What Is a Literature Review? How to Write a Research Literature Review

What Is a Literature Review

A literature review in a thesis is a systematic presentation of existing studies, theories, models, and findings directly related to your topic. You read the literature to answer three questions: what have other researchers studied, what approaches and data did they use, and which points remain insufficiently explained.

In a quantitative study, a literature review is not a list of articles summarized one after another. It should lead the reader from the research problem to your model, hypotheses, indicators, and the methods you will use to test them. For example, if your topic examines the intention to use e-wallets, the literature review should connect behavioral theory, factors such as perceived usefulness or trust, previous empirical findings, and the user context you plan to survey.

You can think of the literature review as an argument. Each paragraph should show how much is already known about a concept or relationship. At the end, you identify a gap specific enough to justify your research question. The gap may concern the context, target group, relationship between variables, scale, or inconsistent findings.

A literature review usually has two layers. The first is the theoretical foundation, such as TAM, TPB, UTAUT, SERVQUAL, or another theory suited to your topic. The second is a review of empirical studies, where you compare samples, locations, variables, methods, and findings. Theory helps explain why variables may be related. Empirical studies show the conditions under which that relationship has been tested.

You should distinguish the literature review from the methodology section. The literature review explains why you selected the variables and why the model needs to be tested. The methodology explains how you will collect data, measure the variables, and run the analysis. If a paragraph starts describing questionnaire distribution, data coding, or running SmartPLS too early, you have moved methodology into the wrong section.

Why the Literature Review Matters in Quantitative Research

The literature review is the foundation for building your research model and hypotheses. When a previous study shows that service quality may affect satisfaction, you need to check how the concept was defined, which indicators were included in the scale, and whether that finding fits the population in your own topic.

This section also helps you avoid selecting variables based only on intuition. A variable that seems reasonable in practice is not enough to include it in your model. You need a theory or empirical evidence explaining the relationship. If you put too many variables into the questionnaire without a clear argument, the model becomes long, data collection becomes harder, and the results become more difficult to explain during the defence.

The literature review also helps define the unit of analysis and the research context. A scale used for bank customers may need adjustment when applied to food delivery app users. Differences in industry, age, location, or level of product use may change how respondents understand an indicator.

When reading sources, create a literature review matrix instead of only saving PDF files. Each row can represent one study, while the columns include author and year, context, sample size, independent variables, dependent variables, method, findings, and limitations. This makes repeated findings, contradictory findings, and understudied groups easier to see.

The literature review also prepares the analysis chapter. If you expect to run Cronbach's Alpha, EFA, regression, or SmartPLS, the review should show how the variables have been conceptualized and measured. The material on what a scale is is useful when you examine the connection between theoretical concepts and a set of indicators.

A good literature review does not need to contain every source you find. It needs the sources that directly serve your research question. You may read 30 articles but use only 10 to 15 that are genuinely relevant. The selection criterion is not the number of citations. It is relevance and the source's ability to support your argument.

How Much Literature Review Is Enough

There is no single number of pages or sources that applies to every thesis. An empirical research paper, an undergraduate thesis, and a master's thesis have different requirements. Give priority to your department's guidelines, your program's thesis template, and your supervisor's direct instructions.

A sufficient literature review usually has several signs. First, the sources are grouped by topic or relationship between variables instead of being arranged in the order in which you downloaded them. Second, each group includes comparison rather than only a series of summaries. Third, the concepts in the review match the variables in the model and questionnaire. Fourth, the research gap is written as an argument that can be tested with quantitative data.

The table below is a practical checklist. The thresholds in the table are sourced methodological guidance, while the structural criteria should be checked against your program's requirements.

If you use PLS-SEM, the literature review should prepare the reader for a two-stage assessment: assess the measurement model first, then assess the structural model. This approach is presented in (Anderson and Gerbing, 1988), and the current PLS-SEM procedure is presented in (Hair et al., 2022). Do not include CFI, TLI, or RMSEA indices from CB-SEM in an explanation intended for SmartPLS.

You also need to distinguish academic adequacy from having enough pages. A long literature review with no comparison, no identified gap, and no connection to hypotheses has still missed its purpose. A shorter section with a clear argument, suitable sources, and a direct connection to the model can be more convincing.

How to Read and Organize the Literature

When you open an article, do not begin by copying its abstract word for word. Read it through the questions of your own study and record information that can be used in your model or methodology. You can work through the following steps.

Step 1: Fix the Research Question and Keyword Groups

Write the research question as one sentence containing the population, context, and outcome to be explained. Then create keyword groups for the dependent variable, each independent variable, the theory, and the context. For example, for intention to use a banking app, the keyword group might include intention to use, perceived usefulness, perceived ease of use, trust, mobile banking, and Vietnam.

Do not search only in Vietnamese. Academic sources often use English terminology more consistently. Record synonyms as well so you do not miss studies that use a different name for the same concept.

Step 2: Read the Theory and Model Sections

In each article, mark the definition of the concept, the theoretical foundation, the model diagram, and the argument for each hypothesis. Record separately which indicators the authors used to measure each variable. If an article uses only a concept similar to your variable, note the similarities and differences instead of transferring it directly into your questionnaire.

For models such as TAM, TPB, or UTAUT, identify which components are original and which variables were added in later studies. For example, (Davis, 1989) is a foundational source for TAM, while a study applying TAM to e-commerce may adapt the scale to that context. Do not treat every variable appearing in an applied study as an original component of the theory.

Step 3: Record Methods and Findings Using the Same Template

Record each study in the same order: survey population, sample size, sampling method, scale, software, tests, and main findings. For quantitative research, state whether the study used regression, CB-SEM, or PLS-SEM. This prevents you from directly comparing indices produced for different objectives and methods.

When reading a results table, record both the direction of the effect and the reported significance level. Do not turn a correlation into a causal relationship. Also record statistically nonsignificant findings, because contradictory results often help explain why the relationship needs to be tested again in a new context.

Step 4: Group Sources into Argumentative Paragraphs

After taking notes, remove the “summary” column if it makes you merely retell each article. Instead, group sources by dependent variable, influencing factor, population, or finding. A paragraph should contain a topic sentence, evidence from several studies, a comparison, and a transition to the next point.

For example, a paragraph about perceived usefulness might begin with a definition, compare studies showing an effect on intention to use, and then state that the result may depend on users' familiarity with technology. The final sentence can connect this issue to the reason you will test the relationship among the user group selected for your study.

Step 5: Write the Gap and Research Model

The gap must answer the question, “What is missing, and how will this study test it?” You can identify one or more types of gap: a context that has not been tested, an understudied population, a mediator or moderator that has not been included, a scale that has not been tested in a new context, or inconsistent findings across previous studies.

If your topic includes a mediator variable or moderator variable, explain the mechanism through which you expect that variable to operate. Do not add a mediator or moderator simply because another paper's model includes one. Your argument must show which relationship the new variable explains or changes.

What to Do When the Literature Review Is Not Adequate

If your literature review is currently a chain of individual article summaries, you do not need to rewrite the entire chapter immediately. Mark the topic sentence of each paragraph and write beside it which variable or hypothesis the paragraph supports. Move paragraphs that do not serve the research question to your reference list or remove them.

If you lack sources, search by individual concept rather than by the full title of your topic. Each variable should have a definition source, a scale source, and several relevant empirical studies. Start with foundational sources, then follow the studies that applied or retested the model.

If studies have conflicting findings, do not select the findings that support your hypothesis and omit the rest. Record differences in sample, context, measurement, or method. This may provide a reasonable basis for testing the relationship, but you still need to write the hypothesis from a theoretical argument rather than assuming the result in advance.

If your model contains too many variables, review the scope of your research question. A model with more variables is not automatically more valuable. It makes the questionnaire longer, increases the chance that respondents will abandon it, and makes the results harder to explain. If your data contain outliers, you can review the guidance on handling outliers, but do not delete data simply to make the results fit the literature review.

If your supervisor says that the literature review is “too general,” ask which part lacks specificity: the concept, context, findings, scale, or gap. Then revise the missing section. Adding many citations without changing the argument usually makes the chapter longer without making it clearer.

Distinguishing the Literature Review, Theoretical Foundation, and Research Review

The theoretical foundation focuses on the concepts and theories used to explain the model. The research review focuses on what previous studies did and found. In many thesis templates, the two sections appear in the same chapter, but their functions remain different.

“Literature review” is a broad term that may include theory, empirical studies, scales, and context. “Research review” usually emphasizes previous studies, patterns in the findings, and the research gap. “Theoretical foundation” answers why a relationship between variables may exist.

You should also distinguish the literature review from the reference list. The literature review is the argument inside the thesis. The reference list is the complete list of cited sources. A source included in the reference list but never used in the argument may lead your supervisor to ask why it is there.

The literature review also differs from the results section. In the review, you report previous studies' findings as existing evidence. In the results chapter, you present findings from your own data file, such as the KMO and Bartlett's Test, Rotated Component Matrix, Coefficients, or Path Coefficients tables. Make the source of each result clear so readers do not confuse the two types of evidence.

Common Errors

The first error is writing in the style of “Author A says, Author B says, Author C says” without comparison. After two or three studies, add a sentence explaining the agreement, difference, and implication for your model.

The second error is using a source that is not directly relevant simply because its title contains a keyword similar to your topic. Read the method and sample to check its fit. A study of purchase intention may not provide enough support for actual usage behavior if the two concepts were measured differently.

The third error is citing a theory without using it to create hypotheses. If you selected TPB, TAM, or UTAUT, identify which theoretical concepts appear in your model and explain why they can be adapted to your context.

The fourth error is copying wording or machine-translating an entire scale. Read the original definition, adjust the language for your respondents, and state clearly that the scale is a reference version before testing it. Reviewing what a scale is helps you separate concepts, latent variables, and indicators in your writing.

The fifth error is writing an excessively broad gap, such as “no study has examined this topic.” A statement like that is difficult to prove. Narrow it to a specific context, population, variable, or relationship that the literature you reviewed has not fully explained.

The final error is using different variable names in the literature review, model, and questionnaire. Create a cross-reference table containing the Vietnamese name, English name, variable code, and scale source before finalizing the chapter. This reduces errors when you later enter the data into SPSS or SmartPLS.

Frequently asked questions

What is a literature review in a thesis?

A literature review is the section that organizes theories and previous studies related to your topic. It explains the foundation of the model, the choice of variables, the evidence for relationships, and the gap that your study will test.

How many pages should a research review be?

There is no fixed page count for every program. Follow your department's template and your supervisor's instructions. In terms of content, the section is adequate when relevant groups of sources have been compared, the model variables have a basis, and the gap leads to the research question.

Does every sentence in a literature review need a citation?

You need a citation when defining a concept, presenting a theory, reporting a finding, or using a scale from a previous study. A paragraph synthesizing several sources can place citations at the end of a sentence or point, but the reader must be able to tell which source supports which claim.

Should a literature review be organized by author or by topic?

Organize it by topic, variable, or relationship in the model. Authors and years should appear as evidence within each group. Organizing the entire review by author usually creates a list of summaries and makes comparison and the research gap difficult to show.

What should I do if I cannot find a research gap?

Create a comparison table covering the context, population, variables, scales, methods, and findings of the studies you have read. The gap may involve a different survey group, location, untested relationship, or inconsistent findings. You do not need to claim that the entire field has never been studied.

Can I use another thesis's literature review?

You can use another thesis to find keywords and original sources, but you should not copy its argument or reference list without reading the original sources. Check the source, context, method, and citation before including it in your own thesis.

Open your thesis file and create a table with the source, concept, variable, method, finding, and gap. Then reorganize the paragraphs around your research model. If you need help turning those notes into a cited, structured literature review chapter, use M5 thesis writing from DoThesis.