13 Research methods, statistics, critical review and evidence-based practice13.7 Qualitative research
13.7.7 Appraising qualitative and mixed methods research
2 min read
Written by Resident Doctor Committee · Published
Coherence of a qualitative study
The question, methodological assumptions, sampling, collection and analysis should hang together, and findings should trace back to the data.
A study claiming grounded theory should show categories built up iteratively and theoretical sampling.
The CASP qualitative checklist has ten questions; "can't tell" marks a reporting gap, not an omitted method.
Sampling and reflexivity
A small purposive sample can answer a focused question, but its size and recruitment still need justifying.
Reflexivity asks who interviewed and how they knew participants; an independent interviewer eases some pressures but removes none entirely.
Analysis and rigour
Verbatim extracts alongside each theme keep participants' meaning intact and let the reader check that interpretations are grounded in the data.
Negative or divergent cases strengthen an analysis by testing its limits.
Rigour is judged as credibility, dependability and transferability, not statistical representativeness.
Technique | What it does |
|---|---|
Member checking (respondent validation) | Takes the researchers' interpretation back to participants for confirmation, establishing credibility |
Triangulation | Checks findings against an independent source, method or investigator |
Re-interviewing the same participants cannot provide triangulation, because the source is not independent.
Reporting guidance | Year | Items | Scope |
|---|---|---|---|
COREQ | 2007 | 32 | Interview and focus group studies |
SRQR | 2014 | 21 | Any qualitative research |
Both guide reporting; neither certifies quality.
Validity of instruments
Validity is the extent to which an instrument measures the construct it was designed to measure, as against reliability, the consistency of its scores.
Type | What shows it |
|---|---|
Face validity | The measure looks, on the surface, as if it assesses what it should; an informal judgement without statistical testing shows face validity only |
Criterion (concurrent) validity | A translated instrument agrees with an accepted reference measure of the same construct taken at the same time |
Mixed methods designs
Mixed methods research combines qualitative and quantitative approaches with a stated rationale and real integration; putting quotations beside trial results is not integration.
Design | Sequence | What the second component does |
|---|---|---|
Convergent | Concurrent | Views one phenomenon from two angles, joined at interpretation |
Sequential explanatory | Quantitative, then qualitative | Explains the quantitative result, such as interviews after a null trial |
Sequential exploratory | Qualitative, then quantitative | Develops an instrument or intervention that is then tested |
A joint display sets quantitative results, qualitative findings and the combined interpretation side by side.
Appraising the combination
MMAT 2018 asks two screening questions, then applies design-specific criteria; judgements should not be collapsed into one score.
A weak component limits the combined claims that rest on it but does not invalidate the stronger component's findings.
What an interviewee attributes to treatment is not a causal effect estimate.
GRAMMS covers rationale, design, integration, limitations and added insights, but following it does not make the mixing work.