FMHU5003 Chap.2 Designing Qualitative Research
Designing Qualitative Research
This chapter builds qualitative design as an alignment problem. A defensible study connects the phenomenon of interest, the relevant macro, meso or micro level, a focused research question, an information-rich sample and a transparent stopping claim.
Students compare purposeful, snowball, maximum-variation and convenience strategies, then practise judging adequacy through relevance, variation, analytic depth and the scope of the conclusion rather than sample size alone. The chapter also separates a study-level research question from the questions asked of participants.
It shows how samples can include people, texts, images, objects and time in places, provided the selection helps illuminate the problem. Flexibility is treated as documented iteration rather than improvisation: analysis may change later recruitment, refine the question or expose a missing social level, and the researcher should make that development traceable.
What this chapter covers
- 01
Phenomena suited to qualitative inquiry
- 02
Macro, meso and micro social levels
- 03
Focused and researchable qualitative questions
- 04
People, texts, objects and places as samples
- 05
Purposeful, snowball and maximum-variation selection
- 06
Thematic and theoretical saturation claims
Worked example · free
Redesign an access question
- +1Replace frequency with a process or experience that needs explanation.
- +1Identify the relevant group and setting without narrowing the answer in advance.
- +1Choose a purposive sampling logic that can illuminate meaningful contrast.
- +1Bound the conclusion to contextual explanation rather than prevalence.
Key terms
- Purposive sampling
- Selection based on relevance to the research question rather than statistical representativeness.
- Maximum variation
- Purposeful inclusion of meaningful contrasts so similarities and differences can be examined.
- Saturation
- A bounded claim that further sampling is no longer developing the specified analytic account.
Designing Qualitative Research FAQ
What makes a qualitative question strong?
A strong question is worthwhile, researchable, focused and open to depth. It asks how a process works or how people interpret experience, identifies the relevant social world and avoids installing a preferred explanation. It also supports a feasible sample and method. The wording should invite discovery while remaining precise enough to guide selection and analysis.
Does qualitative sampling aim to represent a population?
Usually it aims for relevance and insight rather than statistical representativeness. Participants, documents or settings are selected because they can illuminate the problem. The researcher should explain the selection logic, meaningful variation and exclusions, then keep the conclusion within those boundaries. Transfer depends on contextual understanding, not a claim that the sample is average.
Is saturation a fixed sample size?
No. The course material places sample size in relation to the progress of analysis. A useful saturation claim names what became sufficiently developed, for which sample and analytic purpose. Simply announcing saturation or relying on a universal number provides little methodological evidence. Different subgroups or theoretical categories may remain thin even when one pattern is well developed.
How do macro, meso and micro levels affect design?
They identify where the relevant social process is located. Macro questions concern wider systems or policy, meso questions concern organisations and groups, and micro questions concern individual experience or interaction. Some studies need connections across levels. The sampling and data-generation plan must include evidence capable of addressing each level named in the question.
Exam move
Draw the design chain from phenomenon through social level, question, sample, method and claim. For every practice case, test the chain in both directions. If the conclusion does not fit the sample, narrow it. If the sample cannot illuminate the question, change the selection logic. Compare sampling strategies by the relationships they make visible and the exclusions they create.
When reading a paper, underline every sentence that supports its saturation claim and note what the authors actually show. Build a comparison table for purposeful, snowball, maximum-variation and convenience sampling with columns for analytic purpose, likely strength, likely exclusion and ethical pressure. Take one broad topic and write separate macro, meso and micro questions, then design a relevant sample for each.
Practise distinguishing the research question from an interview question by explaining how several conversational prompts contribute evidence toward one study-level aim. During appraisal revision, replace the phrase small sample with a precise judgement about relevance, variation, analytic depth or the scope of the claim.
End each design exercise by writing a two-sentence audit trail: what evidence caused the design to change and why the change improved alignment. Add one negative case to every sampling exercise and decide whether it calls for a revised explanation, a different participant group or a narrower conclusion.
Explain aloud why information power depends on specificity, dialogue and analytic purpose rather than a magic participant count. Finally, compare the recruitment route with the population named in the conclusion, because an apparently varied sample may still share one access pathway and exclude people whose experiences would challenge the emerging account.
