Quantitative vs. Qualitative vs. Mixed Methods: Which Approach Should You Use?

Should Your Study Be Quantitative, Qualitative, or Mixed Methods?

You have a research problem and perhaps an emerging research question. At some point, someone will probably ask: Is your study quantitative or qualitative?

Then comes the increasingly common third option: Why not mixed methods?

The question can sound like a choice among competing camps. Quantitative research is sometimes portrayed as objective and generalizable, qualitative research as deep and subjective, and mixed methods as the comprehensive option that gives you “the best of both worlds.” Those descriptions are too crude to guide a serious methodological decision.

Each approach can produce rigorous and useful research when it fits the question and is implemented well. Each can also be badly mismatched to a question.

The more productive starting point is therefore not “Which approach is better?” but “What kind of evidence do I need to answer this research question?”

Choose the Approach That Produces the Evidence Your Question Needs

What Quantitative, Qualitative, and Mixed Methods Research Actually Offer

Quantitative research works with numerical evidence

Quantitative research systematically represents relevant phenomena using numerical data and applies mathematical or statistical procedures to describe patterns, estimate quantities, examine relationships, compare groups, predict outcomes, or evaluate effects.

It is particularly useful when the question requires answers such as:

• How common is this phenomenon in a defined population?
• How much does an outcome differ between groups?
• Are two measured variables associated?
• How accurately can an outcome be predicted?
• Did an intervention change a measured outcome compared with an appropriate alternative?

Quantitative research often requires researchers to define constructs in measurable terms. “Academic engagement,” for example, must become something observable through indicators, scales, behavioral records, or other measures if it is to enter a quantitative analysis.

This creates considerable analytic power, but it also means that measurement choices matter. A precise statistical estimate of a poorly operationalized construct remains a precise estimate of something that may not adequately represent what the researcher intended to study.

Qualitative research works with meaning, experience, process, and context

Qualitative research is particularly useful when the question requires detailed understanding of how people experience, interpret, construct, negotiate, or respond to a phenomenon, or when processes and contexts cannot be adequately represented through predefined numerical variables alone.

Evidence may come from interviews, focus groups, observations, documents, images, field materials, naturally occurring interactions, or other sources appropriate to the research tradition.

Qualitative inquiry can be especially valuable when researchers need to understand:

• how participants interpret an experience;
• why a process unfolds differently across contexts;
• how practices, identities, or meanings are constructed;
• how an unfamiliar or insufficiently understood phenomenon operates;
• what mechanisms, perceptions, or contextual conditions might explain an observed pattern.

Qualitative research is not simply quantitative research with fewer participants and no statistics. Its sampling logic, evidence, analytic reasoning, and standards for evaluating claims may differ substantially from those used in quantitative research.

Mixed methods intentionally integrates quantitative and qualitative research

Mixed methods research involves more than placing a questionnaire and several interviews in the same project.

A central feature is integration: the quantitative and qualitative components are intentionally connected, combined, compared, embedded, or otherwise brought into relation so that together they answer the research question more effectively. Methodological literature consistently identifies this integration as a defining characteristic of mixed methods research.

The components can relate in different ways. Quantitative results might identify an unexpected pattern that subsequent interviews investigate. Qualitative findings might be used to develop a survey instrument that is then examined in a larger population. Quantitative and qualitative evidence might be collected concurrently and brought together during interpretation. One form of evidence might also be embedded within a larger intervention or evaluation design.

The key question is not whether both forms of data appear somewhere in the project. It is what is gained by integrating them.

The three approaches answer different versions of a problem

Consider research on doctoral students' use of generative AI.

The approaches are not competing to answer exactly the same question in exactly the same way. Changing the form of evidence often changes the answer you can produce.

Quantitative does not mean objective and qualitative does not mean subjective

The familiar objective-versus-subjective contrast is too simplistic.

Quantitative research involves human decisions at every stage: defining constructs, choosing measures, determining sampling procedures, deciding how missing data will be handled, selecting models, setting analytic thresholds, and interpreting results. Numerical data do not remove judgment from research.

Qualitative research, meanwhile, does not mean that researchers may simply report whatever interpretation they prefer. Rigorous qualitative inquiry uses systematic procedures appropriate to its methodological tradition and may incorporate reflexivity, transparent documentation, careful sampling or case selection, attention to discrepant evidence, and other strategies for strengthening the credibility of interpretation.

Different approaches manage researcher judgment, uncertainty, and evidence differently. Neither escapes them.

Quantitative does not automatically mean generalizable

Numerical evidence is not automatically representative of a wider population.

A quantitative survey distributed to 100 volunteers may provide precise numerical descriptions of those respondents while offering a weak basis for estimating characteristics of all university students. Generalizability depends on how the sample relates to the target population, how participation occurs, how measurements perform, and what inferential assumptions are justified.

Similarly, qualitative research is not simply “non-generalizable.” Qualitative studies may seek forms of theoretical, conceptual, analytic, or contextual transfer that differ from statistical generalization to a population. The intended scope of claims should be evaluated according to the purpose and logic of the study.

Qualitative does not mean exploratory by definition

Qualitative methods are often associated with exploratory research because they can be powerful when researchers need to understand a phenomenon before imposing predetermined categories.

But qualitative research can also be descriptive, explanatory, interpretive, evaluative, or theory-oriented. Quantitative research can likewise serve exploratory purposes, such as examining an unfamiliar dataset or identifying patterns that generate hypotheses.

The distinction among exploratory, descriptive, explanatory, and evaluative purposes therefore should not be mechanically mapped onto quantitative or qualitative approaches.

Mixed methods is not simply “more complete” research

Mixed methods is attractive because many important research problems have both measurable and contextual dimensions. Yet using two approaches does not automatically create a stronger study.

Mixed methods introduces additional design decisions: Why are the components being combined? Which component has priority? Will they occur concurrently or sequentially? Does one phase depend on the results of another? At what point will integration occur? What will researchers do if the quantitative and qualitative findings appear to diverge?

Common mixed methods structures include convergent designs, explanatory sequential designs, and exploratory sequential designs, although the methodological literature contains additional and more complex typologies.

The labels are useful, but integration is more important than memorizing the typology. Mixed methods guidance emphasizes that researchers should begin with the research question and a clear purpose for combining the components.

Mixed methods is different from merely using multiple methods

A study can use several methods without being mixed methods research.

For example, a qualitative study might use interviews, observations, and documents. That is multiple-method qualitative research, but it does not become mixed methods merely because more than one technique is present.

Likewise, a quantitative study might combine questionnaires, administrative records, physiological measures, and standardized assessments without containing a qualitative component.

Definitions vary somewhat across the literature, but integration is widely treated as central to mixed methods rather than the mere coexistence of different data types.

Choose the approach after clarifying the question

Starting with “I want to conduct quantitative research” reverses the preferred logic of design unless there is a defensible reason for that constraint.

Instead, clarify what the research question requires. If you need an estimate, comparison, measured relationship, prediction, or intervention effect, quantitative evidence may be central. If you need an in-depth account of experience, meaning, process, or context, qualitative evidence may be more appropriate. If answering the question requires both and there is a clear reason to integrate them, mixed methods may be justified.

This is part of the broader process of choosing a research design from the question rather than from a preferred technique.

Feasibility can legitimately affect the choice

A mixed methods study may appear ideal on paper but require expertise in two methodological traditions, additional participant contact, multiple forms of analysis, integration across components, and substantially more time.

Similarly, a large quantitative study may require access to a sample that a researcher cannot realistically obtain. An intensive qualitative design may require more sustained field engagement than a project timeline permits.

Feasibility does not mean choosing the easiest approach. It means identifying the strongest approach that can answer the question credibly under actual ethical and practical conditions.

If the approach needed to answer the original question cannot realistically be implemented, the more defensible response may be to narrow or revise the question rather than retain the question and collect evidence incapable of answering it.

How the Same Topic Produces Three Different Studies

Common Misconceptions About the Three Approaches

Choose Based on the Evidence You Need, Not the Label You Prefer

Before deciding that your study is quantitative, qualitative, or mixed methods, write down the answer you hope the research will eventually provide.

If that answer requires a numerical estimate, measured comparison, prediction, or estimate of an effect, you probably need substantial quantitative evidence. If it requires understanding how participants interpret an experience, how a process unfolds, or why context matters, qualitative evidence may be central.

If the answer genuinely requires both, ask one more question: What will I learn by integrating the two forms of evidence that I could not learn adequately from either alone?

If you cannot answer that question, mixed methods may not yet be justified.

Once the broad approach is clear, you still need to determine which specific research design is appropriate for the question. “Quantitative,” “qualitative,” and “mixed methods” do not by themselves specify every structural decision in the study.

Before Choosing Quantitative, Qualitative, or Mixed Methods

Frequently Asked Questions About Quantitative, Qualitative, and Mixed Methods Research

There Is No Best Approach Independent of the Question

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.