AI Prompts for Research Methods and Academic Writing
Prompts for sharpening research questions, planning a study, drafting questionnaires, organising literature and writing methods and limitations. Use them to think and draft faster, not to replace your own reading and writing.
Use AI responsibly in research
- Read the sources yourself. AI tools can invent references and misstate findings. Never cite a paper you have not opened.
- Follow your institution’s AI policy and declare AI help where required.
- Keep your voice and reasoning. Use AI to suggest options, then decide and write in your own words.
- Do not paste unpublished or confidential material unless the tool’s data terms allow it.
1. Sharpen a research question
My topic is [topic] and my context is [country, population, setting]. Here is my draft research question: [question]. Suggest three sharper versions that are specific, answerable with data and feasible in [time and budget]. For each one, say what data and method it would need.
2. Link questions to methods
My research objectives are: [list]. For each objective, suggest a suitable design (survey, interviews, experiment, secondary data), the variables or topics to measure and the analysis I would use. Present it as a table.
Our questionnaire design guide continues from here.
3. Draft questionnaire items
I need 8 survey items measuring [concept] among [population]. Use a 5-point agree/disagree scale, plain language, one idea per item, and no leading wording. Mark which items should be reverse-coded. Then list possible problems with each item.
Pilot every item with real respondents. For validated scales, use the published wording instead.
4. Review your questionnaire
Review the questions below as a survey methodologist. Flag double-barrelled, leading, vague, jargon-heavy or overlapping items and suggest rewrites. Questions: [paste]
5. Build a literature matrix
I will paste abstracts of papers I have already selected and read. For each, extract into a table: aim, setting, design, sample, key finding, limitation. Use only what the abstract says and write “not stated” when information is missing. Abstracts: [paste]
Verify every row against the full paper before using it.
6. Find gaps and themes across studies
Here is my literature matrix: [paste]. Group the studies by theme, point out where findings agree or conflict, and suggest what is missing (populations, methods, outcomes). Do not add studies that are not in the matrix.
7. Plan the methodology section
Outline the methodology section for a study with this design: [design], sample: [who, how many, how selected], data collection: [tools], analysis: [methods]. Give the headings, what each should cover and typical reviewer questions I should answer.
8. Draft the sampling justification
Write a short paragraph justifying a [sampling method] sample of [n] from a population of [N], including the sample size calculation inputs: confidence [95%], margin of error [5%], expected proportion [50%], and non-response of [20%]. Do not change my numbers.
9. Write limitations and ethics
Based on this study description: [paste], list the likely limitations and the ethical issues a review board would ask about (consent, confidentiality, vulnerable participants, data storage). Suggest how I can address each one.
10. Improve clarity, not content
Edit the paragraph below for clarity, grammar and concision. Keep my meaning, terminology and citations unchanged. Show your changes in a list rather than silently rewriting. Paragraph: [paste]
11. Prepare reviewer responses
Below are reviewer comments and my current manuscript text. For each comment, suggest a polite, specific response and what change to make. Mark any comment where I should add analysis instead of rewording. Comments: [paste]
Tips for better research prompts
- Say who you are, the field and the level, for example master’s thesis in public health.
- State the output you want: table, outline or 150-word paragraph.
- Tell the tool what not to do, such as not inventing references or numbers.
- Ask follow-up questions: “What is the weakest part of this?”
When you reach the analysis stage, our statistical analysis prompts and qualitative analysis prompts continue the workflow. For hands-on support, see questionnaire design and survey data analysis.
Frequently asked questions
Are AI prompts allowed for academic research?
Many universities and journals allow limited, declared use, but rules differ. Check your institution’s policy and disclose how you used AI.
Can AI write my literature review?
It can help organise papers you have read, but it should not be your source of facts. It may invent citations, so read and cite the original papers yourself.
Will AI-written text be detected?
Detection is unreliable, but that is beside the point. Submit work that is your own reasoning and that you can explain and defend.
What is the safest way to start?
Use AI for brainstorming, structure and editing, and keep reading, analysis and conclusions in your own hands.
Related guides and services
Want an expert to check or run your analysis?
Send us your question and data description. We reply with a clear plan and a quote.
Discuss your project on WhatsAppAlso related: EndNote course: reference management · Free data analysis templates