06. Build an interview guide and integrate mixed-method findings
Lesson
LEARNING OBJECTIVES
Write neutral interview questions and probes, explain why two methods are needed, and use an integrated display to examine agreement and disagreement.
INTERVIEW-GUIDE PRACTICE
Begin with an accessible invitation to describe an experience, move to relevant topics and close with an opportunity to add something important. Avoid questions that assume a preferred explanation.
For a fictional safe-water study:
Opening: “Tell me how your household obtained drinking water yesterday.”
Process probe: “What happened next?”
Context probe: “What made that easier or more difficult?”
Clarifying probe: “What do you mean by ‘reliable’?”
Closing: “What have we not discussed that would help us understand this?”
Avoid “Why don't you use the correct treatment?” because it assumes a correct practice and may sound accusatory. Pilot the guide, consent process and interview setting. Record relevant context and a reflexive memo without unnecessary identifiers.
MIXED METHODS REQUIRES INTEGRATION
Using a questionnaire and interviews in one project is not enough. Specify what each contributes, their timing, sampling relationship and where the results will be brought together.
An explanatory sequential plan might begin with a survey and use interviews to explore unexpected patterns. An exploratory sequential plan might use qualitative work to develop concepts or measures for later quantitative study. A convergent plan might collect strands in parallel and compare their insights. Choose because of the question, not because combining methods sounds automatically stronger.
WORKED FICTIONAL INTEGRATED DISPLAY
Survey finding: 40 of 100 responding households reported treating water during a defined period.
Interview insight: constructed teaching accounts describe fuel cost, time and uncertainty about instructions as barriers. These are invented examples, not real quotations or measured community frequencies.
Integrated interpretation: resource and communication issues may help explain patterns, but the interview sample does not establish how common each barrier is.
Remaining question: do reported practices match actual practices, and were non-responding households different?
Potential next step: refine measurement and assess a feasible implementation strategy before claiming a programme effect.
If interviews suggest distrust while a survey shows high satisfaction, investigate wording, social desirability, sampling and context. Do not suppress the disagreement or declare that one strand “validates” the other automatically.
YOUR TASK
Draft six interview questions with neutral probes and a short reflexivity note. Then create a joint display with columns for quantitative finding, qualitative interpretation, relationship between strands, unresolved issue and possible implication.
Use three fictional entries, including one disagreement. Explain how participants would be selected for each strand, whether records need linkage, what consent permits and how privacy would be protected.
SELF-CHECK AND FEEDBACK
Can interviews explain a pattern without estimating its prevalence? Yes, within the design's interpretive limits. Does agreement between methods prove causation? No. Should mixed-method conclusions acknowledge a weak strand? Yes; integration cannot erase its limitations.
PORTFOLIO OUTPUT
An interview guide, a mixed-method design rationale and a transparent joint display.
FURTHER READING
NIH OBSSR: Best Practices for Mixed Methods Research in the Health Sciences.
https://obssr.od.nih.gov/research-resources/mixed-methods-research
EQUATOR: COREQ.
https://equator-network.org/reporting-guidelines/coreq/
Write neutral interview questions and probes, explain why two methods are needed, and use an integrated display to examine agreement and disagreement.
INTERVIEW-GUIDE PRACTICE
Begin with an accessible invitation to describe an experience, move to relevant topics and close with an opportunity to add something important. Avoid questions that assume a preferred explanation.
For a fictional safe-water study:
Opening: “Tell me how your household obtained drinking water yesterday.”
Process probe: “What happened next?”
Context probe: “What made that easier or more difficult?”
Clarifying probe: “What do you mean by ‘reliable’?”
Closing: “What have we not discussed that would help us understand this?”
Avoid “Why don't you use the correct treatment?” because it assumes a correct practice and may sound accusatory. Pilot the guide, consent process and interview setting. Record relevant context and a reflexive memo without unnecessary identifiers.
MIXED METHODS REQUIRES INTEGRATION
Using a questionnaire and interviews in one project is not enough. Specify what each contributes, their timing, sampling relationship and where the results will be brought together.
An explanatory sequential plan might begin with a survey and use interviews to explore unexpected patterns. An exploratory sequential plan might use qualitative work to develop concepts or measures for later quantitative study. A convergent plan might collect strands in parallel and compare their insights. Choose because of the question, not because combining methods sounds automatically stronger.
WORKED FICTIONAL INTEGRATED DISPLAY
Survey finding: 40 of 100 responding households reported treating water during a defined period.
Interview insight: constructed teaching accounts describe fuel cost, time and uncertainty about instructions as barriers. These are invented examples, not real quotations or measured community frequencies.
Integrated interpretation: resource and communication issues may help explain patterns, but the interview sample does not establish how common each barrier is.
Remaining question: do reported practices match actual practices, and were non-responding households different?
Potential next step: refine measurement and assess a feasible implementation strategy before claiming a programme effect.
If interviews suggest distrust while a survey shows high satisfaction, investigate wording, social desirability, sampling and context. Do not suppress the disagreement or declare that one strand “validates” the other automatically.
YOUR TASK
Draft six interview questions with neutral probes and a short reflexivity note. Then create a joint display with columns for quantitative finding, qualitative interpretation, relationship between strands, unresolved issue and possible implication.
Use three fictional entries, including one disagreement. Explain how participants would be selected for each strand, whether records need linkage, what consent permits and how privacy would be protected.
SELF-CHECK AND FEEDBACK
Can interviews explain a pattern without estimating its prevalence? Yes, within the design's interpretive limits. Does agreement between methods prove causation? No. Should mixed-method conclusions acknowledge a weak strand? Yes; integration cannot erase its limitations.
PORTFOLIO OUTPUT
An interview guide, a mixed-method design rationale and a transparent joint display.
FURTHER READING
NIH OBSSR: Best Practices for Mixed Methods Research in the Health Sciences.
https://obssr.od.nih.gov/research-resources/mixed-methods-research
EQUATOR: COREQ.
https://equator-network.org/reporting-guidelines/coreq/
My study notes
Stored only in this browser. Do not enter patient information.Saved in this browser. No account needed.