Can One Study Have More Than One Research Design?

Can a Single Study Legitimately Contain Two or More Research Designs?

Your project starts simply. Then the research question becomes more ambitious.

Perhaps you want to estimate whether an educational intervention improves learning outcomes and conduct a qualitative case study to understand how implementation differs across classrooms. Perhaps you begin with a cross-sectional survey and later follow the same participants longitudinally. Or you embed a qualitative process evaluation within a randomized trial.

At that point, what is the research design?

Do you have to choose one label? Can you say the study has two designs? Does combining designs automatically make the project mixed methods? Or have you inadvertently created two separate studies?

There is no universal rule that every study must be represented by exactly one design label. Complex investigations can contain distinct design components. The more important question is whether those components belong together as one coherent investigation.

Yes, but Multiple Components Do Not Automatically Mean Multiple Designs

Start by Asking What You Mean by “More Than One Design”

Research design can be described at several levels

Part of the confusion comes from the flexibility of the term research design.

A researcher might describe a project broadly as mixed methods, then identify an explanatory sequential structure, then describe the quantitative component as quasi-experimental and the qualitative component as a case study. Another researcher might call the entire project a mixed methods design containing quasi-experimental and qualitative components.

Those descriptions are not necessarily contradictory. They operate at different levels.

The methodological literature does not impose one universal taxonomy across every discipline. Complex designs are often nested: an overarching strategy can contain components with their own sampling, timing, comparison, data-generation, and analytic logic.

This is why how specifically you name a research design matters less than accurately communicating how the study is structured.

Multiple research questions do not automatically mean multiple designs

A study can have several research questions that are all answered within one design.

Imagine a cross-sectional survey investigating university students' generative AI use. Researchers ask how frequently students use AI, whether usage differs across disciplines, whether attitudes toward AI are associated with frequency of use, and which measured characteristics predict reported use.

Those questions may require different analyses, but they can all operate within the same cross-sectional structure using the same population, sampling strategy, measurement occasion, and dataset.

More questions do not necessarily create more designs.

Multiple methods do not automatically mean multiple designs

The same is true of methods.

A qualitative case study might use interviews, observations, documents, and institutional records. Those are several methods within one case-study design.

A randomized experiment might use achievement tests, questionnaires, behavioral logs, and administrative records. Again, multiple methods do not imply multiple research designs.

This follows from the distinction between research design and research methods. Methods are procedures for generating and analyzing evidence; design concerns the structure and logic through which that evidence answers the research question.

Multiple analyses do not mean multiple designs either

Researchers may conduct descriptive statistics, regression models, subgroup analyses, sensitivity analyses, qualitative coding, and other forms of analysis within one design.

Analysis does not independently determine design.

A dataset does not become six research designs because the researcher opens six statistical menus. The relevant question is how the evidence was generated and structured, not how many analytic procedures were subsequently applied to it.

Multiple phases can still belong to one overarching design

Some research is deliberately phased.

Mixed methods research provides clear examples. An explanatory sequential design commonly begins with a quantitative component and follows it with qualitative inquiry intended to explain or elaborate the quantitative results. An exploratory sequential design reverses that broad sequence, using initial qualitative work to inform a subsequent quantitative component.

Mixed methods design guidance treats decisions about sequencing, priority, sampling relationships, implementation settings, and integration as parts of constructing the overall study.

You could therefore describe such a project as one mixed methods design with distinct quantitative and qualitative components. Depending on the components, each may also have a recognizable design of its own.

When does it make sense to say that a study contains multiple designs?

The case becomes stronger when components have meaningfully different structural logics.

For example:

If two or more such components are intentionally incorporated into one project, describing their component designs separately may be informative.

The important question is what each design contributes and how the components relate.

Mixed methods is one important case, but not the only one

Multiple-design research is sometimes assumed to mean mixed methods research. That is too narrow.

Mixed methods specifically involves quantitative and qualitative components and, importantly, their intentional integration. A project containing two quantitative design components is not mixed methods merely because two designs are present.

For example, a research program could embed a randomized experiment within a larger longitudinal quantitative follow-up. Another project could combine a cross-sectional baseline study with a prospective cohort component.

Conversely, using quantitative and qualitative methods does not automatically produce a strong mixed methods design. The components need a purposeful relationship, and integration should contribute to answering the research problem. Mixed methods scholarship emphasizes integration rather than the mere presence of two forms of data.

The broader choice among quantitative, qualitative, and mixed methods approaches should therefore be distinguished from the number of component designs within a project.

Nested designs are common in complex research

Large or complex studies often contain designs within designs.

A randomized trial may include a longitudinal follow-up, an implementation study, a process evaluation, or a qualitative component. A cohort study may contain a nested case-control study. A mixed methods evaluation may contain a quasi-experimental outcome component and qualitative case studies of implementation.

Researchers may describe these structures using terms such as embedded, nested, multiphase, or hybrid, depending on the methodological tradition and the exact relationship among components.

The label should communicate structure rather than merely make the study sound elaborate.

Different purposes do not necessarily require different designs

Suppose a longitudinal cohort study has both descriptive and explanatory aims. Researchers describe how student engagement changes over four semesters and then investigate which measured factors are associated with those trajectories.

There are two research purposes, but not necessarily two research designs.

This is why a study can be both descriptive and explanatory without automatically becoming a multiple-design study.

Purpose and design answer different questions. Adding a purpose does not mechanically add a design.

The components need a reason to belong together

The central issue in a complex study is coherence.

Suppose researchers conduct a randomized trial of an online tutoring intervention and interviews with participating students. The interviews could be genuinely integrated with the trial if they investigate mechanisms, implementation, acceptability, unexpected outcomes, or differences observed in the quantitative findings.

But imagine the interviews instead investigate students' unrelated career aspirations simply because the participants are already available.

The same people and project budget do not make those questions one coherent study.

A useful test is whether the components need one another to answer a common higher-order research problem. A complex project can remain one study despite several questions, methods, phases, populations, or datasets when those components are conceptually and methodologically integrated. When clusters of questions can be conducted and interpreted largely independently, the project begins to resemble several studies sharing a topic.

One topic is not enough to create one study

This distinction matters because a broad research topic can support many separate investigations.

“Generative AI in higher education,” for example, could support studies of prevalence, learning outcomes, academic integrity, faculty practices, institutional policy, accessibility, assessment, student experience, and implementation.

Those studies can belong to the same research program without being forced into one design. A research topic defines an area of inquiry rather than one predetermined study.

Adding another design should therefore be justified by the needs of the current study, not merely by the fact that another interesting question can be asked.

More designs can increase inferential reach, but also complexity

Combining complementary designs can be powerful.

An experiment may provide strong evidence about whether an intervention produces an outcome under specified conditions. A qualitative component may illuminate how participants experience the intervention or why implementation varies. Longitudinal follow-up may reveal whether effects persist.

Together, those components may answer a broader question than any one could answer alone.

But every additional component creates methodological obligations. Researchers may need additional sampling strategies, expertise, data-management procedures, ethical considerations, analytic plans, integration strategies, reporting space, and time.

More designs do not automatically mean a stronger study

Complexity is not a methodological virtue in itself.

A single well-chosen design may answer a focused question more convincingly than a project containing several underdeveloped components. Additional designs are justified when they solve a real evidentiary problem.

Ask what becomes possible after adding the second design. Does it answer another necessary part of the overarching question? Does it explain a result the first component cannot explain? Does it address implementation or context essential to interpretation? Does it extend observation over a time scale required by the question?

If the answer is merely “it makes the study more comprehensive,” the rationale needs more work.

One study or several studies?

There is no numerical threshold.

Two designs can belong comfortably within one study. A project with one nominal design can still contain questions so disconnected that it functions like several investigations.

Consider the degree of dependence among components.

That boundary can involve judgment. Research programs, multiphase projects, trials with embedded studies, and dissertations can all organize related investigations differently. What matters is that the reporting structure does not conceal conceptual independence or falsely imply integration.

Name the structure in the way that best helps the reader understand it

You do not need to force every complex study into one enormous compound label.

Often the clearest approach is to identify the overarching design or strategy and then describe each component explicitly.

For example, you might explain that a mixed methods evaluation contains a quasi-experimental quantitative component and a qualitative multiple-case component, followed by a description of how the findings are integrated.

Alternatively, if one design clearly dominates and another is embedded for a limited purpose, you might name the primary design and describe the embedded component separately.

The principle behind naming a research design with appropriate specificity applies especially strongly here: use terminology to reduce ambiguity, not to display the maximum number of methodological adjectives available.

One Study With an Experiment and a Qualitative Case Component

Common Misunderstandings About Multiple Research Designs

Add Another Design Only When the Research Problem Requires It

Before adding a second design, identify the gap left by the first.

What important question remains unanswered? Why can the existing design not answer it? What distinct evidence would the additional design provide? How will that evidence relate to the first component?

If those questions have clear answers, a multiple-component or multiple-design study may be justified.

Complexity should follow necessity. This is another reason the strongest feasible design may be preferable to an idealized but unmanageable one.

Before Combining Multiple Research Designs

Frequently Asked Questions About Multiple Research Designs

One Study Can Contain Several Designs When the Components Form One Coherent Investigation

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.