Conflicting participant accounts in a nursing dissertation are not automatically errors that must be reconciled or removed. They may reveal differences in role, experience, context, power, timing or interpretation that are central to the research question. The academic task is to analyse those differences transparently without forcing agreement, counting opinions as though qualitative data were survey results, or deciding that one participant represents the “truth”.
This guide provides a step-by-step method for handling contradictory interview or focus-group accounts in qualitative nursing research. It explains how to verify the apparent contradiction, compare cases, refine codes and themes, use negative-case analysis, write a defensible findings section and discuss what divergence means for nursing practice.
Quick answer: return to the complete transcripts, define exactly what conflicts, compare the speakers’ contexts, test the pattern across the dataset, document your analytical decision and report both the dominant account and meaningful divergence. Contradiction should change or qualify the interpretation when it is relevant; it should not be hidden merely because it complicates a neat theme.
What Are Conflicting Participant Accounts?
Conflicting participant accounts occur when two or more participants describe, explain or evaluate the same phenomenon differently. One nurse may describe a discharge tool as helpful, while another experiences it as inflexible. A patient may report feeling involved in a decision, whereas a family carer describes the process as rushed. The accounts may appear mutually incompatible, but they can also represent different positions within the same care pathway.
Contradiction can occur between participants, between professional groups, within one participant’s interview, or between what someone says and what another data source shows. The response depends on the study design and epistemological position. Therefore, do not treat every difference as the same methodological problem.
Key principles
- Difference is not automatically poor-quality data.
- A majority view is not automatically the most analytically important view.
- Context should be examined before accounts are combined.
- Quotations illustrate an interpretation; they do not replace analysis.
- Researchers must show how themes and conclusions were produced.
- Confidentiality remains essential when describing unusual cases.
Why Contradictory Accounts Matter in Nursing Research
Nursing research often examines complex care delivered across roles, settings and relationships. People can experience the same service differently because their information, authority, needs and contact with the service are different. Variation may therefore reveal something that a simple consensus statement would miss.
For example, ward leaders may describe a new safety huddle as consistently implemented, while bedside nurses report that staffing pressures make it irregular. These accounts do not require the researcher to select a winner. Instead, the tension may indicate a difference between policy-level expectations and operational experience.
Divergence can strengthen a qualitative dissertation when it:
- identifies conditions under which an experience changes;
- reveals unequal access to information or decision-making;
- shows that a theme is conditional rather than universal;
- challenges an assumption held at the beginning of the study;
- distinguishes professional intention from patient experience; or
- generates a more precise recommendation for nursing practice.
However, an apparent contradiction can also result from vague interview questions, inconsistent coding, loss of context or quotations being read too narrowly. Analysis must begin by checking the data rather than immediately creating a new theme.
First Decide What Kind of Conflict You Have
| Type of conflict | Example | Analytical question |
|---|---|---|
| Between participants | Two nurses evaluate the same handover process differently | What differs in their roles, shifts, experience or exposure? |
| Between stakeholder groups | Patients report limited choice while staff describe care as collaborative | Do power or expectations shape the accounts? |
| Within one account | A participant first calls a system useful but later describes avoiding it | Is this inconsistency, ambivalence or context-dependent experience? |
| Across time | An early interview is positive and a later interview is critical | Did experience, policy or circumstances change? |
| Across data sources | Interview claims differ from observations or documents | What can each source legitimately show? |
| Interpretive disagreement | Researchers understand the same extract differently | Which assumptions or theoretical positions produce each reading? |
This classification prevents premature conclusions. It also stops the researcher from presenting all disagreement as evidence that participants are unreliable.
Step 1: Verify the Apparent Contradiction
Return to the complete transcript, not only the coded extract. Read the material before and after the quotation. Check the question that prompted it, whether the participant was discussing the same setting and whether the time frame changed.
Next, listen to the recording where permitted and necessary. Tone, pauses and emphasis may clarify whether a statement was uncertain, ironic or qualified. Correct any transcription error through the approved audit process; never silently alter a quotation to make the accounts agree.
Ask four verification questions:
- Are the participants discussing the same phenomenon?
- Are they referring to the same point in time?
- Were the interview questions sufficiently comparable?
- Does the apparent conflict remain when both extracts are read in context?
If the answer to the last question is no, revise the code. If the conflict remains, retain it as an analytical lead.
Step 2: Build a Participant-Divergence Matrix
A compact matrix helps compare accounts without reducing them to votes. Use participant identifiers that comply with the approved confidentiality plan. Record only contextual characteristics that are relevant to the research question.
| Analytical field | What to record |
|---|---|
| Issue | The precise point on which the accounts differ |
| Account A | A concise, contextualised summary |
| Account B | The contrasting or qualifying summary |
| Relevant context | Role, setting, timing or experience linked to the difference |
| Supporting extracts | Transcript locations and candidate quotations |
| Dataset test | Where else the pattern appears or fails to appear |
| Interpretive decision | Retain, split, qualify, rename or reject the proposed theme |
| Reflexive note | How the researcher’s assumptions may affect interpretation |
The matrix is a thinking tool, not a substitute for immersion in the data. Do not include identifiable clinical details simply because they appear relevant. Unusual combinations of role, ward, diagnosis and incident may enable deductive identification even when names have been removed.
Step 3: Compare Cases Without Counting Opinions
Qualitative comparison asks how meaning varies and under what conditions. It does not automatically ask which view received the most votes. Return to all cases coded to the issue, then search for confirming, qualifying and disconfirming material.
For each account, examine:
- the participant’s relationship to the phenomenon;
- what information or authority the participant possessed;
- the clinical or organisational setting;
- whether the account concerns routine practice or an exceptional event;
- the language used to express certainty, hesitation or ambivalence; and
- whether the difference answers or falls outside the research question.
Frequency may be described when it genuinely assists the analysis, but avoid claims such as “most participants believed” unless your method supports the comparison and the denominator is clear. The importance of an account can come from its capacity to change the interpretation, not only from how often it appears.
Step 4: Use Negative-Case Analysis Carefully
A negative or disconfirming case is data that does not fit a developing explanation. Its purpose is not to destroy a theme whenever one participant differs. Instead, it tests the boundaries of the interpretation.
Suppose a developing theme states that electronic handover increased nurses’ confidence. One participant reports lower confidence because temporary staff could not access the system. That account may show a condition: the benefit depended on reliable access and orientation. The stronger theme becomes “digital handover supported confidence when access and preparation were adequate.”
Possible analytical responses include:
- Qualify the theme: state the circumstances in which the pattern holds.
- Split the theme: create distinct patterns when the accounts reflect different meanings.
- Create a subtheme: retain a coherent exception that explains variation within a broader pattern.
- Rename the theme: replace an overgeneralised label with a more accurate central concept.
- Reject the theme: remove an interpretation that cannot explain the dataset adequately.
- Retain a bounded divergence: report a meaningful contrasting account without claiming it is a separate theme.
The choice must fit the adopted qualitative methodology. Reflexive thematic analysis, framework analysis, grounded theory, interpretative phenomenological analysis and qualitative content analysis do not make identical assumptions. Follow the approach declared in the methodology chapter rather than borrowing techniques inconsistently.
Step 5: Review Codes and Theme Boundaries
Contradiction sometimes appears because an initial code is too broad. A code such as “good communication” may combine timely information, emotional reassurance, opportunity to ask questions and involvement in decisions. Participants may agree on one dimension and disagree on another.
Review the code against the original extracts. Ask whether it:
- captures one coherent meaning;
- distinguishes experience from evaluation;
- preserves who did what and in which context;
- contains both semantic statements and latent interpretations without explanation; or
- has been shaped by the interview guide rather than the dataset.
Then test the developing theme across the complete dataset. Byrne (2022) explains that reflexive thematic analysis treats the researcher as actively involved in interpretation and organises codes around a central concept. Therefore, a theme should not become a container for every response to one interview question.
Students needing broader chapter-level troubleshooting can use the separate guide on nursing dissertation discussion chapter problems. The present article remains limited to participant-level contradiction.
Step 6: Keep a Transparent Reflexive Record
Reflexivity requires more than declaring that bias exists. Record the specific assumption, decision and consequence. For example, a researcher with experience in acute nursing may initially interpret rapid communication as efficient, whereas a participant experiences it as excluding questions. The reflexive note should explain how revisiting the account changed or qualified the analysis.
An audit entry can contain:
- the date and stage of analysis;
- the conflicting extracts reviewed;
- the initial interpretation;
- the alternative interpretation considered;
- the contextual evidence examined;
- the decision made; and
- the effect on codes, themes or conclusions.
If more than one researcher is involved, discuss differences in interpretation in a way that fits the declared methodology. In reflexive thematic analysis, Byrne (2022) notes that collaboration can enrich interpretation rather than simply force coder consensus. Do not report inter-rater reliability automatically when it conflicts with the analytical approach.
Step 7: Write Conflicting Accounts in the Findings Chapter
The findings section should make the analytical pattern clear before presenting quotations. Begin with the theme or claim, show the dominant or shared pattern where relevant, introduce the divergence and explain how it changes the interpretation.
A practical paragraph sequence
- Analytical claim: state the theme’s central meaning.
- Pattern: explain how the account operated across relevant cases.
- Illustration: use a concise quotation with an anonymised identifier.
- Divergence: introduce the contrasting account neutrally.
- Interpretation: explain the condition, tension or boundary revealed.
- Transition: connect the refined interpretation to the next point.
Worked nursing example
Weak reporting: “Most nurses liked bedside handover, but Participant 6 disagreed.”
Stronger reporting: “Bedside handover was commonly described as improving continuity because patients could clarify information during transfer. However, this benefit was conditional rather than universal. One nurse working with patients requiring sensitive psychosocial discussion described moving parts of the handover away from the bedside to protect privacy. The contrasting account reframed bedside participation as a negotiated practice shaped by confidentiality, patient preference and clinical context.”
The stronger version does not dismiss the exception. It also avoids claiming that one format is always superior. If you need models of how dissertation elements perform different jobs, consult the site’s nursing dissertation examples and adapt only the analytical principle to your own data.
Step 8: Discuss What the Divergence Means
The discussion should move beyond announcing disagreement. Explain what the contradiction contributes to the research question and how it relates to previous evidence, methodology and nursing practice.
A useful discussion sequence is:
- Restate the refined finding without repeating all quotations.
- Explain the context that appears to produce the difference.
- Compare the interpretation with genuinely relevant literature.
- Consider an alternative explanation.
- State what the study design can and cannot establish.
- Develop a proportionate implication for nursing practice or research.
Use cautious verbs such as “suggests”, “indicates”, “may reflect” and “was interpreted as” where appropriate. Avoid “proves”, “demonstrates for all nurses” or “confirms” when a small contextual qualitative study cannot justify those claims.
Contradictory patient and professional accounts require particular care. A professional account should not automatically be treated as more accurate because it uses clinical language. Equally, the researcher should not invent a patient preference or generalise one person’s account. The guide on integrating patient voice in nursing academic work provides additional safeguards for representing preferences and experience.
How Reporting Guidelines Support Transparency
Reporting guidelines are useful checks, although completing a checklist does not by itself prove that an analysis is rigorous. The Standards for Reporting Qualitative Research cover the whole qualitative report and emphasise transparent reporting of the approach, methods, findings and interpretation (O’Brien et al., 2014).
For studies using interviews or focus groups, COREQ includes reporting areas such as research-team reflexivity, study design, derivation of themes and the use of participant quotations (Tong, Sainsbury and Craig, 2007). Use the guideline that fits the design and your university’s instructions. Do not cite COREQ as though it were the method used to analyse the data.
Protect Confidentiality When Reporting an Unusual Account
A divergent case may be memorable precisely because it is unusual. That makes deductive identification more likely. Remove or generalise non-essential details while preserving the meaning needed for interpretation.
Before including a quotation, check:
- whether the approved consent covers quotation;
- whether the identifier follows the study protocol;
- whether names, locations or rare roles appear in the extract;
- whether combined details could identify the participant;
- whether redaction changes the meaning; and
- whether paraphrase is permitted and methodologically appropriate.
Never invent a quotation, change its meaning or merge several participants into one quotation without transparent methodological justification. Ethical academic support should help you analyse your own approved data; it should not fabricate participants, transcripts, codes or findings.
Common Mistakes to Avoid
- Deleting accounts that do not fit the preferred conclusion.
- Calling every difference a separate theme.
- Using participant frequency as the only measure of importance.
- Assuming contradiction means somebody was dishonest.
- Removing context to make quotations appear more decisive.
- Choosing one dramatic quotation without checking the full dataset.
- Mixing incompatible qualitative methods.
- Claiming that member checking guarantees a single correct interpretation.
- Reporting disagreement without explaining its analytical meaning.
- Including identifiable details from an unusual case.
- Adding new data collection after approval has ended.
- Overgeneralising a context-dependent account to all nursing settings.
Final Checklist for Conflicting Participant Accounts
- Define the exact point of contradiction.
- Re-read complete transcripts and verify transcription accuracy.
- Check whether participants refer to the same context and time.
- Classify the type of divergence.
- Compare the relevant cases across the dataset.
- Review whether codes are too broad or inconsistent.
- Test the developing explanation against disconfirming material.
- Record reflexive and analytical decisions.
- Refine, split, qualify or reject the theme as justified.
- Select concise quotations that preserve meaning and confidentiality.
- Explain the divergence rather than merely displaying it.
- Keep claims within the study design and dataset.
- Check COREQ or SRQR where appropriate.
- Confirm alignment between findings, discussion and conclusion.
Frequently Asked Questions
Do conflicting accounts make a qualitative study unreliable?
No. Different accounts can reflect real variation in position, context or experience. Credibility depends on how transparently the researcher examines and reports the divergence, not on producing artificial agreement.
Should every contradictory account become a theme?
No. A contrasting account may qualify a theme, form a subtheme, reveal a boundary or remain a meaningful exception. The decision depends on its relevance and relationship to the complete dataset.
Can I say that most participants agreed?
Only when that statement is supported and useful within the chosen method. Explain the denominator and context where needed. Do not turn qualitative analysis into an informal vote or assume the most frequent account is automatically the most significant.
What if one participant contradicts themselves?
Read both statements in context. The difference may reflect time, circumstances, ambivalence or the complexity of experience. Avoid labelling the participant inconsistent until those possibilities have been examined.
How many quotations should illustrate the disagreement?
There is no universal number. Use enough well-chosen extracts to support the interpretation without turning the chapter into a transcript. Follow the assessment brief and protect participant confidentiality.
Should my supervisor decide which account is correct?
Your supervisor can challenge the logic and methodological fit of your interpretation. However, the dissertation should transparently show how you reached the analytical decision from the data and declared approach.
Get Support with a Qualitative Nursing Dissertation
If contradictory interviews have left your coding or themes unclear, specialist guidance can help you review the research question, methodology, anonymised analysis and chapter structure. Explore our nursing dissertation writing and research support, review the broader nursing research services, or send your brief and approved project details for a focused plan. Support must remain consistent with university rules, ethics approval and your ownership of the research.
Conclusion
Conflicting participant accounts in a nursing dissertation should be examined as potential evidence about context, difference and the boundaries of an interpretation. Verify each apparent contradiction, compare cases, revisit codes, test themes against disconfirming material and keep a transparent reflexive record. Then report the divergence in a way that explains what it changes rather than hiding it or treating it as a vote. This approach produces a more credible, nuanced and defensible account of qualitative nursing data.
References
Braun, V. and Clarke, V. (2021) ‘One size fits all? What counts as quality practice in (reflexive) thematic analysis?’, Qualitative Research in Psychology, 18(3), pp. 328–352. Available at: https://doi.org/10.1080/14780887.2020.1769238 (Accessed: 6 September 2026).
Byrne, D. (2022) ‘A worked example of Braun and Clarke’s approach to reflexive thematic analysis’, Quality & Quantity, 56, pp. 1391–1412. Available at: https://doi.org/10.1007/s11135-021-01182-y (Accessed: 6 September 2026).
O’Brien, B.C., Harris, I.B., Beckman, T.J., Reed, D.A. and Cook, D.A. (2014) ‘Standards for reporting qualitative research: a synthesis of recommendations’, Academic Medicine, 89(9), pp. 1245–1251. Available at: https://doi.org/10.1097/ACM.0000000000000388 (Accessed: 6 September 2026).
Tong, A., Sainsbury, P. and Craig, J. (2007) ‘Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups’, International Journal for Quality in Health Care, 19(6), pp. 349–357. Available at: https://doi.org/10.1093/intqhc/mzm042 (Accessed: 6 September 2026).
