Before writing a bachelor's, master's or doctoral thesis in China, students submit a thesis proposal and defend it in front of a panel. Panels reject proposals for the same reasons: a topic too broad to finish, a literature review that lists papers without synthesis, research questions that do not match the methods, and a schedule with no slack. This skill builds a proposal that is narrow enough to finish and argued well enough to pass.
Write in Simplified Chinese unless asked otherwise. The university's own template and word limits override this structure. Never invent references: every citation must come from the student's own reading or a source they confirm.
Required Inputs
Ask for these if not provided:
- Degree level (undergraduate, master's, doctoral), discipline and the university's template, if any
- Topic idea, and the supervisor's guidance
- Papers already read, with full details, or the databases the student can use (CNKI (China National Knowledge Infrastructure), Wanfang, Web of Science)
- Data, equipment or field access the student actually has
- Deadlines: proposal defence, mid-term check, submission
Output Structure
1. Topic narrowing
Turn the idea into a researchable question: object, variable or problem, method, scope (place, period, sample). Offer two narrower versions and say which is safer to finish in time.
2. The thesis proposal
- I. Research background and significance: the problem, why it matters now, theoretical and practical significance
- II. Research in China and abroad: grouped by theme, not paper by paper; ends with the gap this study fills
- III. Research content and objectives: three to five research contents that answer the question
- IV. Research methods and technical route: methods mapped to each content; the technical route as numbered steps (a text version of the technical route diagram)
- V. Innovations: one to three, each specific and defensible
- VI. Existing research groundwork and resources: what the student already has
- VII. Schedule: a table by month with milestones and a buffer
- VIII. Expected outcomes
- IX. References: GB/T 7714 format; see
cn-citation-gbt7714
3. Literature review plan
Search terms in Chinese and English, databases, inclusion criteria, and a reading matrix: | Reference | Research question | Method | Conclusions | Relationship to this study |
4. Defence preparation
The likely panel questions with answer points: why this topic, is the scope feasible, why this method, what is new, what if the data is not available.
Quality Checks
- The research question names an object, a method and a scope
- Every research content is matched to a method in section four
- The literature review is grouped by theme and ends with a stated gap
- Every innovation point is specific enough to be checked
- The schedule has a buffer before each university deadline
- No reference is invented; unconfirmed sources are marked to be verified
Anti-Patterns
- A topic the size of a field. "the application of artificial intelligence in education" cannot be finished; narrow it.
- A list of abstracts as a review. Synthesis by theme is what panels look for.
- Innovation by adjective. "the first systematic study" claims need evidence that nobody has done it.
- Fabricated citations. Panels check; a fake reference can fail the whole proposal.
Example Trigger Phrases
- "Help me write a master's thesis proposal on the influence of short videos on university students' consumption behaviour."
- "What will the panel ask at my proposal defence? Help me prepare."
- "My literature review reads like a chronological list. Help me reorganise it by theme."
- "Help me write a Chinese university thesis proposal."