From topic to investigation

How do you turn an overly broad topic into a precise, neutral question that can be investigated using documentary sources?

1250 € per session, excl. VAT

In short

This half-day session of 3 hours takes participants from a general topic to a workable question for documentary research. Participants discover how to define the scope, clarify terms and compare formulations using generative AI, then assess their question using a simple evaluation grid.

Why this training?

A topic such as « digital workshops » opens up too many avenues to guide research effectively. A clearly scoped question helps with choosing keywords, identifying useful information and excluding documents that address a different problem. AI can suggest alternative wording, but it can also introduce an assumed cause or broaden the scope unnecessarily. Learning to discuss its suggestions critically helps you retain control over your information needs, without confusing convincing wording with a question that can genuinely be investigated using documentary sources. Running case study: The fictitious organisation « Avelune Médiathèques » wants to understand the low attendance of adult beginners at its digital workshops. Using a fictitious training dossier containing schedules, workshop descriptions and anonymised feedback, participants turn this concern into a documentary research question, without assuming that the schedules explain the situation. Application at work: In the following week, revisit a documentary research need from your work, draft an initial question and then apply the evaluation grid. Ask a colleague to review the revised version and retain the justified changes before starting to search for sources.

Objectives of this training

  • Distinguish between a topic, a research question and a hypothesis.
  • Define a relevant population, context and period.
  • Draft an open question without presupposing its answer.
  • Compare three alternative formulations suggested by generative AI.
  • Assess a question and justify the corrections made.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Staff who need to prepare documentary research or a synthesis report.
  • Junior researchers and documentation staff wishing to clarify a request.

Prior skills

  • No prior knowledge of AI required.
  • Ability to describe a professional information need in a few sentences.

What you will work on

  1. Recognising a workable question

    Objective: Distinguish between the initial concern, topic, hypothesis and question. Differences between a request for a solution and a need for knowledge; descriptive, comparative and explanatory questions.

  2. Narrowing the scope without losing substance

    Objective: Define a scope compatible with the information that can be accessed. Population concerned, geographical area, period, phenomenon under study; definition of ambiguous terms and useful exclusions.

  3. Requesting alternative wording without delegating judgement

    Objective: Discover a simple interaction with generative AI. An instruction specifying context, scope and neutrality; risks of presuppositions, added concepts and fabricated references.

  4. Checking whether the question can be investigated using documentary sources

    Objective: Assess the quality of the scoping before starting research. Evaluation grid: clarity, scope, neutrality, defined terms and potential sources; distinction between a relevant question and an available answer.

What you take away

  • Scoping sheet for the documentary research need.
  • Comparison table of three formulations with comments.
  • Final question with definitions of its terms and exclusions.
  • Before/after evaluation grid and initial leads for sources.

Method and assessment

  • Short teaching inputs followed by immediate practice using the running case study.
  • Individual work followed by work in pairs, with participants reviewing each other’s outputs.
  • Guided group discussion: what works, what needs correcting and what remains a human decision.
  • Preparation: Arrange access to a generative AI assistant authorised by the organisation; a guided demonstration can replace individual hands-on use. Choose a fictitious or anonymised topic, with no personal data or confidential documents to be shared.

Observable criteria: The question addresses one clearly identifiable main subject. ; The population, context and period are made explicit where necessary. ; The wording does not present any assumed explanation as a fact. ; Important terms are defined and two relevant types of sources are identified. ; The before/after comparison shows corrections justified in the evaluation grid..

Frequently asked questions

Do you need to be familiar with academic research methods?

No. Scoping is introduced through an everyday documentary research need, with annotated examples.

Does the AI answer the question during the session?

No. It helps with rewording; searching for evidence and synthesising findings remain subsequent steps.

Can you work on a real request?

Yes, after anonymisation and only with authorised tools; the fictitious case study remains available.

Is a very precise question always preferable?

No. An overly narrow scope can make sources impossible to find; the evaluation grid helps identify this risk.

Last updated: