Module 08 · 3 lessons

Research design

Free — no account needed to read

The shape of your question tells you which designs become appropriate — and why.

Why it matters

Choosing a design by habit rather than by fit is the most expensive error in research, because you usually discover it after collecting the data.

What it teaches

• Reading the shape of your own question
• Which designs become appropriate for which shapes
• What each design buys and what it costs
• Feasibility: time, access, money and ethics as real constraints
• Saying honestly what your design cannot deliver

When you reach for it

• You know your question and are choosing how to answer it
• You have access to one kind of data and want to know what it can support
• Somebody asks why you did it this way

The common misreading

That there is a hierarchy of designs with randomised trials at the top. A trial is the strongest tool for a causal question and the wrong tool for a descriptive or interpretive one.

What this module will not tell you

This module will not hand you a design. It shows which designs become appropriate for which shapes of question, and the choice depends on your access, ethics, time and field conventions — all of which you know and it does not.

Think first

You want to know whether a training programme improves performance. You cannot randomise who receives it. What can you still do, and what will you have to give up?

Write your question, then write what you would need to observe in the world for the answer to be clear. Not how you would get it — just what would have to be observable.

The lessons

• Read the shape of the question — Name the shape first. It rules out most designs immediately and tells you what the rest would each cost. (9 min)
• What each design buys, and what it costs — You are not choosing the strongest design; you are choosing which limitation you can live with and will declare. (9 min)
• Feasibility is part of the design, not an afterthought — Test the binding constraint first. A design that survives contact with access, time, money and ethics is a real design. (7 min)

Practise it: Walk your question to a design

Do the reasoning end to end, and keep the alternatives visible rather than collapsing to one answer.

1. Write your question and name its shape.
2. List every design that could speak to that shape.
3. For each, write what it buys and what it costs.
4. Cross out the ones your access, time, money or ethics rule out — and say which constraint did it.
5. From what remains, choose one, and write the sentence you would use to defend the choice.
6. Write the sentence that says what it will not be able to tell you.

What a considered answer tends to contain (not a mark scheme): A considered answer keeps more than one design alive until a named constraint removes it, and ends with both a defence and a declared limitation.

Use it on real work: Commit a design to a real project

Write the design and its reasoning into a project you are actually running.

1. Record the design in the project brief with the one-sentence defence.
2. Record the limitation in the same place, so it reaches your write-up later.
3. Name the binding constraint and the smallest test of it you can do this week.
4. Do that test and record what you found.

Reflect

Which design did you rule out for a good reason, and which did you rule out because it was unfamiliar?