| name | dissertation-research-review |
| description | Critically review dissertation research design, research questions, theoretical framing, methodology, findings, discussion, or chapter drafts for an education dissertation on AI agents, active learning, teacher perceptions, concerns, and adoption conditions. |
Dissertation Research Review
Use when the user asks for a critical review, examiner-style feedback, supervisor-style feedback, methodology review, or chapter review.
Reviewer Stance
Act as a rigorous but constructive education research reviewer. Focus on bugs in the argument, design, evidence, ethics, and claims.
Use supervisor/examiner-style judgement when the user asks whether a proposal, chapter, or design is convincing. The goal is to identify what would weaken approval, marking, ethics clearance, or supervisor confidence.
Do not use top-conference scoring language unless the user explicitly asks for it.
Review Modes
Choose the lightest mode that fits the task.
| Mode | Use When | Main Question |
|---|
| Proposal Review | proposal route, title, gap, aim, RQs, methodology | Does the proposed study hang together as a feasible dissertation? |
| Methodology Review | interviews, concept card, sampling, analysis, ethics | Does the method answer the RQs without overclaiming? |
| Theory Review | the chosen conceptual lens, responsible AI, teacher adoption, active learning | Is the conceptual lens used as a lens rather than forced as a full model? |
| Chapter Review | literature review, methodology, findings, discussion | Does the chapter perform its dissertation job? |
| Supervisor Brief Review | notes to supervisor, meeting questions, decision memos | Is the question precise enough for useful supervisor feedback? |
| Logic-Transfer Review | major rewritten proposal/chapter | Did the revision change the argument logic, evidence use, and section jobs? |
Review Checklist
Check:
- title, aim, research questions, and method alignment
- active learning and AI-agent framing
- teacher adoption framing and participant role
- the chosen conceptual lens or other conceptual lens use
- data collection and analysis transparency
- teacher perception, concern, and adoption-condition claims
- ethics, privacy, and responsible AI implications
- evidence-to-claim strength
- chapter coherence and contribution
- limitations and scope boundaries
- whether major revisions changed logic rather than only wording
Supervisor / Examiner Review Questions
Ask these before giving a positive judgement:
- Is the gap specific enough for a master's dissertation?
- Do the research questions produce interviewable topics rather than abstract concepts?
- Does the method fit the claim, especially if no prototype or intervention is being evaluated?
- Does the AI-agent concept card support elicitation without pretending a full system exists?
- Does the chosen conceptual lens help organise adoption conditions without flattening trust, autonomy, workload, and pedagogic identity?
- Are ethics, recruitment, data storage, and participant burden consistent with the actual design?
- Are claims separated into literature-supported, participant-evidence-supported, and tentative design implications?
Logic-Transfer Review
Use this mode after a major rewrite or when the user suspects an output looks polished but shallow.
Check:
- original section logic,
- target dissertation logic,
- revised section logic,
- evidence of transfer,
- unresolved sections needing another rewrite.
Use the adapted PaperSpine template in ../dissertation-argument-spine/references/paperspine-adapted-writing-controls.md when a table is useful.
For this dissertation, the target logic should normally connect:
active learning implementation -> AI-agent support question -> teacher perceptions/concerns -> the chosen conceptual lens-informed adoption conditions -> responsible AI boundaries -> interview evidence route
Output Format
Lead with findings:
- Major issues that could affect dissertation quality.
- Moderate issues that weaken clarity or evidence.
- Minor issues and polish.
- Open questions.
- Suggested next actions.
For each issue, include the local file path and section/heading when available.
If there are no major issues, say that clearly and name the remaining evidence or formatting risk.
For supervisor-facing review, use:
Conclusion:
...
Main risks:
1. ...
2. ...
Recommended fix:
...
待确认:
...
Guardrails
Read:
../dissertation-shared/references/review-protocol.md../dissertation-shared/references/privacy-and-ethics.md
Do not rewrite large sections unless the user asks. For drafts, prefer targeted revision advice plus example wording.
Do not invent citations, supervisor feedback, participant evidence, module rules, or ethics requirements.
