Monday, 15 December 2025

A note on citation practice related to Perplexity-generated contents in the context of MBA and Housing Studies dissertation works

A note on citation practice related to Perplexity-generated contents in the context of MBA and Housing Studies dissertation works (information from perplexity.ai)


My question: "Provide 5 illustrative examples of citation practice on Perplexity-generated content in the context of MBA and Housing Studies dissertation report works so as not to make academic dishonest acts."

The reply: "Many universities now allow limited use of tools like Perplexity, but require transparent disclosure, correct citation style, and clear separation between AI support and your own intellectual work. Below are five short, MBA/Housing Studies–oriented examples you can adapt (check your own school’s policy and style guide first).


Example 1 – Methodology note (MBA dissertation)

“In the early stages of this dissertation, an AI assistant (Perplexity AI, 2025 version) was used to help generate preliminary lists of key terms related to ‘dynamic pricing in Hong Kong private housing’ and to identify potentially relevant academic and policy sources. All final source selection, critical evaluation, and analysis were conducted by the author, and every cited work was retrieved and read in its original form.”

If required by your style (e.g. APA), add a reference like:
OpenAI. (2025). Perplexity AI (Version X.X) [Large language model]. https://www.perplexity.ai


Example 2 – Methods section (Housing Studies mixed methods)

“Perplexity AI was used during instrument development to suggest additional survey items on ‘perceived housing insecurity’ among PRH tenants, based on prompts describing the theoretical framework and target population. The researcher refined, rephrased, and piloted all items, and only questions validated through expert review and pilot testing were retained in the final questionnaire.”

Short in‑text attribution:
“AI-assisted brainstorming (Perplexity AI, 2025) supported item generation, but the final instrument was researcher-designed.”


Example 3 – Literature review support (MBA / Housing policy)

“An AI tool (Perplexity AI) was consulted to generate an initial list of articles on ‘institutional investors and build‑to‑rent housing in Asia’ and to provide brief topic overviews. These overviews were used only as background orientation; all arguments in this literature review are based on primary sources that were independently accessed, read in full, and cited directly.”

Optional footnote:
“Perplexity AI was used as a search assistant, not as a source to be cited for housing market findings or theoretical claims.”


Example 4 – Transparency statement on writing support

“Perplexity AI was employed for language editing and suggestions to improve clarity and coherence in several sections (e.g. Chapters 3 and 5).” “No AI-generated text was inserted verbatim into the final dissertation; all substantive content (research design, analysis, interpretation, and conclusions) remains the author’s original work.”

If your institution prefers acknowledgements:
“Acknowledgements: The author used Perplexity AI for grammar checking and stylistic suggestions. Responsibility for all content and any remaining errors lies solely with the author.”


Example 5 – Directly citing an AI-generated explanation

If your supervisor explicitly allows citing AI output (less common for dissertations), one possible APA-style approach is:

In text (e.g. when explaining a generic concept):
“An AI assistant defines ‘generative AI’ as software that produces novel text, images, or code in response to user prompts, based on patterns in its training data (Perplexity AI, 2025).”

Reference list entry:
Perplexity AI. (2025). Perplexity AI (Version X.X) [Large language model]. https://www.perplexity.ai

Across all five examples, the key integrity principles are: disclose that you used AI, state exactly what it did, cite original academic sources instead of the AI wherever possible, and ensure the core thinking and writing are your own."



A collection of blog notes on using chatgpt for research purpose.

No comments:

Post a Comment