MAST20034 · Critical Thinking With Data
Qualitative Methods
Week 6 widens the toolkit beyond numbers: some questions are about why and how, not how many, and for those, qualitative methods are the right tool. The chapter sets out when qualitative is appropriate (exploring meaning, generating hypotheses, understanding process and context) and the four data sources — interviews, focus groups, observation, and document/artefact analysis — with what each is good and bad at. You learn coding as the core analytic act: inductive (bottom-up) coding that lets themes emerge from the data versus deductive (top-down) coding against a pre-set framework, feeding into thematic analysis. Crucially, you learn that qualitative rigour is judged by its own criteria — credibility, transferability, dependability and confirmability, with convergence (triangulation) as the key quality signal — not by sample size or p-values. The chapter closes on mixed methods and the honest qual-vs-quant trade-offs: depth and context versus breadth and generalisability. Exam prompts here ask you to pick the right approach and justify it.
What this chapter covers
- 016.1 When qualitative is the right tool (why / how, not how many)
- 026.2 The four sources of qualitative data — interviews, focus groups, observation, documents
- 036.3 Coding: inductive (bottom-up) vs deductive (top-down) and thematic analysis
- 046.4 Rigour (credibility / transferability), convergence and mixed methods
Choosing the right method and justifying it, mark by mark
- +1Match question to method: the question is about why — meaning, experience and process — which a survey count cannot reach, so a qualitative approach is appropriate.
- +1Pick a source + justify: use semi-structured interviews (or focus groups) with recent leavers/stayers — they elicit reasons, context and unanticipated factors in the participants' own words.
- +1Name the analysis: apply thematic analysis, coding transcripts inductively so reasons emerge from the data rather than being forced into a pre-set list.
- +1Add rigour: establish credibility via triangulation — e.g. member-checking themes with participants, or converging interview findings with exit-survey data.
Key terms
- Qualitative research
- Methods that study meaning, experience and process through non-numerical data (words, images, observations). Suited to ‘why’ and ‘how’ questions and hypothesis generation, not to measuring prevalence or testing effects.
- Inductive vs deductive coding
- Inductive (bottom-up) coding lets themes emerge from the data; deductive (top-down) coding applies a pre-existing framework. The choice shapes what you can find — inductive for discovery, deductive for testing a known frame.
- Thematic analysis
- Systematically coding qualitative data and grouping codes into themes that answer the research question. The workhorse analysis for interview and focus-group data.
- Credibility / transferability
- Qualitative analogues of internal and external validity: credibility = are the findings believable for these participants (triangulation, member-checking)?; transferability = could they apply to other contexts, given thick description?
- Triangulation (convergence)
- Combining multiple data sources, methods or analysts so that agreement strengthens confidence in a finding. Convergence across qual and quant evidence is a key quality signal in mixed methods.
Qualitative Methods FAQ
When should a study be qualitative rather than quantitative?
When the question is about meaning, experience, process or context — the ‘why’ and ‘how’ — or when you are exploring and generating hypotheses. Quantitative suits ‘how many’, ‘how much’ and testing pre-specified effects.
How is qualitative rigour judged if not by sample size?
By its own criteria — credibility, transferability, dependability and confirmability — supported by practices like triangulation, member-checking, an audit trail and thick description. A small, well-grounded qualitative study can be highly rigorous.
What do mixed methods add?
They let depth and breadth reinforce each other: qualitative work explains the ‘why’ behind a quantitative pattern, and convergence between the two strands raises overall confidence. The cost is greater complexity and design care.
Exam move
Lead with the question-type decision rule — ‘why/how/meaning’ → qualitative, ‘how many/how much’ → quantitative — because most exam prompts here ask you to choose and justify a method. Put the four data sources (one strength + one weakness each) and the credibility/transferability rigour criteria on your notes sheet. Practise distinguishing inductive from deductive coding in a sentence, and always pair a recommended method with a named rigour step (triangulation, member-checking). Remember the trade-off line: qualitative buys depth and context at the cost of breadth and generalisability.
