Plagiarism in a PhD Research Proposal: Source Audit

A PhD research proposal should distinguish your proposed contribution from the literature, methods, and feedback you used to develop it. Cite borrowed ideas and wording appropriately, and follow the program's authorship rules. A similarity scan can help locate text matches, but it cannot establish that the proposed research is original.

Plagiarism Checker Online | Published October 7, 2026 | Sources checked October 7, 2026

A proposal can have no obvious copied paragraphs and still blur who developed its central idea. You might describe a published framework as though you devised it, present a supervisor's suggestion as your independent plan, or label another team's findings “preliminary results.” A text report won't settle those questions.

This guide concerns a PhD application proposal, not a general proposal-writing template. US programs differ in whether they request one. Check the instructions for your particular application first. Our thesis and dissertation guide covers the larger graduate-writing process; the audit below focuses on contribution and source provenance before admissions submission.

Text originality and research originality are different

Text originality concerns the wording you submit and how you identify borrowed language. Research originality concerns the contribution you propose to make in relation to existing work. You can quote a source accurately and cite it, yet still need to explain what your study would add. Conversely, you can express a borrowed idea in new wording and still need to credit its origin.

Leeds University Business School's proposal guidance asks applicants to explain the research's originality, justify methods through the literature, and cite the literature and contribution appropriately. That's useful guidance from one institution, not a universal application format or word limit.

Don't treat a low similarity score as evidence that nobody has investigated your question. Search the relevant scholarship, read it, and discuss the proposed contribution with an appropriate faculty advisor. If the question closely resembles a published study, explain the relationship rather than disguising the overlap with different terminology.

Audit each proposal section

The following matrix is an original worksheet. It doesn't prescribe which sections your program requires. Apply the rows to the corresponding claims in your document, even when the proposal is short. Record a source location that lets you return to the evidence, not just an author's surname.

Pay particular attention to the gap statement. “No research exists” is a claim about a literature search, not a statement made true by confidence. If your evidence covers a particular population or set of studies, keep the claim within that boundary. Learn how to separate literature synthesis from patchwriting when your draft follows a source's argument too closely.

Read the applicant-explanation column aloud. Could you explain each choice without the borrowed proposal open beside you? If not, return to the method or theory source and work out the reasoning. Changing a few words in another applicant's explanation doesn't supply that understanding.

Borrow a method, not another applicant's explanation

Protocol source, settings and planned changes

Adopting an established method isn't automatically plagiarism. The problem is an unclear or misleading source trail: borrowed instructions without credit, copied rationale presented as yours, or a claimed modification that the draft never identifies. Keep recognized technical names when they're accurate. Replacing them to evade matches makes the proposal less precise.

Original hypothetical two-sentence example, not an actual study: “I plan to use semi-structured interviews, following the interview-design source identified in my methods notes. I will adapt the topic guide to ask how commuter students use evening study spaces, and explain each change after reviewing the source and the program's research requirements.”

This example deliberately contains no invented author, citation or findings. It is not submission-ready: replace the source description with a real citation to work you have read. Its purpose is to separate an adopted approach from the applicant's proposed setting and adaptations.

Another applicant's successful proposal can help you see how a document is organized. It doesn't become a method source, and their explanation of why a method suits their project isn't yours to adopt as personal reasoning. Cite the research that supports your choice and write the explanation for your own question.

Separate proposed results from existing findings

Tense alone doesn't establish provenance. Changing “the study found” to “my study will find” turns a reported result into a prediction, not an original contribution. Use language that identifies who generated the evidence and what your proposal will investigate.

These sentences are illustrative, not reports of real research. Don't copy “preliminary notes” into your application unless such notes exist and you may describe them. If a laboratory team generated the evidence, identify the team and your contribution instead of relabeling it as your individual work.

Check tables and figure captions too. A borrowed graph under a heading called “Preliminary results” can mislead a reader even when the source appears in the bibliography. Identify it where it is encountered, and distinguish published evidence from your planned investigation.

Feedback and other people's contributions

Oxford's graduate application policy, checked October 7, 2026, requires supporting documents to be the applicant's own work unless clearly indicated otherwise. Its guidance distinguishes ordinary proofreading or tutor/supervisor assistance from other guidance leading to substantive changes that need acknowledgment. Those are Oxford's distinctions; don't assume every program uses them.

Portsmouth's proposal guide encourages applicants to refine proposals with potential supervisors while also requiring their own wording and acknowledgment of information or ideas from sources. Feedback and authorship therefore need to be considered together, not treated as opposites.

Keep a simple development record: who commented, what changed, and whether the change concerned grammar, the question, the design or substantive text. If a potential supervisor suggests an entire approach, ask how the application should identify that contribution. Don't present advice as a confirmed collaboration or endorsement.

Check program rules before using professional editing, shared drafting or AI assistance. A disclosure doesn't make prohibited assistance permissible. Where a disclosure is required, our AI disclosure statement guide can help you describe the activity accurately; your application instructions decide what is allowed.

Final application source audit

1. Open each source actually used. Remove references you haven't consulted or identify secondary reliance correctly.
2. Credit borrowed framing, concepts and methodological arguments where they appear.
3. Mark exact borrowed wording as quotations using the required style.
4. Label the proposed contribution separately from existing findings.
5. Check preliminary evidence, your role and any restrictions on sharing it.
6. Handle feedback, editing and AI assistance under the program's rules.
7. Save the reviewed submission version and its source notes.

If material comes from earlier work of your own, check reuse conditions for your own work. If you're adapting a completed thesis, review an existing thesis manuscript for sources and old claims before carrying passages over. Reuse permission and accurate attribution aren't resolved simply by rephrasing.

Use university resources first. After the manual audit and confirmation that outside uploading is permitted, you may compare readable proposal text with public sources and order an optional text pre-check. A match can prompt a useful revision. A low score cannot certify research novelty, applicant authorship or admission readiness.

Frequently asked questions