Quality, Trustworthiness and Critical Appraisal

Introduction

Qualitative health research can illuminate experiences, meanings, relationships and organisational processes that are difficult to capture through numerical measurement alone. Its value, however, depends on more than interesting quotations or a compelling narrative. Readers must be able to judge whether the interpretation is credible, whether the research process was sufficiently systematic, whether conclusions are grounded in the data, and whether insights may be useful beyond the original setting. These concerns are addressed through quality, trustworthiness and critical appraisal.

In quantitative research, quality is often discussed through validity, reliability, precision and generalisability. Qualitative research operates with different assumptions about reality, knowledge and the role of the researcher. Consequently, quality cannot be reduced to applying quantitative criteria to non-quantitative data. A trustworthy qualitative study is transparent about its philosophical position, context, participants, relationships, analytic decisions and limitations. It demonstrates a reasoned connection between the research question, design, data generation, analysis and claims.

This chapter examines the major criteria of trustworthiness—credibility, dependability, confirmability, transferability and authenticity—and the practices used to support them. It also explains triangulation, member reflection, peer debriefing, audit trails and thick description. Finally, it introduces critical appraisal and reporting standards, especially CASP, COREQ and SRQR, and considers their appropriate use in health-care management research.

Learning objectives

By the end of this chapter, the reader should be able to:

1. Explain the meaning and philosophical basis of qualitative research quality.
2. Distinguish credibility, dependability, confirmability, transferability and authenticity.
3. Select and apply appropriate strategies for strengthening trustworthiness.
4. Critically appraise a qualitative health research article using CASP and related tools.
5. Use COREQ and SRQR to evaluate or improve reporting completeness.
6. Identify common threats to quality, including superficial triangulation, tokenistic member checking and reflexivity failure.
7. Judge whether findings are sufficiently supported for clinical, organisational or policy decisions.

1. Quality in qualitative inquiry

1.1 Quality is a chain of coherence

Qualitative quality is best understood as a chain of coherence rather than a checklist. The following elements should fit together:

• the research question and its assumptions;
• the methodological tradition, such as phenomenology, ethnography, grounded theory or case study;
• participant selection and access;
• the method of data generation;
• the researcher–participant relationship;
• data management and analysis;
• interpretation and reflexivity; and
• the form and scope of the conclusions.

For example, a study claiming to develop a grounded theory should show theoretically purposeful sampling, iterative analysis and development of concepts, not merely use interviews and label the result “grounded theory.” Similarly, an ethnographic study should provide evidence of sustained engagement with a social setting and analysis of culture or practice, rather than a small collection of interviews alone.

1.2 Paradigm and quality criteria

Qualitative researchers do not all share the same view of reality. A post-positivist study may emphasise systematic procedures and evidential warrant. A constructivist study may treat findings as co-constructed between participants and researchers. A critical study may ask whose interests are served by existing arrangements and whether research challenges inequity. These differences affect how quality is demonstrated.

The criteria developed by Lincoln and Guba—credibility, dependability, confirmability and transferability—remain influential because they provide a language suited to naturalistic inquiry. Other scholars have emphasised authenticity, resonance, ethical integrity, criticality, coherence and usefulness. No framework should be applied mechanically. The criteria must be interpreted in relation to the study’s paradigm and purpose.

2. The core dimensions of trustworthiness

2.1 Credibility

Credibility concerns the plausibility and believability of the findings from the perspective of participants, relevant professionals and informed readers. It asks: “Does the interpretation adequately represent the phenomenon studied?” Credibility is not achieved simply by collecting a large number of interviews. It depends on the quality of engagement, the appropriateness of sampling, the depth of data and the adequacy of analysis.

Strategies that may support credibility include:

• prolonged engagement with the setting or topic;
• persistent observation of issues relevant to the research question;
• purposeful or theoretical sampling;
• triangulation of sources, methods, researchers or theories;
• member reflection;
• peer debriefing;
• searching for negative or discrepant cases;
• iterative data collection and analysis; and
• presenting sufficient evidence for interpretations.

Credibility requires attention to disconfirming evidence. If a study reports that all nurses welcomed a new digital system but does not discuss dissenting accounts, the reader cannot assess whether consensus was genuine or manufactured by selective reporting.

2.2 Dependability

Dependability concerns the stability, consistency and traceability of the research process. It does not mean that repeating a qualitative study must produce identical findings. Social contexts change, and researchers are part of the research instrument. Instead, dependability asks whether the process is logical, documented and responsive to changes in the field.

A dependable study records:

• why the design was selected;
• how participants were recruited;
• how interview or observation procedures evolved;
• how coding and category development occurred;
• why analytic decisions were made; and
• how deviations from the original protocol were handled.

An audit trail is central. It may contain versions of interview guides, recruitment materials, field notes, coding frameworks, analytic memos, meeting records, reflexive notes and documented decisions. A supervisor or auditor should be able to follow the pathway from raw material to interpretation, while respecting confidentiality.

2.3 Confirmability

Confirmability addresses the extent to which findings are grounded in participants’ accounts and the data rather than being inventions of the researcher. Complete objectivity is not assumed. Instead, the researcher demonstrates how interpretations were reached and how personal assumptions, professional interests and power relationships were considered.

Confirmability is supported by:

• reflexive journals;
• explicit positionality statements;
• an audit trail;
• examination of negative cases;
• team discussion of competing interpretations;
• retaining links between codes, categories and excerpts; and
• distinguishing participant accounts from researcher interpretation.

A quotation does not automatically prove confirmability. A quotation may be selected because it is vivid but atypical. Researchers should explain whether an excerpt illustrates a dominant pattern, a minority position or a contradiction.

2.4 Transferability

Transferability concerns the extent to which readers can judge whether findings may be relevant in another context. The researcher does not usually claim statistical generalisation from a qualitative sample. Rather, the researcher provides enough contextual detail for readers to make an informed comparison.

Transferability depends on thick description, including:

• the health-care setting and service model;
• geographical and socioeconomic context;
• organisational history and culture;
• participant characteristics relevant to the question;
• professional roles and power relationships;
• timing and duration of fieldwork; and
• the conditions under which the findings were produced.

Thick description is not simply lengthy description. It is analytically relevant detail that enables interpretation. For instance, “staff resisted the intervention” is less useful than an account of staffing levels, implementation history, decision-making structures, prior experiences of change and the meanings staff attached to resistance.

2.5 Authenticity

Authenticity refers to whether the research represents a fair range of perspectives and contributes to understanding, self-reflection or action. Authenticity is particularly important in studies involving marginalised groups, contested organisational decisions or unequal power.

Dimensions commonly associated with authenticity include:

• fairness: representing different stakeholder perspectives;
• ontological authenticity: helping participants understand their own situation;
• educative authenticity: increasing understanding of others’ perspectives;
• catalytic authenticity: stimulating action or change; and
• tactical authenticity: supporting participants’ capacity to act.

Authenticity does not mean that every viewpoint receives identical space regardless of relevance. It requires an ethically defensible account of whose voices were included, whose were absent and how the researcher dealt with unequal ability to speak.

3. Practical strategies for establishing trustworthiness

3.1 Triangulation

Triangulation involves examining a question through more than one perspective or source. It can take several forms:

Triangulation is not a vote-counting exercise. If three sources agree, that does not automatically make a finding true; they may all reflect the same institutional narrative. Conversely, disagreement is not a failure. Differences may reveal role-based interests, temporal change or unequal power. A strong study explains what convergence and divergence mean.

3.2 Member reflection

Member reflection—sometimes called member checking—invites participants to consider summaries, interpretations or emerging findings. It may occur during data collection, after preliminary analysis or during dissemination. It can identify factual errors, clarify meaning and create dialogue about interpretations.

However, participants are not automatically the final arbiters of analysis. A participant may reject an interpretation because it is uncomfortable, because organisational consequences are feared or because the interpretation concerns patterns not visible from an individual position. Member reflection should therefore be treated as an additional source of insight, not as a simple validity vote. Researchers should document who was invited, who responded, what changed and why.

3.3 Peer debriefing

Peer debriefing involves discussing methodological and interpretive decisions with a knowledgeable colleague who is not directly responsible for all aspects of the study. The peer may question assumptions, identify gaps, challenge premature closure and ask whether conclusions exceed the evidence.

Effective debriefing is more than asking a colleague to proofread the manuscript. It should address questions such as:

• What alternative explanation fits these data?
• Which participants or cases are missing?
• Is the concept supported across the dataset?
• How might the researcher’s role have shaped the interaction?
• Does the conclusion claim more than the design can support?

3.4 Audit trails

An audit trail creates a transparent record of methodological and analytic development. It is particularly valuable in team-based studies and research commissioned by health-care organisations. A practical audit trail may include four linked layers:

1. Raw data: recordings, transcripts, observations and documents.
2. Data reduction: coding, summaries, matrices and case profiles.
3. Interpretive work: memos, diagrams, category definitions and rival explanations.
4. Decisions: changes in sampling, analytic disagreements, ethical responses and final claims.

The audit trail must protect privacy. Identifiers should be removed or encrypted, access should be controlled, and retention should comply with institutional policy and applicable data-protection law.

3.5 Thick description

Thick description connects events to meanings and context. It allows readers to assess transferability and understand why a finding matters. In reporting an interview study about missed appointments, a thin account might state that transport difficulties were mentioned. A thick account would describe the service location, public transport availability, appointment flexibility, work and caring responsibilities, the meaning of “non-attendance” to staff and patients, and how these factors interacted.

Thick description should not expose participants to deductive disclosure. In a small specialist service, a combination of job title, age, rare diagnosis and quotation may identify a person. Researchers must balance contextual richness with confidentiality.

3.6 Reflexivity

Reflexivity is the systematic examination of how the researcher’s identity, assumptions, relationships and decisions shape the research. It includes personal reflexivity—how biography and values matter—and epistemological reflexivity—how theories of knowledge shape what is treated as evidence.

A reflexive account should be specific. “The researcher was aware of bias” is weak. A stronger account explains that the interviewer was a former manager in the participating service, how this affected recruitment and disclosure, what safeguards were used, and how interpretations were challenged by colleagues or participants.

4. From data to claims: judging evidential support

A useful appraisal asks whether the study shows a clear progression from data to claim. Consider the following sequence:

Research question → sampling → data generation → coding → categories/themes → interpretation → conclusion.

At each stage, the reader should ask whether the next step is warranted. For example:

• Does the sample include people with relevant experience?
• Were data sufficiently rich to answer the question?
• Are themes more than interview topics or headings?
• Are themes internally coherent and distinct from one another?
• Is contradictory evidence addressed?
• Are quotations contextualised rather than presented as isolated proof?
• Do conclusions remain within the scope of the sample and design?

Qualitative analysis is not merely a process of finding topics. A topic such as “communication” may be descriptive. A theme such as “communication as a negotiated form of safety work” offers an interpretive explanation, provided it is supported by the data and conceptually developed.

5. Critical appraisal of qualitative studies

Critical appraisal is a disciplined assessment of a study’s quality, relevance and usefulness. It is not an attempt to assign a universal numerical score. A low score on one reporting item may reflect poor reporting rather than poor research, while a well-reported study may still have a weak design.

5.1 CASP qualitative checklist

The Critical Appraisal Skills Programme (CASP) qualitative checklist is widely used in education, evidence synthesis and health-care decision-making. Its questions address:

• clarity of the research aims;
• appropriateness of qualitative methodology;
• suitability of the research design;
• appropriateness of recruitment strategy;
• adequacy of data collection;
• consideration of the researcher–participant relationship;
• ethical issues;
• rigour of data analysis;
• clarity of findings; and
• value of the research.

CASP is useful because it is accessible and encourages an overall judgement. It should not be used as a mechanical tick-box or as a substitute for understanding the methodology. For example, “Was the data analysis sufficiently rigorous?” requires examination of coding, theme development, negative cases and the link between evidence and interpretation.

5.2 COREQ

The Consolidated Criteria for Reporting Qualitative Research (COREQ) is a 32-item checklist designed primarily for interviews and focus groups. It is organised into three domains:

1. Research team and reflexivity: interviewer characteristics, credentials, occupation, experience and relationship with participants.
2. Study design: theoretical orientation, participant selection, setting, sample size, non-participation and data collection procedures.
3. Analysis and findings: data analysis, derivation of themes, participant checking, quotations and clarity of reporting.

COREQ is a reporting guideline, not a quality score. Failure to report an item does not necessarily mean that the procedure was not undertaken. It means that readers cannot readily judge it.

5.3 SRQR

The Standards for Reporting Qualitative Research (SRQR) provide a broader reporting framework applicable across qualitative approaches. They address title and abstract, problem formulation, purpose, qualitative approach, researcher characteristics and reflexivity, context, sampling, ethics, data collection, analysis, results, discussion, limitations and other information.

SRQR is especially useful for multidisciplinary health research because it accommodates interviews, observations, document studies and other designs. It encourages authors to explain methodological orientation rather than presenting methods as a list of techniques.

5.4 Other relevant standards

Depending on the design, other standards may be appropriate:

• ENTREQ: reporting qualitative evidence syntheses;
• RATS: qualitative research review criteria, including relevance, appropriateness, transparency and soundness;
• COREQ: interviews and focus groups;
• SQUIRE: quality-improvement studies, including qualitative components;
• GRIPP2: reporting patient and public involvement; and
• design-specific guidance from professional and methodological bodies.

Reporting standards should be selected according to the study, used early in protocol and manuscript development, and supplemented by methodological judgement.

6. Appraisal workflow for health-care managers and researchers

A practical appraisal can proceed in five stages.

Stage 1: Establish relevance

Identify the decision the evidence may inform. Is the study about patient experience, workforce retention, implementation, safety culture or service access? A methodologically strong study may still be irrelevant to a particular managerial decision.

Stage 2: Identify the methodological fit

Ask whether the research question requires qualitative inquiry and whether the selected tradition fits the question. Questions about meaning, experience, interaction, culture and process commonly justify qualitative methods.

Stage 3: Examine trustworthiness

Assess credibility, dependability, confirmability, transferability and authenticity. Look for evidence rather than labels. A paper that states “data saturation was reached” should explain how adequacy was judged and what role sampling and analytic sufficiency played.

Stage 4: Examine ethics and power

Consider consent, confidentiality, participant burden, vulnerability, conflicts of interest and the consequences of disclosure. In organisational research, employees may feel that participation is compulsory or that managers can identify criticism. Recruitment through line management requires particular caution.

Stage 5: Judge use and limits

Decide how the findings can responsibly be used. Qualitative evidence may explain why a policy is resisted, identify implementation mechanisms or reveal unintended consequences. It should not automatically be used to estimate prevalence or claim that all service users share one experience.

7. Illustrative case: evaluating a study of electronic rostering

The following is a fictional illustrative case.

Background

A regional hospital introduced an electronic rostering system intended to improve staffing visibility and reduce last-minute shift changes. Six months later, managers commissioned a qualitative study to understand why some wards used the system consistently while others continued to rely on informal messaging.

Problem

The commissioned report claimed that “staff resistance and poor digital literacy” explained limited adoption. It was based on eight interviews with ward managers and a short online staff survey containing two open-ended questions.

Analysis

The study had a relevant qualitative aim, but several quality concerns emerged. The sample excluded health-care assistants, nurses who worked night shifts and staff who had left the organisation. The commissioning department recruited participants, creating a risk that critical views were suppressed. The report did not describe the interviewer’s relationship to the hospital or whether reflexivity was considered.

Data analysis was described as “themes were identified,” with no explanation of coding, theme development or discrepant cases. The report offered only two quotations, both from managers, and did not explain how the open-ended survey responses were integrated with interviews. No audit trail or member reflection was reported.

A more credible interpretation might have examined whether the technology fit staffing practices. Subsequent observations showed that the system required approval from a central office that was unavailable after normal hours. Staff used messaging applications not because of low digital literacy but because they provided rapid feedback during emergencies. The implementation problem therefore involved workflow, accountability and system design as well as user capability.

Improved approach

A stronger study would use maximum-variation or purposive sampling across wards, roles, shifts and levels of adoption. It could combine interviews, observations of rostering work and document analysis of escalation procedures. Researchers would maintain a reflexive log, compare accounts across roles, discuss interpretations with peers and invite participants to reflect on preliminary findings. A transparent audit trail would document changes in sampling and analytic decisions. Reporting would use COREQ or SRQR as appropriate.

Outcome

The revised findings led to changes in after-hours approval, system configuration and training. Training was retained, but it was no longer treated as the sole solution. The hospital also established a user group including night-shift staff and health-care assistants.

Learning points

1. A plausible managerial explanation is not necessarily a well-supported qualitative finding.
2. Sampling must include perspectives relevant to the process under investigation.
3. Technology adoption is often shaped by workflow and power, not only individual competence.
4. Reporting standards can expose missing information, but they cannot replace good inquiry.
5. Findings are more useful when they explain mechanisms and conditions rather than merely naming barriers.

8. Common limitations and misuse

Checklist compliance without methodological integrity

A manuscript may tick every reporting item while using an unsuitable design or superficial analysis. Reporting quality and research quality are related but not identical.

Tokenistic member checking

Sending a final report to participants and recording “no comments received” does not demonstrate meaningful member reflection. Participants may lack time, authority or safety to respond.

Misunderstanding saturation

Saturation is not a universal numerical rule. It may refer to no new codes, sufficient depth, theoretical saturation or informational adequacy. Researchers should define the sense used and explain how it guided sampling.

Overreliance on inter-coder agreement

Agreement between coders can support consistency, especially in large teams, but qualitative quality does not depend solely on calculating coder agreement. Discussion, interpretive reflexivity and conceptual development may be more appropriate than forcing consensus.

Confusing participant validation with truth

Participants provide important perspectives but may not see institutional patterns, may revise accounts or may disagree with interpretations for legitimate reasons. Their responses should be analysed, not treated as a simple pass–fail test.

Excessive claims

A study of one hospital cannot automatically establish that a finding applies to all hospitals. Researchers should distinguish contextual transferability from population generalisation.

Ethical risks of thick description

Detailed context can increase identification risk. Researchers should use composite accounts cautiously, explain any alteration of identifying details and avoid quotations that reveal vulnerable participants.

9. Governance, technology and contemporary developments

Digital recording, transcription software and artificial intelligence can improve organisation and accessibility, but they introduce risks. Automated transcription may misrecognise accents, clinical terminology, pauses or emotionally significant speech. AI-assisted coding may reproduce hidden assumptions, flatten ambiguity or expose confidential data to unauthorised systems. Researchers remain responsible for checking transcripts, documenting software use, protecting data and retaining interpretive accountability.

Contemporary qualitative quality assessment increasingly recognises co-production, patient and public involvement, intersectionality and decolonising approaches. These perspectives ask whether participants influence the research agenda, whether standard categories obscure difference, and whether academic conventions privilege researcher authority. Quality therefore includes not only technical adequacy but also ethical and political responsiveness.

In implementation and health-care management research, qualitative findings are often integrated with quantitative indicators. Integration should be explicit. Qualitative evidence may explain mechanisms behind a numerical trend, identify unintended effects or reveal why an intervention works in one context but not another. It should not be treated merely as illustrative quotation added after statistical analysis.

10. Best-practice principles

A high-quality qualitative health study generally follows these principles:

1. Align the question, paradigm, methodology and methods.
2. Make sampling purposeful and explain inclusion, exclusion and non-participation.
3. Describe the researcher’s position and relationship to the field.
4. Collect data with sufficient depth and responsiveness.
5. Analyse iteratively, including disconfirming evidence.
6. Preserve a clear audit trail without compromising confidentiality.
7. Use triangulation to deepen interpretation, not to manufacture agreement.
8. Treat member reflection as dialogue rather than certification.
9. Provide thick, relevant description and carefully selected quotations.
10. Use CASP for appraisal and COREQ or SRQR for reporting, while exercising methodological judgement.
11. State limitations and avoid claims beyond the evidence.
12. Show how findings may inform practice, policy or further research.

Chapter Summary

Qualitative research quality depends on coherence, transparency, reflexivity and ethical responsibility. Trustworthiness is commonly assessed through credibility, dependability, confirmability and transferability, while authenticity adds attention to fairness, understanding and the potential for change. Triangulation, member reflection, peer debriefing, audit trails, thick description and reflexivity are complementary practices rather than interchangeable checklist items.

Critical appraisal requires readers to examine the fit between the question and methodology, the adequacy of sampling and data generation, the rigour of analysis, the researcher–participant relationship, ethical safeguards and the scope of conclusions. CASP supports structured appraisal; COREQ and SRQR improve reporting transparency. None can substitute for methodological judgement. For health-care managers, the most useful qualitative evidence explains how experiences and organisational processes arise in context, while remaining honest about uncertainty, power and limits of transferability.

Key Terms

Credibility: The plausibility and believability of qualitative findings.

Dependability: The consistency, traceability and logical documentation of the research process.

Confirmability: The extent to which findings are grounded in data rather than researcher invention.

Transferability: The potential relevance of findings in contexts beyond the original study.

Authenticity: Fair and meaningful representation of perspectives, especially those affected by unequal power.

Triangulation: Examination of a phenomenon using multiple sources, methods, researchers, theories or analytic approaches.

Member reflection: Inviting participants to consider summaries or interpretations of their accounts.

Peer debriefing: Critical discussion of research decisions with a knowledgeable independent colleague.

Audit trail: A documented record linking data, analysis, interpretation and methodological decisions.

Thick description: Contextually detailed account that supports interpretation and judgement of transferability.

Reflexivity: Examination of how researcher position, assumptions and relationships shape inquiry.

Negative case: Data that contradict or complicate an emerging interpretation.

Positionality: The researcher’s social, professional and epistemological location in relation to the study.

CASP: Critical Appraisal Skills Programme guidance for evaluating research evidence.

COREQ: A 32-item reporting guideline for interviews and focus groups.

SRQR: Standards for Reporting Qualitative Research across qualitative designs.

Saturation: A claim that additional data are unlikely to add relevant conceptual insight, defined according to the study’s analytic purpose.

Methodological coherence: Alignment among research question, paradigm, design, methods, analysis and claims.

Critical appraisal: Systematic assessment of a study’s credibility, relevance, ethics and usefulness.

Review Questions

5 Short Answer Questions

1. Define credibility and identify two practices that may strengthen it.
2. How does transferability differ from statistical generalisability?
3. What information should an audit trail contain?
4. Why is reflexivity important in qualitative health research?
5. Distinguish COREQ from SRQR.

5 Long Answer Questions

1. Explain credibility, dependability, confirmability, transferability and authenticity, using examples from health-care management research.
2. Critically discuss the benefits and limitations of triangulation and member reflection.
3. Describe how a researcher could establish a defensible chain from interview data to themes and conclusions.
4. Compare the purposes of CASP, COREQ and SRQR in qualitative research.
5. Discuss ethical tensions between thick description and protection of participant confidentiality.

3 Analytical / Application Questions

1. A study claims that clinicians resist a new electronic record because they lack digital skills. It includes interviews only with senior managers and provides no reflexive account. Appraise the claim using at least four trustworthiness criteria and propose a stronger design.
2. You are asked to decide whether findings from a qualitative study of patient navigation in an urban teaching hospital can inform rural hospitals. What contextual information would you require, and how would you judge transferability?
3. A research team disagrees about whether staff accounts indicate “poor communication” or “communication as safety work.” Explain how peer debriefing, negative-case analysis, member reflection and an audit trail could help resolve or productively retain the disagreement.

Multiple Choice Questions with Answers

Question 1

Which criterion concerns whether qualitative findings are believable and plausible to informed readers?

A. Credibility

B. Dependability

C. Transferability

D. Authenticity

Correct Answer: A. Credibility

Question 2

What is the primary purpose of an audit trail?

A. To increase the number of participants

B. To document the pathway from data to interpretation

C. To replace ethical review

D. To calculate statistical generalisability

Correct Answer: B. To document the pathway from data to interpretation

Question 3

Which practice most directly supports transferability?

A. Using only one data source

B. Providing thick description of the context

C. Removing all contextual details

D. Requiring identical findings in replication

Correct Answer: B. Providing thick description of the context

Question 4

Which statement best describes triangulation?

A. Counting how many participants agree with the researcher

B. Using multiple perspectives or sources to deepen analysis

C. Selecting only data that support the main theme

D. Converting all qualitative data into numerical scores

Correct Answer: B. Using multiple perspectives or sources to deepen analysis

Question 5

Why should member reflection not be treated as a simple validity vote?

A. Participants cannot provide any useful information

B. Participants may disagree for reasons that require interpretation

C. Researchers must never share findings

D. Member reflection is relevant only to quantitative studies

Correct Answer: B. Participants may disagree for reasons that require interpretation

Question 6

Which guideline is specifically designed for reporting interviews and focus groups?

A. COREQ

B. CASP

C. ENTREQ

D. SQUIRE

Correct Answer: A. COREQ

Question 7

A study reports themes but does not explain coding, theme development or contradictory cases. Which area is most directly weakened?

A. Data analysis rigour

B. Participant recruitment numbers

C. Statistical power

D. Randomisation

Correct Answer: A. Data analysis rigour

Question 8

What does confirmability primarily require researchers to demonstrate?

A. That every participant gave identical answers

B. That findings are grounded in data and analytic decisions are transparent

C. That the sample is statistically representative

D. That the researcher had no perspective

Correct Answer: B. That findings are grounded in data and analytic decisions are transparent

Question 9

Which is an example of investigator triangulation?

A. Studying participants at two time points

B. Combining interviews with observation

C. Having multiple researchers examine and discuss the data

D. Applying two theories to the same phenomenon

Correct Answer: C. Having multiple researchers examine and discuss the data

Question 10

What is a major risk of using AI-assisted transcription in qualitative health research?

A. It guarantees unbiased interpretation

B. It may misrecognise speech and create confidentiality risks

C. It eliminates the need for data checking

D. It makes reflexivity unnecessary

Correct Answer: B. It may misrecognise speech and create confidentiality risks