Strata Academy

How to Appraise Qualitative Research: Complete Guide for Students

Paradigms, interviews, focus groups, thematic analysis, trustworthiness, CASP, JBI, COREQ, and qualitative synthesis

1. What qualitative research is (and is not)

Qualitative research systematically collects and interprets non-numerical data – interview transcripts, field notes, documents, images – to understand experiences, cultures, and processes. It answers how and why questions: how patients navigate a new diagnosis, why staff resist a pathway change, what stigma feels like in help-seeking.

It is not a 'weaker' version of quantitative research. Poor qualitative work exists; so does poor RCT reporting. The error is applying the wrong appraisal framework – judging an ethnography by power calculation standards, or an interview study by ROB 2 domains.

Medical students encounter qualitative papers in SSC portfolios, health services dissertations, implementation science, and mixed-methods theses. NHS organisations increasingly value patient experience data alongside trial evidence.

2. Paradigms: positivism, interpretivism, and pragmatism

Research paradigms shape what counts as evidence. Post-positivism underpins hypothesis testing and probabilistic inference in trials. Interpretivism holds that meaning is constructed socially – researchers interpret participant accounts rather than measuring a single objective truth.

Pragmatism, common in applied health research, chooses methods by the question: a process evaluation may combine surveys with interviews. Critical paradigms foreground power, equity, and structural racism – increasingly relevant in public health qualitative work.

Appraisal should check alignment: if authors claim grounded theory but present only descriptive themes without iterative sampling, the stated methodology and actual methods may not match.

• Interpretivism – depth, context, participant voice (most health qualitative primary studies)
• Pragmatism – mixed methods, programme evaluation
• Critical theory – structural determinants, participatory methods
• Reflexivity – authors state their position and influence on data collection

3. Framing questions with PEO

PICO suits comparative effectiveness. Qualitative studies often use PEO: Population (who), Exposure or phenomenon (what experience is studied), Outcome of interest (what understanding is sought). Example: In adults with chronic pain attending community physiotherapy (P), how is trust in the therapist constructed (E), to improve person-centred commissioning (O)?

Review questions for qualitative syntheses follow similar logic but search for qualitative evidence on a phenomenon – registered on PROSPERO with qualitative review methods specified.

4. Sampling strategies and saturation

Random sampling is rare. Purposive sampling selects information-rich cases. Maximum variation sampling spans diverse settings. Theoretical sampling (grounded theory) adds participants to refine emerging categories. Snowball sampling uses networks – useful but can limit diversity.

Saturation means new data add little thematic novelty. Authors should explain how they judged saturation (code redundancy, no new codes in final interviews), not only cite a number. Single-site homogeneous samples weaken transferability even if saturation is claimed.

5. Semi-structured and in-depth interviews

One-to-one interviews dominate health qualitative research. Topic guides list domains with open questions; skilled interviewers probe without leading. Appraise whether guides were piloted, interviews audio-recorded and transcribed verbatim, and language of interview matches participants.

Telephone and video interviews expanded access post-pandemic but alter rapport and non-verbal data. Short interviews (<20 minutes) may not reach depth for phenomenological claims.

• Leading questions – 'You found the clinic helpful, didn't you?'
• Recording and transcription quality – omissions obscure appraisal
• Interview fatigue and sensitive topics – ethics and distress protocols
• Researcher characteristics – gender, clinical role, power imbalance

6. Focus groups and group methods

Focus groups exploit group interaction – norms, disagreement, and shared language. They suit exploring attitudes to public health messaging or service prototypes. They are poor for deeply personal trauma where privacy dominates.

Group composition (mixing junior and senior staff, patients and managers) shapes what can be said. Facilitator neutrality and managing dominant voices are methodological skills, not incidental details.

7. Observation, ethnography, and fieldwork

Ethnography involves prolonged field immersion – wards, clinics, communities. Participant observation ranges from overt note-taking to more immersive roles. Time in field, gatekeeper access, and how observations were recorded and analysed must be reported.

Structured non-participant observation may count activities or behaviours with qualitative interpretation – distinct from time-motion quantitative studies.

8. Documents, diaries, and visual data

Policy texts, social media (with ethics approval), patient diaries, and photo-elicitation extend beyond talk. Document analysis asks who produced the text, for what audience, and what was omitted. Triangulation combines interviews with observation or documents to strengthen credibility.

9. Thematic analysis (Braun & Clarke)

Reflexive thematic analysis is widely taught in UK medical schools. Six phases: familiarisation, coding, generating themes, reviewing themes, defining/naming, writing up. Themes are researcher interpretations of patterned meaning – supported by quotations, not bullet summaries alone.

Descriptive themes restate data ('barriers to access'); interpretive themes explain mechanisms ('moral identity threatened by dependence'). Appraisal checks whether authors distinguish description from interpretation.

1. Familiarisation with transcripts
2. Systematic coding across dataset
3. Collating codes into candidate themes
4. Reviewing themes against coded extracts
5. Defining and naming themes
6. Writing results with vivid examples

10. Grounded theory, IPA, phenomenology, framework analysis

Grounded theory aims to generate theory from data through constant comparison and theoretical sampling – Glaserian and Straussian traditions differ. Interpretative Phenomenological Analysis (IPA) uses small homogeneous samples to interpret lived experience through a double hermeneutic. Phenomenology seeks essence of experience (descriptive or interpretive). Framework analysis uses matrix charts common in policy research with a priori and inductive codes.

Mislabelling is common – 'grounded theory' used for any inductive coding. Read methods for iterative sampling and constant comparison before accepting the label.

11. Trustworthiness (Lincoln & Guba)

Credibility – are findings believable? Prolonged engagement, triangulation, member checking (used cautiously), peer debriefing. Transferability – thick description lets readers judge if insights apply elsewhere – not statistical generalisation. Dependability – audit trail, code-recode, team analysis. Confirmability – reflexivity; findings traceable to data not solely researcher preference.

• Do not map blindly to 'internal validity' – terminology differs by design
• Member checking can be inappropriate for power-sensitive topics
• Inter-coder agreement helps but is not mandatory in reflexive TA

12. CASP and JBI qualitative critical appraisal

CASP qualitative checklist structures journal club into: (1) Is the study valid? (2) What are the results? (3) Are results useful locally? Work through all sections – students often stop after section 1.

JBI qualitative critical appraisal is used when synthesising qualitative evidence in reviews. Items address methodological congruity, participant voice, analysis logic, and conclusion fit. Dissertation examiners may specify JBI for qualitative systematic reviews.

13. COREQ and SRQR reporting

COREQ (Consolidated Criteria for Reporting Qualitative Research) covers 32 items: research team reflexivity, study design, and analysis/reporting. Check interviewer training, setting, relationship established before study, saturation, and use of software.

Incomplete COREQ reporting mirrors CONSORT gaps – you cannot appraise what is not described. Flag missing ethics approval, absent quotations, or undefined analysis approach as major limitations.

14. Qualitative evidence synthesis

Qualitative systematic reviews aggregate findings across primary studies using meta-ethnography, thematic synthesis, critical interpretive synthesis, or JBI convergent methods. ENTREQ and eMERGe guide reporting of syntheses. Search strategies must locate qualitative literature – often supplementary to Medline, including CINAHL, PsycINFO, and grey literature.

Mixed-methods reviews combine quantitative and qualitative streams – require explicit integration methods and may use GRADE-CERQual for qualitative certainty.

15. Step-by-step student appraisal workflow

1. Confirm design is qualitative – not a mislabelled survey
2. State PEO in one sentence
3. Assess sampling, ethics, and data collection fit
4. Name analysis approach; check quotes support themes
5. Complete CASP sections 1–3
6. Note applicability to your NHS context
7. State one practice implication and one research gap

16. Strata Academy qualitative module

The paid Clinical Research Methods course includes a dedicated qualitative research module (three interactive lessons): paradigms and design, methods and analysis, and appraisal with trustworthiness checklists. Free guides remain open – the module adds scenarios, quizzes, and progress tracking toward your certificate.

Interactive version (quizzes, walkthroughs) loads when JavaScript is enabled.