What Is Citation? How to Cite Sources Correctly in a Thesis

What Is Citation in a Thesis

Citation is the way you identify which ideas, data, definitions, or findings in your writing come from a published source. That source may be a journal article, a methods book, an official report, or a thesis that you have read and checked. In a quantitative thesis, citations usually appear in the theoretical foundation, research model, measurement scales, methods, and discussion of results.

You can think of a citation as having two parts. The first is a short marker placed immediately after the sentence or paragraph that uses the source, usually consisting of the author's surname and publication year, such as (Hair et al., 2010). The second is the full reference information at the end of the thesis. These two parts must match. A source listed in the references but never cited in the text may look like a reference added only for formality. A source cited in the text but missing from the reference list prevents the reader from checking it.

What citation means in practical thesis writing also depends on the type of content you use. If you explain an author's idea in your own words, that is an indirect citation. If you keep the original wording, that is a direct quotation and usually requires quotation marks and a page number under your university's rules. For data in .sav or .csv files that you collected yourself, you do not cite someone else for those numbers. You still need to cite the source of the scale, formula, or analytical criterion you used.

Treat citation as part of your argument rather than as formatting work to complete at the end. Every sentence that gives a definition, evaluation threshold, or conclusion from previous research needs to be checked for an appropriate source. This makes it easier to answer when your supervisor asks, “Where did this number come from?”

The standard structure of a citation section

A strong citation section in a thesis is more than a list of author names. Organize sources according to the function of each paragraph, then present them consistently in the style required by your department or supervisor. The table below gives you a practical structure for checking a draft.

Your university may require APA 7, Harvard, or an internal format. Do not mix several styles in the same reference list. For example, one reference may place the year in parentheses after the author's name in APA style, while another places the year at the end according to a different style. This inconsistency often leads to revision requests even when the research itself is sound.

Before you start, open your department's thesis formatting guide. If the rules are unclear, ask your supervisor about three points: the in-text citation style, how to record electronic sources, and how to arrange sources in English and other languages. You should also keep a source-tracking table with columns for source ID, author, year, title, URL or DOI, the section used, and the location in your draft.

Steps for writing your citations

Step 1: Identify which sentences need sources

Read each paragraph and mark sentences that do not come from data you analyzed yourself. A sentence such as “perceived value affects purchase intention” needs support from previous research. A sentence such as “the sample included 250 valid responses” reports your own data and does not need an external citation, although the way you selected the sample size or excluded questionnaires may need one.

Even common statements should be checked in an academic context. You can cite the definition of Cronbach's Alpha with (Cronbach, 1951). The criterion that Alpha at 0.7 or above is acceptable can be cited with (Nunnally, 1978), while Alpha from 0.6 may be acceptable for a new or exploratory scale according to (Hair et al., 2010). Do not place one general source at the end of a long paragraph when it is unclear which sentence that source supports.

Step 2: Choose the appropriate original source

Prioritize the original source for a model, scale, or method. If you use SERVQUAL, look for the original article rather than citing only a thesis that reused the scale. If you use TAM, an appropriate theoretical source is (Davis, 1989). For TPB, use (Ajzen, 1991). For UTAUT, you can check (Venkatesh et al., 2003).

Methodological sources must also match the procedure you performed. When writing about observations per item, the permitted source is (Hair et al., 2010). When writing about regression sample size based on the number of independent variables, you can use (Tabachnick and Fidell, 2013). When writing about Durbin-Watson in regression diagnostics, use (Field, 2013). Read the relevant part of the source instead of selecting it only because it appears frequently in other theses.

Step 3: Rewrite the idea in your own language

Indirect citation is often more suitable for the literature review. Read the main idea, close the source, write the sentence according to your own argument, and then compare it with the source again. This keeps the paragraph connected to your research instead of turning it into a chain of translated sentences from several papers.

For example, instead of copying a full definition of perceived value, you could write: “In this study, perceived value is understood as a customer's overall evaluation of the benefits received compared with the cost paid.” Place the source that matches the definition you used after the sentence. If the wording remains almost identical to the original, use a direct quotation according to your university's rules.

Step 4: Record the citation while writing

Do not wait until the end of the chapter to add sources. When you create a new paragraph, you can use a temporary code such as [Davis1989] in your working file, then convert it to the required style when the draft is complete. For every code, save the full source information as well. This prevents the situation where you remember the idea but no longer remember which document supplied it.

If you use reference-management software, check the author names and Vietnamese diacritics after importing them. The software may misread metadata, capitalize a title incorrectly, or place a Vietnamese author's name in the wrong field. The software can help you import and synchronize references, but you are still responsible for deciding whether the source actually fits your argument.

Step 5: Compare the text with the reference list

After finishing a chapter, create two lists. The first contains every citation in the body. The second contains every source at the end of the thesis. Compare them by author surname and year, then check cases where the same author has several publications in one year.

You should also open a few sources at random. Does the link lead to the correct article? Does the article actually discuss the point you used? Are the journal title and publication year correct? A plagiarism check may detect repeated wording, but it does not replace checking the quality of your sources. You can consult the guide to checking Turnitin and read more about plagiarism when reviewing a draft before submission.

Reusable sentence templates

You can replace the text in square brackets with information from your own study. These sentences are writing frames, not ready-made results for your research.

When citing several sources, include only documents that genuinely support the same point. Five references do not automatically make a sentence more reliable if they concern different contexts or variables. One clear, accessible source that matches the sentence is more useful than a long list added to make the writing look academic.

You also need to distinguish the author of a model from the author of a paper you found. A recent study may apply TAM, but that does not make it the original source of TAM. In the theoretical foundation, cite the original source. In the literature review, cite the applied study to show how the model has been tested in a similar context.

A shortened citation example

Suppose your topic concerns factors affecting students' intention to use e-wallets. You build a model with perceived usefulness, perceived ease of use, and usage intention. A paragraph could follow this sequence:

“Davis (1989) proposed the Technology Acceptance Model to explain users' acceptance of technology. Under this approach, perceived usefulness reflects the extent to which users believe that a technology improves their task performance, while perceived ease of use reflects the extent to which users believe that using the technology requires little effort. In the context of e-wallets, this study adopts these two concepts and tests their effects on students' usage intention.”

This paragraph performs three clear tasks. The first sentence identifies the theoretical origin. The next two sentences explain the concepts according to the selected source. The final sentence connects the theory to the topic without claiming that the hypothesis has already been supported. If you propose a hypothesis, the next section needs to add suitable empirical evidence, such as: “Based on the argument above, the study proposes H1: Perceived usefulness has a positive effect on students' intention to use e-wallets.”

In the methods section, the writing can be more specific: “The scale was assessed using Cronbach's Alpha and Corrected Item-Total Correlation. Alpha at 0.7 or above is considered acceptable according to (Nunnally, 1978), while Corrected Item-Total Correlation from 0.3 or above is used according to (Nunnally and Bernstein, 1994). For exploratory research, a lower Alpha threshold may be considered according to (Hair et al., 2010), but the decision should also take into account the number of items and the content meaning of the scale.”

How to write this in your thesis: “Based on [theory/scale source], the study adopts [number] items to measure [construct] in the context of [target group]. After adjusting the wording and checking the data, the scale is assessed using [method], with the criterion [threshold] according to ([author, year]).”

The numbers in the example illustrate the structure of the argument. Replace them with the actual number of items, results, and sources in your file. Do not write that a scale has acceptable reliability before running Reliability Statistics and Item-Total Statistics on the actual data.

If your study uses SmartPLS, the criteria must also be supported by the correct group of sources. For example, outer loading from 0.7 or above can be discussed with (Chin, 1998) or (Hair et al., 2022), CR from 0.7 and AVE from 0.5 can be cited with (Fornell and Larcker, 1981), and HTMT below 0.85 or 0.90 depending on how close the constructs are can be cited with (Henseler et al., 2015). Do not include CFI, TLI, or RMSEA in the SmartPLS assessment section, because those indices belong to CB-SEM according to (Hu and Bentler, 1999).

You can add internal links when readers need a broader section on undergraduate thesis, want to view the Undergraduate Thesis topic, or want to compare the implementation with a master's thesis. These links support navigation. They do not replace the academic sources that must appear in the reference list.

Common mistakes that lead to follow-up questions from the committee

Citing a secondary source as the original source

You read a thesis about a model and record the thesis author as the person who proposed the model. When questioned, you cannot point to the original article. Trace the secondary source's reference list and check the original document before crediting it as the source.

Listing sources without linking them to a point

A long paragraph that places five references at the end without showing which source supports which sentence is difficult to defend. Break the paragraph into smaller points, place the citation immediately after the relevant sentence, and remove sources that do not fit.

Using one threshold for every study

Alpha, factor loading, outer loading, AVE, and HTMT belong to different assessment steps. State clearly which index you are using, whether it comes from SPSS or SmartPLS, and which source supports the threshold. In particular, do not take a CB-SEM criterion and apply it to PLS-SEM.

A reference list that does not match the text

This happens when you delete a paragraph but forget to delete its source, or add a citation without updating the reference list. Check manually after every chapter revision instead of waiting until the night before submission.

Copying wording and only adding a source

A citation does not turn copied wording into a paraphrase. If you keep the original sentence, mark it as a direct quotation according to your university's rules. If you do not need to preserve every word, understand the source and rewrite it using your own sentence structure.

Citing a source that does not exist

A link found online may have the wrong article title, publication year, or no original document. Do not include a source merely because its title appears relevant. Check the author, year, title, and actual content before using it.

Frequently asked questions

What is citation, and why does a thesis need citations?

Citation is the way you acknowledge the source of an idea, concept, data point, or method that you use. It lets readers check your argument, distinguish your findings from earlier research, and reduce the risk of being accused of plagiarism.

What makes a citation correct?

A correct citation matches the content in the body with the source in the reference list, follows the style required by your university, and identifies a source that genuinely supports the sentence. Record the citation immediately after the point, then check the author's name, year, and full source information in the reference list.

Does every sentence in a thesis need a citation?

Results that you collected and analyzed yourself, such as the number of valid questionnaires or a regression coefficient in your SPSS file, do not need an external citation. Definitions, models, scales, formulas, sampling criteria, and evaluation thresholds usually do need appropriate sources.

What is the difference between direct and indirect citation?

A direct citation keeps the source's exact wording and usually requires quotation marks and a page number under your university's rules. An indirect citation explains the source's idea in your own words while still giving the author and year. In a thesis, paraphrasing usually creates a smoother argument as long as the meaning is not changed.

Can one sentence have several citations?

Yes, when several sources support the same point and you have read them. Avoid adding references only to make a sentence look academic. If the studies report different findings, state the difference rather than placing every source at the end as though they reached the same conclusion.

What should I do if a reference is missing after writing?

Review every citation in the body, make a list of authors and years, and compare it with the reference list. For a source that cannot be verified, find the original document or remove the claim. Once you have run the analysis and started writing the chapters, recording sources directly in the working file will reduce this problem.

Open your draft, choose one chapter you are writing, and check each paragraph with three questions: where did this point come from, does the citation match, and is the full source in the reference list. If you need to standardize your sources, connect the citations to the model, scales, and questionnaire in DoThesis M3 model, hypotheses, scales, and questionnaire.