When nursing findings disagree across quantitative and qualitative strands, preserve the difference and investigate what it means. Do not change data, remove inconvenient accounts or choose the result that best fits the expected conclusion.
Define the apparent disagreement
Write the numerical finding and qualitative finding separately in accurate language. Then state the specific contradiction you think exists. A high average satisfaction score and some negative interviews may describe variation rather than incompatible findings.
Check whether the measures address the same issue. Satisfaction with an appointment is different from ease of accessing it; confidence in a task differs from observed performance. The first job is to establish whether the comparison is meaningful.
Examine the participants and timing
Compare eligibility, recruitment, response patterns and sample characteristics. Interview volunteers may include people with particularly strong experiences. A survey taken immediately after a visit and interviews held later may capture different perspectives.
Record these possibilities as explanations to examine, not facts about your participants unless the records support them. If the groups cannot be linked, avoid writing as though one person’s interview explains another person’s numerical response.
Review the measurement and analysis
Check scoring instructions, reverse-coded items, missing-value rules and the output used to describe the quantitative pattern. Review qualitative coding and look for accounts that challenge the proposed theme.
Correct a documented error transparently. However, legitimate disagreement is not an error to repair. Keep the original material and an audit trail of any analytical change. Do not recode solely to obtain a preferred relationship between the strands.
Consider context and different meanings
A hypothetical community nursing survey might show high reported access to advice, while interviews describe difficulty obtaining advice outside office hours. The strands may define access at different levels.
A useful discussion would explain that difference and assess what each result actually establishes. It should not claim that the survey is false or that the interviews represent everyone in the service. Clearly label hypothetical examples in your own planning notes.
Write a cautious integrated interpretation
Use language that identifies both the tension and its boundaries: the survey suggests one pattern in its sample, while interviews illuminate a difficulty experienced by some participants. Explain the design or context that may account for the difference.
Fetters, Curry and Creswell (2013) place integration at several levels of mixed-methods work. In your project, make the relationship between strands explicit and distinguish supported explanations from unresolved questions.
Report what remains uncertain
You may not be able to explain the difference fully. State what information would be needed, such as linked observations, a different measurement period or a broader sample. An honest unresolved finding is stronger than an unsupported reconciliation.
Use a joint display to show comparable findings and limitations. For differences within qualitative accounts alone, use the existing participant-account guide rather than treating that as a mixed-methods problem.
When nursing findings disagree: compare possible explanations
Illustrative planning example: First, imagine favourable appointment ratings alongside interview reports of difficult access. Check whether both strands actually examine the same issue. Someone may value a consultation after struggling to obtain it. Next, compare sampling and timing. Finally, preserve the difference and explain the most defensible interpretation. Do not alter scores or remove accounts to manufacture agreement.
| Element | Example | Meaning or limit |
|---|---|---|
| Different concepts | Satisfaction versus ease of access | Check whether the apparent contradiction is meaningful |
| Different timing | Immediate questionnaire versus later interview | Describe the relevant periods |
| Different samples | All respondents versus interview volunteers | Avoid assuming individual linkage |
Practical checklist
- State each finding accurately.
- Compare concepts and samples.
- Check documented analysis errors.
- Preserve legitimate disagreement.
Frequently asked questions
Does disagreement invalidate the study?
Not automatically. Assess whether it reflects comparability, measurement, sample differences, error or a meaningful tension.
Should I remove contradictory interview extracts?
No. Relevant contrary accounts should inform the analysis and its limitations.
Must I explain every discrepancy?
Explain what you can support and acknowledge what remains uncertain.
Should I choose whichever finding has more participants?
No. Sample size alone does not decide which account answers the question. Assess design, relevance and the meaning of each source.
Conclusion
When nursing findings disagree, investigate comparability before treating the difference as an error. A transparent interpretation preserves both strands and explains what remains unresolved.
Get feedback on your project
For a review of your own question, plan or draft, use mixed-methods integration support. Send your university brief and supervisor comments through our contact page. Follow your university’s rules on external support and share only material you are authorised to disclose.
References
Fetters, M.D., Curry, L.A. and Creswell, J.W. (2013) ‘Achieving integration in mixed methods designs—principles and practices’, Health Services Research, 48(6 Pt 2), pp. 2134–2156. Available at: Full article.
