Sequential vs. Concurrent Mixed Methods: Which Should Come First and Why?

Should the Quantitative or Qualitative Component Come First, or Can You Conduct Both Together?

You have established that your research problem genuinely requires both quantitative and qualitative evidence. The next design decision concerns timing.

Should you collect survey data first and use the results to shape interviews? Should qualitative exploration come first so that you can develop variables or an instrument? Or can both components proceed during roughly the same phase and be integrated afterward?

The answer should not depend on which method you prefer or which dataset happens to become available first. Timing in mixed methods research has a methodological function. One component should precede another when the later component needs something from the earlier one. When neither component depends on prior results from the other, concurrent collection may be more appropriate.

Sequence the Components When One Needs to Inform the Other

Timing Should Follow the Relationship Between the Components

Sequential Mixed Methods Creates a Deliberate Dependency

In sequential mixed methods research, the components occur in an intentional order. The defining feature is not merely that one dataset happens to be collected earlier. Findings or decisions from the first component contribute methodologically to what happens in the second.

The first component might identify participants for follow-up, expose findings that require explanation, generate concepts for measurement, help construct an instrument, or refine questions for subsequent investigation.

This creates a dependency: the second component cannot be fully designed or implemented until relevant information from the first becomes available.

Two common patterns are explanatory sequential and exploratory sequential designs. Their names sound similar, but the logic runs in different directions.

Explanatory Sequential Design Usually Begins Quantitatively

An explanatory sequential design commonly begins with quantitative data collection and analysis. Qualitative inquiry follows because some aspect of the quantitative findings requires explanation, elaboration, or contextualization.

Suppose a survey of university instructors finds that institutional AI training is positively associated with reported adoption of generative AI, but the relationship is strong among early-career academics and nearly absent among senior academics. Rather than designing generic interviews before seeing the survey results, researchers can use that unexpected pattern to determine whom to interview and what needs explanation.

The qualitative component is not simply an appendix to the survey. Its questions and sampling should respond to the quantitative findings. That connection is what gives the sequence methodological purpose.

Exploratory Sequential Design Usually Begins Qualitatively

An exploratory sequential design reverses the logic. Researchers begin qualitatively because the phenomenon is insufficiently understood, existing constructs may not fit the context, relevant variables are uncertain, or an appropriate measurement instrument is unavailable.

The qualitative findings then inform a quantitative phase. Researchers may use concepts developed from interviews, observations, or other qualitative evidence to construct items, define dimensions, generate hypotheses, identify variables, or develop an intervention or instrument that is subsequently examined quantitatively.

Qualitative-first sequencing therefore makes sense when the quantitative component would be premature without initial exploration.

Concurrent or Convergent Designs Do Not Require One Component to Wait

In a concurrent arrangement, qualitative and quantitative components are conducted during approximately the same phase of the research rather than one being designed from the completed findings of the other.

A convergent mixed methods design is a common example. Researchers collect and analyze qualitative and quantitative evidence separately and then bring the results together to examine how they relate.

For instance, a study of faculty experiences with a new learning-management system could administer a structured survey while also conducting interviews during the same semester. If neither component requires the other's results before data collection, simultaneous implementation may be feasible. Researchers can later compare what the survey suggests with what participants describe qualitatively.

Concurrent does not mean disconnected. The integration may occur after separate analyses, but the components should still be designed around a common research problem and a clear plan for bringing the findings together.

The Central Difference Is Dependency, Not the Calendar

A project does not become sequential mixed methods simply because interviews happened in March and a survey happened in June. If the March interviews had no methodological influence on the June survey, chronological order alone provides weak justification for calling the design sequential.

Which Comes First Depends on What Is Missing

A useful way to choose the order is to ask what you do not yet know.

If you already have quantitative measures but do not understand a statistical pattern, quantitative-first sequencing may make sense. If you do not yet know which constructs, dimensions, or language adequately represent the phenomenon, qualitative-first sequencing may be more appropriate.

The uppercase and lowercase notation is sometimes used in mixed methods literature to indicate relative priority, while arrows indicate sequence. Notation conventions vary, however, so researchers should explain their design in words rather than assuming symbols alone communicate the methodological logic.

Integration Between Sequential Phases Must Be Visible

A common weakness in sequential studies occurs at the transition between components. Researchers complete a survey, announce that interviews followed, and never explain how the quantitative findings shaped the qualitative component.

The connection should be explicit. Did the first phase determine who was sampled? Which results were followed up? How were interview questions constructed from those results? Were unusual cases deliberately selected? Did the qualitative phase investigate a particular statistical relationship?

Likewise, in qualitative-first research, researchers should explain exactly how qualitative findings informed instrument development, variables, hypotheses, sampling, or other features of the quantitative phase.

This is part of the broader requirement that qualitative and quantitative evidence must actually be integrated rather than merely coexist in the same project.

Concurrent Designs Still Need a Plan for Bringing Findings Together

Concurrent collection solves the sequencing question, but it does not solve integration automatically.

After separate analyses, researchers may compare findings through narrative integration, matrices, joint displays, data transformation, or other appropriate strategies. They may examine whether results converge, complement one another, reveal different dimensions, or appear inconsistent.

A joint display can be particularly useful when it places related quantitative and qualitative findings in the same analytical structure and supports integrated conclusions. Simply placing a statistical table next to several quotations is not necessarily integration if the relationship between them remains unexplored.

Divergent Findings Are Methodologically Useful

Concurrent designs are sometimes justified as a way to “validate” one dataset with another. Corroboration may be useful, but treating agreement as the only desirable outcome impoverishes the design.

Suppose survey respondents report high confidence in using an educational technology, while interviews reveal frequent reliance on colleagues to complete basic tasks. The discrepancy may reflect social desirability, different interpretations of confidence, limitations in the survey measure, contextual variation, or a distinction between perceived and demonstrated competence.

The disagreement is not necessarily something to eliminate. It may be the most interesting integrated finding in the study.

Sequential Designs Often Take Longer, but Concurrent Designs Can Be Harder to Coordinate

Sequential research may require considerable elapsed time because researchers must collect and analyze one component before fully developing the next. Recruitment, ethics amendments, instrument development, and follow-up can extend the project further.

Concurrent research can shorten the overall data-collection period, but it creates other demands. Teams may need expertise in both approaches at the same time, sampling decisions must often be made before either dataset has been analyzed, and researchers need a credible plan for integrating findings generated on parallel tracks.

Practical convenience matters, but it should not reverse the logic of the research question. Running components concurrently simply to finish sooner is a poor trade if the second component genuinely needs information from the first.

Timing Is One Design Dimension, Not the Entire Mixed Methods Design

Sequential versus concurrent tells readers when components occur relative to one another. It does not fully specify priority, integration strategy, sampling relationship, theoretical framework, or the precise function each component serves.

A complete mixed methods design therefore requires more than a timing label. Researchers should describe why both components are needed, how they relate, when they occur, where integration happens, and what combined inference the design is intended to support.

The Same Research Problem Can Require Different Timing

There is no universally correct order for the topic. The correct sequence depends on the methodological dependency. Asking “Should qualitative or quantitative come first?” without specifying what each component needs to accomplish is therefore rather like asking which chapter of a dissertation should be written first. Academics have opinions, naturally, but the useful answer depends on the work that must happen.

Common Mistakes When Sequencing Mixed Methods Research

Ask What One Component Needs From the Other

The simplest way to choose the timing is to identify the dependency between components before deciding on the calendar.

Then make the integration point visible in the protocol. In sequential research, specify what information moves from the first phase to the second. In concurrent research, specify how independently generated findings will eventually be brought together.

Do not let project management quietly become methodology. If sequential logic is necessary, allow enough time for analysis between phases. If concurrent collection is appropriate, ensure the team can conduct both components rigorously at the same time.

When timing alternatives could both answer the question, consider what the timing of each measurement allows the study to establish and whether the additional demands of a particular sequence materially improve the evidence.

Before Choosing Sequential or Concurrent Mixed Methods

Questions About Sequential and Concurrent Mixed Methods

Let Methodological Dependency Determine the Order

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.