Is a Study Really Mixed Methods if It Merely Reports Two Separate Analyses?

Can two separate analyses still constitute mixed methods research?

You reach the results section of a paper and find two familiar territories. First come the quantitative findings: descriptive statistics, models, effect estimates, or hypothesis tests. Then the paper changes methodological gears and presents interviews, themes, observations, or qualitative interpretations.

Both components may be rigorous. Both may address the same broad topic. But if they are analyzed and reported separately, is the study genuinely mixed methods, or are you simply reading two studies sharing one manuscript?

The distinction matters because separate analysis is not itself a methodological flaw. Mixed methods research often requires quantitative and qualitative data to be analyzed using their own appropriate procedures before they are integrated. The real issue is whether the separation persists throughout the study or whether the two components eventually interact in a purposeful way.

Separate analysis is compatible with mixed methods, but separation cannot be the whole story

Separate analysis and methodological separation are not the same thing

Quantitative and qualitative data usually require different analytic procedures

There is nothing suspicious about seeing statistical analysis and qualitative analysis conducted separately. The two forms of evidence generally require different analytic procedures because they answer questions through different forms of data and inference.

A quantitative component might estimate associations, distributions, differences, or effects. A qualitative component might identify patterns of meaning, processes, experiences, mechanisms, or contextual explanations. Forcing both into a single analytic procedure merely to demonstrate that the study is “mixed” could weaken rather than strengthen the research.

Recent methodological work on mixed methods analysis explicitly describes separate quantitative and qualitative analysis as part of the process before researchers identify points of integration and draw insights from their combination. In other words, separate analysis can be a stage in mixed methods research rather than evidence against it.

The problem begins when separate analyses remain separate

Suppose a paper reports a survey showing that 68% of participants prefer a particular learning approach. It then presents interviews identifying four themes about students' learning experiences. The discussion summarizes the survey result, summarizes the themes, and ends.

What relationship exists between those findings?

If the study never answers that question, the mixed methods contribution is unclear. The reader knows what the quantitative component found and what the qualitative component found, but not what becomes understandable by considering them together.

This concern is not merely theoretical. O'Cathain, Murphy, and Nicholl evaluated mixed methods studies in health services research and found that researchers often described the separate components while giving insufficient attention to the mixed methods design. In many completed studies, integration was difficult to judge because data and findings from the different components had not been integrated.

Integration can occur even when the results appear in separate sections

Do not equate manuscript organization with research design. A paper may have a section titled “Quantitative Results” followed by “Qualitative Results” and still contain substantial integration elsewhere.

Fetters, Curry, and Creswell describe integration at several levels. At the methods level, researchers may connect components through sampling, build one component from results of another, merge datasets or results for analysis, or embed one form of data within another design. At the interpretation and reporting level, findings can also be brought together narratively or through joint displays.

This means that evaluating whether the study actually integrated its quantitative and qualitative components requires looking beyond the headings in the results section.

A sequential design may require separation before integration

Consider an explanatory sequential design. Researchers first collect and analyze quantitative data. Those findings then inform a qualitative phase intended to investigate selected results more deeply.

The analyses are necessarily separated in time. Yet the design can be strongly integrated if quantitative results determine whom researchers interview, what they ask, which unexpected findings they investigate, and how the eventual interpretation is constructed. Fetters and colleagues describe this type of relationship through mechanisms such as connecting and building.

Similarly, in an exploratory sequential design, qualitative analysis may occur first and subsequently shape a quantitative instrument, variables, hypotheses, or sampling strategy. Again, separate analyses do not imply independent components.

Parallel reporting is weaker evidence than an explicit analytic relationship

Sometimes authors alternate between numbers and quotations without actually integrating them. A statistical result followed by a thematically similar quotation may look integrated because the evidence appears together on the page.

But proximity is not analysis.

Ask what relationship the researchers claim between the findings. Does the qualitative evidence explain variation in a quantitative pattern? Does one component qualify the apparent generality of the other? Does quantitative evidence show how widespread a qualitatively identified phenomenon may be within the studied sample? Does one method reveal an apparent contradiction requiring further interpretation?

A joint display can make such relationships visible by organizing quantitative and qualitative findings together. It is useful when it facilitates comparison and the development of integrated interpretations, rather than functioning as a decorative table.

The mixed methods research question should give the components a reason to meet

A useful way to evaluate the design is to work backward from the research problem. Why were two methodological approaches needed?

Methodological discussions of integration increasingly emphasize that the mixed methods purpose or question should anticipate how the strands will be combined. A study may have separate quantitative and qualitative questions, but a mixed methods question can identify what researchers expect to learn through their relationship.

This also means that each strand should do real intellectual work. Before worrying about integration, ask whether each component addresses a meaningful part of the research question. Two weakly relevant components do not become compelling simply because they are combined.

The strongest evidence may appear in the final inference

One practical test is to examine the study's conclusion. Can you identify a substantive inference that depends on evidence from both components?

Mixed methods scholarship sometimes refers to conclusions developed from integrated evidence as meta-inferences. O'Cathain and colleagues similarly describe the potential “yield” of mixed methods research as knowledge that may not be available from qualitative and quantitative studies undertaken independently.

If the conclusion consists only of “the survey found X, while the interviews found Y,” integration may remain thin. If the researchers establish what X and Y mean when interpreted together, the mixed methods contribution becomes much clearer.

Two studies can contain the same analyses but differ dramatically in integration

Common mistakes when deciding whether separate analyses count as mixed methods

Judge the relationship, not the number of analyses

When reading a purported mixed methods study, do not stop after identifying one quantitative and one qualitative analysis. Trace what happens before, between, and after them.

Ultimately, the question is not whether researchers performed two analyses. It is whether the study gains something from their relationship. Sometimes one method may explain, expand, or contradict the other. That interaction is where much of the potential value of mixed methods research lies.

Check whether separate analyses form a coherent mixed methods study

Questions about separate analyses in mixed methods research

Separate analyses are acceptable; permanently separate reasoning is the problem

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.