How to Fix Plagiarism in a Thesis: Correct Citations and Paraphrasing

What does fixing plagiarism in a thesis mean

Fixing plagiarism means checking passages in your thesis whose content, wording, or structure is too similar to another source, then dealing with them through proper citation, rewriting in your own argument, or removing material that is not needed. The goal is not to replace a few words simply to lower the similarity percentage. The goal is to make it clear which ideas come from a source, which analysis is yours, and which results you produced yourself.

You should separate three tasks that are often grouped together as “fixing plagiarism.” The first is correcting citation errors, such as using an author's idea without identifying the source. The second is rewriting a literature review section that has been copied almost word for word. The third is handling legitimate matches, such as a university name, scale name, technical term, table title, or reference entry. Each task requires a different response.

If you have a similarity report open, save the original version of your thesis before making changes. The report only shows text areas that the system has identified as similar to sources. You and your supervisor still decide which passages need revision. You can also read how to check Turnitin to understand the difference between a similarity percentage and a plagiarism conclusion.

The standard structure for fixing plagiarism

“Fixing plagiarism” is not a separate chapter in a thesis. It is an editing process applied to the literature review, theoretical foundation, methodology, results, and reference list. When you document your own review process, state the tool used, the date of the check, the content groups reviewed, and how you handled them. Do not insert a similarity percentage into the thesis as though it were a research result unless your institution requires it.

The table below helps you classify the work. The number of lines and the suggested length are practical guidance, not compulsory rules for every institution.

A thesis can have a low similarity percentage and still contain an academic integrity problem if it deliberately uses another person's ideas without citation. A high similarity percentage does not automatically mean plagiarism when most matches are in references, variable names, questionnaire items, or required forms. Read each highlighted area instead of chasing one number.

If you are writing your thesis from the beginning, undergraduate thesis can help you map the sections that need to be completed. For a shorter research project, also review research topic so that you do not copy the structure of a long thesis into a shorter paper.

Steps for writing the plagiarism correction section

Step 1: Save the original and read the report by group

Create a clearly named copy, such as luanvan_sua_v1.docx. Open the similarity report and classify the highlighted passages into four groups: matches caused by citations, matches caused by terminology, matches caused by sentence wording, and matches caused by tables or data. This classification prevents you from changing a model name or technical term that should remain unchanged.

Do not begin by deleting every highlighted passage. A fully cited paragraph can still be highlighted because the software compares character strings rather than the writer's intention. Check the identified source, reread the sentence before and after the match, and decide whether the passage presents background knowledge or your own interpretation.

Step 2: Check the source for every idea

For every highlighted passage, answer three questions: Whose idea is this, where is the source, and are you using the original wording or a paraphrase. If you cannot find the source, do not guess the author's name or publication year. Search again for the article, book, report, or document you used while reading.

A good citation appears immediately after the claim it supports. If a paragraph contains several ideas from several sources, split the paragraph or place citations after each group of ideas. Do not put one citation at the end of a page and make the reader guess which sentence it supports.

Step 3: Rewrite around your study's argument

Close the source document after you understand its main point. Rewrite the idea from memory using your own sentence structure, then reopen the source to check that you have not changed its meaning. Replacing a few synonyms, reversing two clauses, or using an automatic paraphrasing tool can still produce near-verbatim copying.

A useful literature review paragraph usually has three layers: what earlier studies found, where those studies agree or differ, and why that gap matters for your topic. If the paragraph only retells each article in sequence without a connecting sentence or interpretation, the similarity may fall after you change the wording while the argument remains weak.

Step 4: Separate quotations from paraphrases

Use a direct quotation only when the original wording has particular value, such as an official definition or a statement that you need to analyse. Put the quotation in quotation marks, give the source, and include the page number if your institution requires it. For most literature reviews, a cited paraphrase is more suitable.

Do not put an original sentence in quotation marks and change only a few words. This is still easy to identify as copying, and the resulting sentence usually sounds unnatural. When you do not need to preserve the author's exact wording, paraphrase the entire idea and cite the source.

Step 5: Review tables, figures, and indicators

A model table, list of indicators, or measurement scale may resemble many other theses because students use the same source. State the origin of the scale, the context in which you adapted it, and the variable codes used in your study. If you translated or adjusted the wording, describe that process in the methodology instead of presenting the table as though it were an entirely new scale.

For a results table, use only the numbers produced by your own analysis. Tables such as Reliability Statistics, KMO and Bartlett's Test, Rotated Component Matrix, or Coefficients have the same column names across many papers. The part you must write yourself is the data, the interpretation, and the conclusion connected to your hypotheses.

Step 6: Run the check again and keep an editing log

After revising, check spelling, match the citations, and export the report again according to your institution's requirements. Create a log with the page number, revised passage, reason for revision, and related source. This log is useful when your supervisor asks why you removed a source, changed a sentence, or kept a matching phrase.

Do not run the report repeatedly just to find a lower percentage. Each revision should clarify the source and make your argument more independent. If the system highlights a university name, table title, or reference list, check whether your institution allows those sections to be excluded and save the settings you used.

Usable sentence templates

You can use the sentences below as frameworks, then replace the bracketed content with real information. A template does not replace reading the source or checking your data.

How to write this in your thesis

You can adapt the following paragraph to describe your source-control process: “The author reviewed the cited passages in [chapter name], compared them with [number] references, and rewrote paraphrased content whose source was not clear. Technical terms, model names, and scale names were retained where necessary, while citations were added at the corresponding locations.” Fill in the number and chapter name only after you have actually completed the review.

When explaining the work to your supervisor, state exactly what you changed. The sentence “I reduced the similarity percentage” does not show the quality of the revision. The sentence “I checked each source, added citations to the definitions, rewrote the synthesis of previous studies, and retained the variable names from the original scale” is more convincing because it describes the process.

A shortened example of fixing plagiarism

Suppose your topic concerns intention to use a digital banking application. The first draft contains this sentence: “Perceived usefulness is the degree to which a person believes that using a particular system would enhance his or her job performance.” This is an original English definition. Translating it closely without a citation does not turn the idea into your own.

One way to handle it is to identify the source and provide a cited translation: “Based on Davis (1989), perceived usefulness refers to the extent to which users believe that a system can improve their work performance. In this study, the construct is placed in the context of digital banking application use, so performance is examined through [specific indicators].” The bracketed section must be built from your actual model and questionnaire.

Another example concerns the literature review. The first draft may list three studies, with one near-verbatim translated sentence for each study. After revision, you can group the findings by theme: “Previous studies have mainly reported a positive relationship between perceived usefulness and intention to use technology. However, the strength of the relationship varies by context, respondent group, and type of technology. This study tests the relationship with a sample of [population] in [location], adding evidence for the [context] setting.” You must still cite the claims taken from the sources.

Compare the revised passage with the original document to ensure that you have not changed its meaning. If the original study reported only an association, do not rewrite it as a definite causal relationship. If the source used a student sample, do not extend its conclusion to all Vietnamese consumers when your data cannot support that claim.

Mistakes that make the committee ask follow-up questions

The first mistake is using a mass synonym-replacement tool. The revised text may contain fewer matching characters but introduce incorrect terminology, sentences without a subject, or a meaning different from the original. The committee will usually ask follow-up questions when you cannot explain the concept you have written.

The second mistake is deleting every highlighted sentence without checking its role. Removing a definition, hypothesis, or method description can break the logic of the chapter. Rewrite or cite the passage correctly before deciding to remove it.

The third mistake is adding sources to the reference list without using them in the text, or citing a source in the text without providing complete information at the end. Check in both directions. Search each reference-list entry in the thesis body, then check each in-text citation against the reference list.

The fourth mistake is rewriting the results section from a sample paper. SPSS output tables may have the same column names, but the numbers and conclusions must come from your data file. If you use SPSS, SmartPLS, or another tool, save the original output and record the version, variables entered, and processing criteria.

The fifth mistake is focusing only on the overall percentage. Read the sources with the highest matching percentages, the longest continuous matches, and text repeated from your own previous work. Reusing content from a paper you submitted earlier may also need to be checked under your institution's rules.

If you need to place the issue in a wider context, you can read plagiarism. When the thesis structure is still unclear, compare it with the guidance on thesis, while giving priority to your department's formatting and citation rules.

Frequently asked questions

What similarity percentage is considered acceptable

There is no single number that applies to every institution, discipline, and software package. The acceptable level also depends on where the matches appear, which sources were identified, and the rules of your training institution. Check your department's official guidance, then read the report passage by passage instead of setting an absolute threshold yourself.

Do I need to change every sentence when fixing plagiarism

You only need to revise passages that are copied, lack a source, or do not reflect your own argument. Variable names, model names, technical terms, and some specialised phrases may need to remain unchanged. The correct approach is to cite them, place each phrase in the right context, and add your study's analysis.

Should I use paraphrasing software to fix my thesis

You can use spelling tools or software that helps identify difficult sentences, but you should not let software decide the academic content. A tool does not know which source supports a claim, does not understand the sample context, and may replace a technical term with a word that changes the meaning. Read every revision and compare it with the original source.

Why is a cited passage still highlighted in the report

The software compares strings of text, so it can highlight a passage that already has a citation. Scale names, standard definitions, university names, table titles, and references can also produce matches. Check whether the citation is in the right place, whether the quotation is clearly identified, and whether your institution provides instructions for excluding the reference list.

Do I need to change variable names and questionnaire items to reduce similarity

Do not change a variable name or distort an item's meaning simply to reduce the percentage. If an indicator was adapted from a published scale, give the source, describe the translation, and explain the adjustment for your context. The results tables must then use the same variable codes entered in your data file.

What should I do if I cannot find the source I used

Set the passage aside temporarily and search your document folders, browser history, reference manager, or older drafts. If you still cannot identify the source, do not assign an author and publication year yourself. Rewrite the background using a verified source, or discuss the passage with your supervisor before keeping it.

Now create a copy of your thesis, set up the six-step editing log above, and first handle the long passages with clear sources. Then complete the chapter presentation with M5 thesis writing by DoThesis.