Turning a vague request into a research question
Listen to this lesson
Nigerian voice narration
Fictional training case. A Research Officer receives a vague brief from the Director-General: 'something on AI and productivity in the civil service,' due in a month. This lesson covers converting vague requests into specific, answerable research questions with defined scope and explicit exclusions. Work through the example, identify the evidence needed and practise the method with cleared or synthetic inputs. Any additional figures in the explanation are separate practice examples, not claimed rows in a downloadable file.
Worked example
Officer Adeyemi asks AI to convert the vague brief into three possible research questions, each with scope, data needs, and rough effort. AI proposes: (1) 'What is the self-reported time saved by officers using AI tools for document drafting in three Federal MDAs?' (data: survey, moderate effort); (2) 'How does AI tool adoption correlate with processing time for FOI requests nationally?' (data: administrative records, high effort, likely unavailable); (3) 'What AI tools are currently used informally by civil servants and for what tasks?' (data: short survey, low effort). Adeyemi and the DG pick option 1 as most feasible, then ask AI to state explicitly what is out of scope: state-level civil servants, non-AI productivity tools, and any period before 2023.
The method
Generate multiple candidate research questions scored by feasibility (available data, effort) before picking one, then explicitly list exclusions (what will not be studied) in writing so scope creep has a documented boundary to be pointed back to later.
Worked practice and common mistakes
A feasible self-reported time-saving survey measures what respondents say, not causal productivity. Note self-report bias in the design. A scope change needs a decision on time, access and method. State population, period, source and outcome so the research question is answerable rather than an impressive title. Do not estimate unavailable data to preserve an otherwise unworkable question.
Review before using the result
A good research question is specific and answerable with data you can actually get. State exclusions explicitly in writing. Choose questions based on feasibility, not ambition.
Your workplace practice
Use the recommended free assistant with this lesson's fictional example. Convert this vague request into 3 research questions with scope, data needed and rough effort for each: 'AI and productivity in the Nigerian civil service.' Then state what is out of scope for the most feasible one. Keep a copy of the input and draft. Compare the result with the verification steps above, record any correction, and retain placeholders where evidence is missing. The expected result is a checked draft for human review, not an automatically approved decision. Course examples and practice responses are fictional; amounts, dates, rules and thresholds are not official guidance or supplier quotations.
Open ChatGPT (free tier) and paste this prompt there. Asks good clarifying questions to sharpen a vague brief.
Convert this vague request into 3 research questions with scope, data needed and rough effort for each: 'AI and productivity in the Nigerian civil service.' Then state what is out of scope for the most feasible one.
Why this tool: Asks good clarifying questions to sharpen a vague brief.
Hit a free limit? Switch to ChatGPT or Gemini and continue — the prompt works in any assistant.