
A hospital team in Multan, led by Dr. Farah, surveyed 300 patients about a new diabetes clinic. The numbers looked good: 78 percent said they were satisfied. Then the team interviewed 24 of those patients, and the interviews told a different story. Patients praised the doctors but described waiting three hours for a ten-minute appointment, and several said they had stopped coming. The survey showed the average; the interviews showed why the average was misleading.
A mixed methods research paper combines qualitative and quantitative approaches in a single study. You might survey a large group of patients and then interview a smaller group to understand why the survey results look the way they do. Neither strand alone would tell the full story; together, they answer a question that one method cannot. Mixed methods papers are common in health, education, and the social sciences, and they are judged on more than the quality of each part. Reviewers want to see a clear reason for combining methods, a design that fits that reason, and findings that are genuinely integrated rather than reported side by side. This guide explains how to design, write, and report a mixed methods paper.
A mixed methods paper reports a study that collects and analyses both qualitative data, such as interviews or observations, and quantitative data, such as surveys or measurements, and integrates the two. The key word is integrates: the paper is not two small studies stapled together, but one study in which each strand informs the other. A survey can show how widespread a problem is, while interviews can explain how people experience it.
Take Dr. Farah's Multan clinic study. The survey asked 300 patients to rate their satisfaction on a five-point scale. The interviews asked 24 of those same patients to describe a recent visit in their own words. The integration happened in the analysis: the team grouped the interview transcripts by satisfaction score, and found that the "satisfied" patients still described the same three-hour wait — they were satisfied with the doctor, not with the visit. That finding became the paper's main recommendation: fix the appointment system, not the doctors. A paper that reported the two strands separately would have missed it entirely.
The three most common designs are convergent, explanatory sequential, and exploratory sequential, and each fits a different reason for mixing methods. In a convergent design, both strands run at the same time and are compared at the end — for example, a survey and interviews on the same topic, run in the same month. In an explanatory sequential design, the quantitative strand comes first and the qualitative strand explains its results — for example, trial results followed by patient interviews. In an exploratory sequential design, the qualitative strand comes first and its findings shape the quantitative instrument — for example, interviews that shape a new questionnaire. A fourth design, embedded, places one strand inside a larger study of the other kind, such as interviews inside a clinical trial.
Choose the design that matches your reason for mixing methods, not the one that looks simplest. A convergent design is efficient but demanding, because both strands must be ready at the same time. Sequential designs take longer, but each stage can build on the last: in an explanatory sequential study, allow four to six weeks between finishing the survey analysis and starting the interviews, because the interview guide must be written from the survey results. Name your design explicitly in the methods section. Our guide on how to choose a research methodology explains how to weigh the options for your question.
A mixed methods study usually needs three kinds of questions: a quantitative question, a qualitative question, and a mixed methods question that asks how the two strands relate. The mixed methods question is the one reviewers look for first, because it shows you planned the integration rather than adding it as an afterthought. Write each question in the language of its own tradition. Quantitative questions ask about relationships, differences, or effects; qualitative questions ask about experiences, meanings, or processes.
A weak mixed methods question says only that the study will use interviews and a survey. A strong question names both strands and the way they relate. Weak: "This study uses a survey and interviews to study patient satisfaction." Strong: "How do patients' accounts of their clinic visits explain the pattern of satisfaction scores in the survey, and where do the two strands disagree?" The strong version tells the reader exactly what the integration will do. If you cannot write such a question, the design probably needs more thought before data collection begins.
Plan data collection so the two strands can speak to each other. If interviews are meant to explain survey results, sample interview participants from the survey respondents, so the two sets of data describe the same people. In the Multan study, the team interviewed 24 patients chosen from the 300 survey respondents: eight very satisfied, eight neutral, and eight dissatisfied. That nested sample let them compare what each group said against what each group had scored.
If the strands use different samples, explain why that is appropriate and what it means for the conclusions. Keep the quality standards of each tradition. Quantitative data need a sampling plan, validated instruments, and a clear analysis plan. Qualitative data need a purposeful sample, a transparent interview or observation guide, and a stated approach to analysis. Cutting corners on one strand weakens the whole paper, because reviewers judge each strand on its own terms. Plan the timing as carefully as the methods: in a sequential design, the second strand cannot start until the first has been analysed, so build that analysis time into the schedule.
Analyse the quantitative strand with the statistics suited to your design and sample size, and report the results with the same rigour as a purely quantitative paper. Analyse the qualitative strand with a named approach, such as thematic analysis or content analysis, and show how themes were developed rather than simply listing them. Do not let one strand dominate the analysis while the other becomes decoration. If the paper claims to be mixed methods, both analyses must be reported in enough detail for a reader to judge them.
A common weakness is a strong survey analysis paired with a few interview quotes and no account of how the qualitative data were analysed. Avoid it by writing the qualitative analysis as carefully as the quantitative one. State the approach — for example, "we used thematic analysis following Braun and Clarke, coding the transcripts in two rounds and resolving disagreements by discussion" — and name the themes with the evidence behind each. A reader should be able to follow how you got from 24 transcripts to three themes, just as they can follow the survey analysis.
Integration is where a mixed methods paper earns its name, and a joint display is the clearest way to show it. A joint display is a table or figure that places quantitative results and qualitative findings side by side, so the reader can see where they agree, where they disagree, and what each adds to the other. Build the joint display around your mixed methods question, not around convenience.
Here is a small joint display from the Multan study, in words. Row one: the survey found that 78 percent of patients were satisfied overall; the interviews showed that satisfaction referred to the doctor, while the same patients described three-hour waits. Row two: the survey found that patients over 60 were the most satisfied group; the interviews showed that older patients were the least likely to complain, even when they waited longest. Row three: the survey found that only 12 percent missed a follow-up appointment; the interviews showed that several patients had stopped coming entirely and so never appeared in the survey at all.
A mixed methods paper usually follows the standard structure — introduction, methods, results, discussion — but the methods and results sections carry more weight. In the methods, describe each strand separately and then explain how they were integrated, including the order in which the strands ran and the point at which they came together. In the results, report each strand's findings and then present the integrated results, often through a joint display. In the discussion, interpret the combined findings rather than summarising each strand separately.
Word limits are a real problem for mixed methods papers, because two strands and their integration take more space than one. Many journals allow a longer manuscript for mixed methods work, so check the limit before you write, and ask the editor if a longer paper is acceptable. If the limit is strict, move detail into supplementary materials: the interview guide, the full survey, or extra joint displays.
Two reporting standards help mixed methods authors cover what reviewers expect. GRAMMS, the Good Reporting of a Mixed Methods Study, is a short checklist covering the justification, the design, the sampling, and the integration. JARS-Mixed, from the American Psychological Association, is a fuller set of reporting standards for mixed methods research in psychology and related fields. You do not need to follow both, but follow at least one and say which in your methods section.
Using a recognised checklist also helps you spot gaps before submission — a missing integration section or an unstated design is much easier to fix at the drafting stage than after peer review. Read the checklist before you write, not after.
Contradictions between strands are common and valuable. A survey may show that most patients are satisfied, while interviews reveal deep frustration with one part of the service. The contradiction does not mean one strand is wrong; it usually means the two methods measured different things, or that different groups had different experiences. Report the contradiction openly and work through what it means.
When the strands disagree, say so plainly in the discussion and explore why. Perhaps the survey measured satisfaction with the service as a whole, while the interviews asked about one difficult part of it. Perhaps different groups had different experiences, and the average hid the difference. A contradiction that is explained adds to the paper; a contradiction that is hidden undermines it. Reviewers respect authors who face a contradiction honestly; they distrust papers in which two strands agree suspiciously well.
The first mistake is adding a small qualitative component to a quantitative study and calling the result mixed methods — twelve interviews quoted once do not make a mixed methods paper. The second is failing to state a mixed methods question, so the integration has no purpose. The third is reporting the two strands in separate silos with no integration. The fourth is choosing a design that does not match the stated reason for mixing methods — for example, running both strands at the same time when the interviews were meant to explain the survey results. Avoid them by deciding the design and the integration plan before data collection, and by checking your draft against GRAMMS or JARS-Mixed before you submit.
| Design | How it works | When to use it | Example |
|---|---|---|---|
| Convergent | Both strands run at the same time | To compare and confirm findings | Survey plus interviews on the same topic |
| Explanatory sequential | Quantitative first, qualitative explains it | To explain surprising survey results | Trial results followed by patient interviews |
| Exploratory sequential | Qualitative first, quantitative tests it | To build an instrument from findings | Interviews shape a new questionnaire |
| Embedded | One strand supports the main strand | To add depth to a primary study | Interviews inside a clinical trial |
These authoritative external resources explain the standards and tools mentioned in this guide:
What is a mixed methods research paper?
A mixed methods research paper reports a study that collects and analyses both qualitative and quantitative data and integrates the two, so that the combined findings answer a question neither method could answer alone. The key word is integrates: the paper is one study in which each strand informs the other, not two small studies stapled together.
What is a joint display in mixed methods research?
A joint display is a table or figure that places quantitative results and qualitative findings side by side, making it easy to see where the strands agree, disagree, or add to each other. Build it around your mixed methods question, not around convenience, and say plainly where the strands agree and where they do not.
What is the difference between convergent and sequential designs?
In a convergent design, both strands run at the same time and are compared at the end. In a sequential design, one strand runs first and shapes the next: quantitative first is explanatory, qualitative first is exploratory. Choose the design that matches your reason for mixing methods, not the one that looks simplest, and name it in the methods section.
Do I need a mixed methods research question?
Yes. Alongside a quantitative and a qualitative question, a mixed methods study needs a question about how the strands relate, such as how interview findings explain the survey results. Reviewers look for this question first, because it shows that integration was planned rather than added afterwards.
What are GRAMMS and JARS-Mixed?
GRAMMS is a short checklist for reporting a mixed methods study, and JARS-Mixed is the American Psychological Association's fuller set of reporting standards. Both help authors cover the design, sampling, and integration that reviewers expect. Follow at least one of them, say which in your methods section, and use it to spot gaps before submission.
Can the two strands of a mixed methods study disagree?
Yes, and disagreement is valuable. Report it openly, check whether the samples or measures differ, and explain what the contradiction reveals. Reviewers respect honest handling of disagreement more than forced agreement. A contradiction that is explained adds to the paper; a contradiction that is hidden or smoothed over undermines it.
Submit your manuscript ← Back to GAP Insights
Useful publishers, databases, funders, tools and institutions for researchers: