ENVM7003 Chap.2 From Broad Topic to a Researchable Question
From Broad Topic to a Researchable Question
A question is scoped when you can look at it and say which method it forces. That is the test this chapter trains. Broad questions feel comfortable precisely because they rule nothing out, and a design built on one ends up with instruments that cannot answer what is printed above them.
The course supplies a narrowing sequence: one tension, a named population and place, a manageable relationship, no causal claim the design cannot support, and a project finishable within the time, access, data and analysis skill the course actually gives you.
Six criteria then test the draft.
It must be clear and focused, neither too broad nor too narrow, not answerable yes or no, not too difficult to answer thoroughly in the time, researchable with materials you can obtain, and analytical rather than descriptive. Over-correction is a real failure: a question cut down to one street and one week leaves nothing to discuss.
After the question comes the chain the course teaches explicitly: break it into three to five aims, decide what data each aim needs, and choose methods last. The chapter finishes on feasibility, where the project clinic questions about scope, exclusions, dependencies and back-up plans do the real work.
What this chapter covers
- 01
2.1 Narrowing until a method is forced rather than chosen
- 02
2.2 Six criteria a draft question has to survive
- 03
2.3 Why too narrow fails as badly as too broad
- 04
2.4 Breaking one question into three to five aims
- 05
2.5 Deciding the data before deciding the method
- 06
2.6 Scope, exclusions, dependencies and a fallback that still answers
One question, four aims, three methods, chosen in that order
- +2Aim one: establish current use. Frequency and activity counts across a resident sample. Closed and category items answer this and nothing else would be cheaper.
- +3Aim two: establish what is valued. Rated or ranked value statements plus one open item in the respondent's own words, because the open item is where the report's quotations come from.
- +3Aim three: surface the trade-off. Reasoning about a concrete redesign choice, argued aloud. This is the only aim that needs people responding to each other, which is the justification for adding a focus group.
- +2Aim four: compare with the stated intent. The published design rationale and consultation record, read as documents rather than collected as new data.
- +2Only now name the methods. A short instrument, one focus group and a document reading. Each is attached to an aim, so the justification writes itself.
Key terms
- Researchable question
- A question you can name the answering data for. If you cannot say what you would need to see, the question is not yet researchable however well it reads.
- Scope statement
- A short statement of who or what is studied, where, with what focus, and what is deliberately excluded. The exclusion clause is the part that makes the rest real.
- Dependency
- A relationship where one piece of work cannot start or finish until another does. Approval before collection is the dependency that most often sits on a student project's critical path.
- Fallback design
- A planned alternative that answers the same question from a different data source. Deciding it in advance produces a stronger design than inventing it under pressure.
- Causal claim
- An assertion that one factor produced a change in another. Most feasible student designs cannot support one, and the topic guide warns against making it anyway.
From Broad Topic to a Researchable Question FAQ
How do I know when my research question is narrow enough?
When it forces a method. Read the draft and ask which two or three methods could answer it; if the honest reply is 'almost anything', it is still too broad. The reverse check matters too: if only one very small dataset could answer it, you have gone past scoped into trivial and the discussion section will have nothing to work with.
How many aims should a project have?
The course guidance suggests anywhere between three and five. The number matters less than the structure: each aim should be answerable by identifiable data, and no two aims should need the same data, because that is a sign you have written one aim twice.
Do I need approval before collecting data?
Yes. The topic guide asks you to obtain approval from your tutor or the course coordinator before you start collecting data or lock down an analysis of an existing dataset. Treat that conversation as free design review and bring the question, the population, the instrument and the fallback.
What goes in the project clinic worksheet?
The research question; the scope including what is excluded; the methods, data, collection approach, ethical considerations and analysis plan; dependencies and back-up plans; and a week-by-week timeline through to the presentation weeks. It is not submitted, which is exactly why weak groups skip the only design review they get for free.
What is the most common reason a project runs out of time?
Recruitment. A design that depends on strangers replying to a student request has one point of failure and no natural warning. The repair is a fallback that changes the data source rather than the question: a secondary dataset, a document set, a consultation record, or an observational method on the same site.
Assessment move
Write your question, then write the five sentences that surround it: the problem and who it affects, the question, the aims, the methods and why those, and what the project will not do. Get every group member to agree on those five in one sitting. If agreement does not come quickly the disagreement is real and will otherwise appear as contradictory drafting late in the semester.
Those same five sentences are also the skeleton of the three-minute pitch, so the work is never wasted.
Organisational Behaviour · Identifying and Assisting Students at Risk · Digital Marketing and Social Media
