model, or an organizing framework that subsequent research adopted.

One of the biggest challenges in publishing literature reviews as journal articles is to create an original contribution to knowledge, not simply a summary of what others have written. Fortunately, originality in a literature review does not usually come from collecting new data. It comes from producing a new understanding of existing knowledge; it also provides a way for a literature-only study to make a new contribution without gathering new field data. However, not all original contributions are equally significant, and the kind of method used tends to affect the size of the contribution. For example, a simple summary normally offers little, an insightful critique offers more, a new conceptual framework even more, and a theoretical model even more.

Here are the main ways a literature review can make an original contribution.

1. A simple summary

Some students simply want to create a comprehensive summary of recent research. It might be valuable if the field is so large that a summary is a major achievement by itself, especialy of the outline creates a structure that classifies component parts. However, most (perhaps all) PhD supervisors consider it inadequate if it has no critique or does not identify a research gap.

2. Identify a genuine gap

The most common approach is to show the current state of knowledge on the topic and show that something important has not yet been studied. In fact, this is the minimum for a dissertation. The contribution is not merely saying "nobody has studied this." The review should explain why the omission matters, which parallels the “Significance” section in the introduction. Examples include an understudied population, an understudied country or culture, a neglected educational level, a new technology, a recent policy change, and a new theoretical perspective.

3. Critique assumptions

Many disciplines inherit assumptions that few people question. If you have reasons to doubt current assumptions, a literature review can examine whether the evidence actually supports these assumptions. In education, for example, everyone might assume that “technology improves learning.” “Active learning is always superior” or “collaboration always helps students.”

4. Develop a research agenda

Some journals publish review papers mainly because they provide an outstanding road map for future research. This becomes a significant contribution.

Rather than merely listing gaps, the review explains which questions matter most, which methodologies are needed, which theories should be integrated, which populations deserve attention, and which kinds of researches might be prerequisite for others.

5. Historical review: Trace the evolution of ideas

Rather than asking “What do we know?” you might ask “How has thinking changed?” Many fields have never had a comprehensive historical review. This kind of review makes a contribution by tracing the evolution of ideas. Identifying turning points, and explaining why and how current practice developed. This can reveal why researchers now ask different questions. For example, you might explore early research, reasons for particular watersheds, paradigm shift, current consensus, and emerging challenges.

Results of comparisons

All the next group of new findings are primarily the result of comparing different views and documents.

6. Evaluate methodological quality

Rather than reviewing findings, review how research has been conducted. For example, sampling methods, qualitative techniques, statistical methods, and measurement instruments. This makes a contribution by identifying methodological weaknesses, explaining different results, and recommending improved research designs. This identifies limitations affecting the entire field. For example, you might discover that:

7. Discover contradictions and inconsistencies, and produce a new synthesis

Resolving contradictions is a valuable scholarly contribution, and these papers are often highly cited because researchers want clarity.

Many fields contain different explanations for findings or conflicting findings for example: “Half the studies report positive effects and half report no effect.” Instead of saying “Study A found X.” and “Study B found Y.” you might ask “What broader principle explains both?” The synthesis becomes the contribution. A literature review can investigate “Why do these studies disagree?”, “Were different methodologies used?” “Were different participant groups studied?” and “Were different definitions being used?” For example, you might find that: “Previous studies appear contradictory because they measure different outcomes at different stages of learning.”

8. Reorganize existing knowledge into a new classification or taxonomy

Sometimes the literature has become fragmented, and a new classification itself becomes a contribution. A review may classify studies by sets of assumptions, by methodology, by discipline, or by country.

A literature review might also re-organize the field according to a more meaningful principle. In education, for example, instead of classifying studies by year, you could classify them according to educational philosophy, assumptions about learning, degree of student autonomy, or level of technological integration.

Taxonomies are among the most influential review contributions, and a useful classification can shape research for years. For example, instead of classifying studies by topic, classify them by:

9. Compare disciplines

Different disciplines often solve similar problems differently. Comparing them can reveal useful ideas that have not crossed disciplinary boundaries. For example, researchers in education, psychology, management, and medicine, might all study feedback.

10. Compare countries or cultures

Research may exist in many countries but remain isolated. A review might reveal important contextual differences. For example, one might compare Western studies, Asian studies, African studies, and Latin American studies.

Comparisons and theory-building

By comparing existing literature, researchers can build new theory.

11. Identify patterns nobody has noticed

A careful review may reveal relationships hidden across many papers, where writers of individual papers have never noticed these broader patterns. For example, certain methodologies consistently produce stronger results, quantitative researchers might ignore issues revealed in qualitative studies, or that positive findings occur only in particular contexts. Two quite different fields of study might unknowing operate on analogous theoretical assumptions, creating an opportunity to produce a new theoretical model explaining the similarities.

12. Build a new conceptual framework

Instead of simply reviewing existing theories, combine them into something new. Many highly cited review papers are remembered for introducing a useful framework. If your review develops or integrates theory, this can become a conceptual paper.

Sometimes a section of the literature review naturally develops into a framework. Framework papers often receive considerable attention. For example, instead of discussing twenty variables independently, develop a model showing how they interact. For example, instead of discussing Theory A, Theory B, Theory C independently, you could develop a model showing how they interact. This might produce a new model, a new taxonomy, a new process, or a new framework.

Introduce a higher-level theoretical model

Sometimes many small theories can be explained by one broader theoretical model. In education, for example, rather than separately discussing motivation, engagement, and persistence, you might propose a higher-level explanation linking them together.

Conceptual frameworks and "theoretical models"

These are closely related, but they are not the same. In fact, one often grows out of the other. Here's a useful distinction.

Conceptual framework Theoretical model
Organizes concepts and relationships relevant to a particular problem Explains how or why those relationships occur
Often combines ideas from several theories Usually derived from or extends a specific theory (or set of compatible theories)
Can be descriptive Should have explanatory or predictive power
May simply identify important variables Specifies mechanisms linking those variables
Often unique to one study May be generalizable to many studies

Conceptual framework

A conceptual framework is essentially a map of the important concepts in a field and how they appear to relate. Suppose you review 300 papers on doctoral completion. You might conclude that completion is influenced by:

You organize these into a framework showing likely relationships. That is already a scholarly contribution if your organization is new or more coherent than previous work.

Theoretical model

A theoretical model goes a step further. Instead of simply saying these factors are related, it explains why, and proposes an explanatory mechanism. For example:

The model can often generate hypotheses such as:

Consequently, a theoretical model has greater explanatory value than a conceptual framework.

How one becomes the other

A literature review can progress through stages.

Stage 1 — Summary
E.g. Researchers have identified many factors affecting doctoral completion.
↓
Stage 2 — Conceptual framework
E.g. These factors can be organized into six major domains with clear relationships.
↓
Stage 3 — Theoretical model
E.g. The relationships occur because self-efficacy acts as the central mechanism linking supervision, motivation and persistence.
↓
Stage 4 — Empirical testing
The proposed model is tested using quantitative or qualitative research.

Many highly cited papers stop at Stage 2 or Stage 3 without collecting any new data. After Stage 4, it might eventually progress to wider acceptance as a theory, when many different studies have supported it, often using different methods and populations. A PhD dissertation is only one of these studies.

This distinction is particularly useful when supervising doctoral students. Many students assume that a literature review can only identify "gaps." In reality, a rigorous review can also construct knowledge. One progression that often works well is:

  1. Identify recurring concepts across the literature.
  2. Organize them into a new conceptual framework.
  3. Explain the relationships among those concepts by proposing a theoretical model.
  4. Use the dissertation's empirical research to evaluate, refine, or extend that model.

That approach allows the literature review itself to make an original scholarly contribution before any data are collected, with the empirical phase providing a second, complementary contribution.

Criteria for a conceptual framework

Many published conceptual frameworks are little more than boxes and arrows. They may look impressive, but a conceptual framework is judged by the quality of its thinking, not by the appearance of the diagram. The diagram is simply a way of communicating the framework.

A good conceptual framework should satisfy several criteria.

Criterion Questions to ask
Clarity Are the concepts clearly defined?
Completeness Does it include all the important concepts?
Parsimony Is it as simple as possible without omitting essential ideas?
Logical coherence Do the relationships make sense?
Evidence Is every relationship supported by the literature?
Originality Does it organize knowledge in a genuinely new or better way?
Utility Does it help researchers think differently or conduct better research?
Testability Can future researchers investigate the proposed relationships?
Generalisability Does it explain more than only its original context?

Let's look at these in more detail.

1. It should solve a problem

A worthwhile framework addresses a gap.; it should answer a question that existing frameworks do not answer well. For example:

2. It should organize complexity

A conceptual framework reduces a confusing literature into something understandable. Imagine reviewing 400 papers. Rather than listing fifty variables, you discover they naturally fall into six domains. That organization itself is valuable; the reader should think, "Now I finally understand how this field fits together."

3. Every concept should earn its place

Every box should have a reason for existing. A common weakness is including every variable mentioned in the literature. Instead, ask:

4. Relationships should be justified

Arrows are not decorations; every relationship should have support. If the literature is contradictory, the framework should acknowledge that.

Instead of Leadership → Innovation the review should explain why dozens of studies indicate that leadership influences innovation.

5. The framework should reveal something new

A framework with explanatory power should produce an "Aha!" moment. Perhaps the strongest criterion is whether the reader learns something they had not previously recognized. Examples include:

6. It should be internally consistent

The concepts should not contradict one another, and definitions should remain stable. Relationships should be compatible, and levels of analysis should not become mixed accidentally. For example: individual motivation, organizational culture, and national policy might all belong in one framework, but their interactions must be explicit.

7. It should be parsimonious

Einstein supposedly remarked: "Everything should be made as simple as possible, but not simpler." Many frameworks contain far too many boxes. A framework with twelve carefully chosen concepts is often stronger than one with fifty.

9. It should have a defined scope.

Rather than asking only "Does the model explain the original context?", ask:

Thinking of scope in terms of boundary conditions is particularly valuable. Every good theoretical model should make clear not only where it works, but also where it should not be expected to work. Explicitly defining those limits makes the model more scientifically rigorous and easier for future researchers to test, refine, and extend.

10. If it has progressed to being a theoretical model, it should also generate and guide future research

Readers should immediately think, "I know how I could test this." A model is distinguable by being testable and generative. After reading it, researchers should not simply say, "That makes sense." They should be able to ask new questions, design new studies, or reinterpret existing findings because of it. That ability to generate further discussion and research is one of the strongest indicators that a conceptual framework is making a genuine contribution to knowledge.

A good framework naturally suggests:

Criteria for an exceptional conceptual framework

The best frameworks do more than summarize—they change how people think about a field. Instead of merely collecting concepts, they reorganize knowledge in a way that becomes the new reference point for future research. They are powerful not because of the diagram itself, but because each one offered a simple, coherent explanation that researchers found useful across many studies. Classic examples include:

A possible grading rubric

A conceptual framework could be graded against five progressively higher standards:

Level Characteristics
Poor A collection of concepts with little organization; arrows are largely unsupported.
Adequate Concepts are organized logically and supported by the literature, but the framework is mostly descriptive.
Good Relationships are well justified, the framework is coherent, and it helps readers understand the field more clearly.
Very Good The framework integrates previously separate ideas, resolves inconsistencies, or offers a clearer organization than existing models.
Excellent / Original Contribution The framework changes understanding of the field by proposing a novel organization of concepts that is theoretically sound, evidence-based, parsimonious, and useful for guiding future research. It becomes a platform that other researchers can build on.

CC BY-NC-ND
This work is released under a CC BY-NC-ND license, which means that you are free to do with it as you please as long as you (1) properly attribute it, (2) do not use it for commercial gain, and (3) do not create derivative works. Link
Disclaimer: Artificial Intelligence (AI) was used in some parts of this book. If AI plagiarized your content, contact the author with evidence to initiate changes.