03 · What You Need to Know
How to Distinguish Genuine Mixed Methods From Parallel Research
Why Is Collecting Two Types of Data Not Enough?
Quantitative and qualitative research can coexist within a project without constituting an integrated mixed-methods investigation.
For example, a researcher might survey teachers about their frequency of generative AI use and separately interview students about their experiences of academic integrity. Both topics concern educational AI, but their relationship is not necessarily sufficient to justify mixed methods.
The quantitative component may estimate the distribution of particular teaching practices. The qualitative component may investigate students' interpretations of academic integrity. Unless the study establishes how these contributions address a shared question, the project may consist of two related but independent investigations.
Mixed-methods methodology generally places substantial emphasis on integration. Fetters, Curry, and Creswell (2013) describe integration at the design, methods, and interpretation and reporting levels, illustrating that the relationship between components is a methodological concern throughout the investigation.
Researchers should therefore distinguish the use of both numbers and words from a genuinely integrated mixed-methods design.
What Makes Two Components Part of One Coherent Study?
Two components belong within one mixed-methods study when they contribute to a shared research problem through an identifiable methodological or interpretive relationship.
That relationship may involve one component informing another, the components examining complementary aspects of the same phenomenon, or their findings being compared to develop an overarching interpretation.
For example, a quantitative analysis may identify students whose online learning participation differs from their academic performance. Interviews with selected students could then investigate learning practices that help interpret those patterns.
The components answer different questions, but their relationship is clear. The qualitative inquiry responds to a limitation or explanatory need arising from the quantitative findings.
Alternatively, a researcher may collect survey measures of student engagement and interviews about meaningful participation during the same academic period, then systematically compare the findings.
In both cases, integration contributes to an overarching research purpose rather than merely combining two methodologies for variety.
What Are the Warning Signs of Two Disconnected Studies?
Several features may indicate that a proposed mixed-methods study lacks a meaningful relationship between components.
| Warning sign |
Why it matters |
Possible correction |
| The components address unrelated questions. |
There may be no shared problem requiring integration. |
Clarify the overarching research purpose and the relationship between questions. |
| Interviews are added only to obtain quotations. |
Qualitative evidence may serve an illustrative rather than analytical role. |
Specify a substantive qualitative question and appropriate analysis. |
| Both datasets are collected without an integration plan. |
Relevant constructs, samples, or time periods may not align. |
Identify integration points before collection where feasible. |
| Results are reported separately without an overarching interpretation. |
The study may not demonstrate what combining the approaches contributes. |
Develop an integrated analysis or interpretation tied to the research question. |
| The conclusion merely repeats both sets of findings. |
The combined evidence produces no identifiable additional understanding. |
Explain how findings correspond, complement, qualify, or challenge one another. |
| One component could be removed without affecting the overarching contribution. |
The rationale for including that component may be weak. |
Reconsider its purpose, the integration strategy, or the mixed-methods designation. |
These are diagnostic indicators rather than automatic disqualifications. Some mixed-methods designs legitimately involve different populations or apparently distinct component questions.
The important issue is whether the researcher can explain and demonstrate their substantive relationship.
Why Is an Overarching Mixed-Methods Question Useful?
An overarching question identifies what the researcher intends to learn by bringing quantitative and qualitative contributions together.
Consider the following component questions:
Quantitative: What is the relationship between recorded online learning participation and academic performance?
Qualitative: How do students describe their learning practices and experiences of participating in online courses?
These questions could be investigated independently. To establish a mixed-methods purpose, the researcher might formulate an overarching question:
Mixed methods: How do students' accounts of their learning practices help explain the observed relationship between recorded online participation and academic performance?
The overarching question makes the intended integration visible.
However, a separately worded mixed-methods question is not mandatory in every methodological tradition. A clearly articulated overarching aim and integration objective may serve the same function.
What matters is whether the study addresses a research question that genuinely warrants mixed methods.
What Is the Difference Between Separate Analysis and Separate Studies?
This distinction is particularly important because researchers sometimes assume that quantitative and qualitative data must be analyzed together from the beginning.
Separate analysis is entirely compatible with many mixed-methods designs.
For example, a convergent study may analyze survey responses statistically and interview transcripts qualitatively before bringing the findings together for comparison.
Each component retains analytical procedures appropriate to its methodological assumptions and evidence.
The study becomes integrated when the researchers systematically examine the relationship between those findings and develop conclusions addressing the overarching question.
Separate component analyses
Quantitative and qualitative evidence are analyzed using appropriate methods before being meaningfully connected or interpreted together.
Disconnected parallel studies
Quantitative and qualitative investigations proceed independently without a defensible integration purpose or substantive combined interpretation.
The distinction is methodological rather than merely structural. Two results chapters can belong to one integrated study, while a single combined chapter can still conceal two unrelated analyses.
How Can Integration Be Built Into the Research Design?
Integration should be considered when the research questions and design are established, rather than left entirely to the final discussion.
Researchers can identify where one component needs to inform another or where the findings will need to be compared.
Fetters and colleagues describe connecting, building, merging, and embedding as methods-level integration approaches.
Connecting uses information from one component to inform sampling in another. Building uses findings from one component to inform subsequent data collection procedures. Merging brings findings together for analysis. Embedding incorporates a component within a broader design, potentially linking evidence at multiple points.
These approaches offer different ways of establishing meaningful relationships.
The appropriate choice depends on the specific integration problem, not the desire to demonstrate as many integration techniques as possible.
How Do Sequential Designs Avoid Becoming Two Studies?
Sequential mixed-methods designs commonly establish integration through a dependency between phases.
In an explanatory sequential study, quantitative findings may determine which participants are interviewed or which results require further investigation.
In an exploratory sequential study, qualitative findings may inform questionnaire development, intervention design, or subsequent quantitative measurement.
The second component therefore responds to something established by the first.
However, collecting data in sequence does not automatically establish integration. A survey conducted in January and unrelated interviews conducted in March remain disconnected if neither informs the other and no overarching interpretation requires their relationship.
Researchers should document how first-phase findings influenced second-phase decisions and how the resulting evidence contributes to the overall conclusion.
How Do Concurrent Designs Avoid Becoming Two Parallel Studies?
Concurrent designs present a different challenge because quantitative and qualitative data may be collected independently during a similar period.
The researcher must ensure that the components investigate sufficiently related aspects of the phenomenon to support meaningful comparison.
For example, a study examining student engagement could collect learning analytics and interviews during the same semester.
The quantitative component may establish recorded participation patterns, while the qualitative component examines how students understand and experience participation.
Integration would involve comparing these findings to determine what recorded participation represents and where the indicators may be incomplete.
A convergent mixed-methods design may be appropriate when the research purpose requires this relationship.
However, concurrent timing alone does not make the components conceptually comparable.
Do Both Components Need the Same Participants?
No. Mixed-methods integration can occur with overlapping or different samples.
When the researcher intends to compare individual quantitative profiles with corresponding qualitative accounts, using the same participants or an ethically appropriate linkage procedure may be important.
When the study compares broader institutional patterns or different stakeholder perspectives, separate samples may be appropriate.
The researcher should specify the level at which integration occurs and avoid unsupported individual-level or population-level inferences.
For example, interviews with teachers cannot automatically explain the survey responses of particular students merely because both groups belong to the same institution.
Sampling decisions should therefore support the intended relationship between components rather than follow an assumption that mixed methods always requires identical participants.
How Can a Joint Display Demonstrate Integration?
A joint display can organize quantitative and qualitative findings alongside an integrated interpretation.
For example, a study of online learning might align participation indicators with students' accounts of engagement and identify where the findings correspond or differ.
Guetterman, Fetters, and Creswell (2015) describe joint displays as a means of supporting integrated analysis and communicating insights derived from multiple forms of evidence.
However, the display must serve an analytical purpose. Placing statistics in one column and interview quotations in another does not automatically produce integration.
The researcher should explain what the comparison reveals and how it changes, extends, or qualifies the interpretation.
What Is the Added Value of Integration?
The added value of mixed methods concerns what becomes possible through the relationship between components that would remain inadequately addressed by separate investigations.
For example, quantitative analysis may establish an association between recorded online participation and academic performance. Qualitative inquiry may reveal that students use the platform for substantially different purposes.
Integrating the findings may suggest that participation indicators should be interpreted cautiously because similar recorded activity can represent different learning practices.
This interpretation depends on considering both forms of evidence together.
O'Cathain, Murphy, and Nicholl (2010) discuss techniques for integrating qualitative and quantitative evidence, emphasizing that combining findings can produce insights beyond separate analyses.
Nevertheless, integration should not be judged solely by whether it produces an unexpected or dramatic conclusion. Clarifying the limits of an interpretation, identifying meaningful agreement, or documenting unresolved divergence can also be valuable.
Can a Study Be Mixed Methods If Integration Occurs Only During Interpretation?
Potentially, yes. Some legitimate mixed-methods designs collect and analyze quantitative and qualitative evidence independently before integrating findings during interpretation.
The essential requirement is that the integration addresses a clearly defined research purpose and produces a substantive combined interpretation.
For example, a convergent study may compare independently analyzed datasets through a joint display or structured narrative.
However, planning integration only after data collection may create difficulties if the components address incompatible constructs, populations, or time periods.
Researchers should therefore identify the intended integration points before collecting evidence whenever feasible.
What If You Discover the Components Are Disconnected After Data Collection?
The appropriate response depends on whether the existing evidence supports a meaningful relationship that was not fully recognized during initial planning.
Researchers can examine the original questions, constructs, samples, and findings to determine whether a defensible integration is possible.
If the components address related aspects of the same phenomenon, a transparent integrated analysis may be justified. Any newly developed analytical procedures should be distinguished from those planned in advance.
However, researchers should not invent an overarching question retrospectively merely to preserve the mixed-methods designation.
If the datasets are genuinely unrelated, reporting them as separate studies or distinct components of a broader research program may be more accurate.
Additional data collection may sometimes address the missing connection, but it should be justified by the research problem and relevant ethical requirements.
Watch Out
Do not manufacture integration through superficial comparisons or selective quotations. If the evidence cannot support a meaningful relationship, acknowledge that limitation rather than claiming an integrated conclusion that the design cannot defend.
How Should Integration Be Demonstrated in the Final Manuscript?
The final report should explain why mixed methods was selected, how the components were related, where integration occurred, and what the integrated findings contributed.
O'Cathain, Murphy, and Nicholl (2008) examined the quality of mixed-methods studies and identified problems involving insufficient design justification, transparency, and integration.
Their work informed the Good Reporting of A Mixed Methods Study recommendations, commonly called GRAMMS, which emphasize reporting the mixed-methods rationale, design, component methods, integration, added value, and limitations.
These recommendations concern transparent reporting rather than guaranteeing methodological quality.
Researchers should make the integrated contribution visible in the results or discussion and ensure that the conclusion addresses the overarching question rather than merely restating two independent sets of findings.
