Can Research Be Scientific Without Being Quantitative?

Does Scientific Research Need Numbers and Statistics?

Numbers carry considerable authority in research. Means, percentages, confidence intervals, regression coefficients, and other statistical results can make a study look unmistakably scientific. By contrast, research based on interviews, observations, documents, field notes, or other predominantly nonnumerical evidence may appear less scientific to someone who equates science with measurement and statistical analysis.

That equation is too narrow.

Some research questions are fundamentally about magnitude, frequency, distribution, difference, or numerical association. Quantitative methods are well suited to many such questions. Others concern meanings, experiences, practices, processes, contexts, interactions, or how a phenomenon is understood. For these questions, converting everything into numbers may not produce better evidence. It may instead remove precisely the information the researcher needs to understand.

Scientific Inquiry Does Not Require Every Study to Be Quantitative

Why Scientific Research Cannot Be Reduced to Quantification

Scientific inquiry is defined by more than its data format

A number is not inherently scientific, just as a transcript is not inherently unscientific. Both are forms of evidence whose usefulness depends on how they were produced, what they represent, how they were analyzed, and what conclusions researchers draw from them.

The OECD's internationally used framework for research and experimental development defines R&D broadly as creative and systematic work undertaken to increase the stock of knowledge and devise new applications of available knowledge. Its framework explicitly includes knowledge of humankind, culture, and society. It does not define research by a requirement that evidence be numerical.

Likewise, the National Research Council's Scientific Research in Education argues that rigorous quantitative and qualitative research share an underlying logic of inference, while recognizing that the chain of reasoning varies according to the research question and design. The implication is important: scientific rigor can be expressed through different methodological practices rather than through one mandatory data type.

Qualitative and quantitative research often ask different kinds of questions

These questions are not interchangeable. Counting how many faculty members use an AI tool cannot, by itself, reveal how they understand its pedagogical value, negotiate institutional expectations, or decide when its use is appropriate. Interviews or observations cannot, by themselves, provide a population prevalence estimate.

The method should therefore follow the question. This is consistent with the broader principle that there is no single scientific method that every research project must follow.

Qualitative research is more than collecting people's opinions

Qualitative research may use interviews, focus groups, participant or nonparticipant observation, documents, archival materials, field notes, and other sources. The resulting evidence is often textual, visual, or otherwise nonnumerical, although qualitative projects may also incorporate numerical information.

What makes the inquiry research is not simply the presence of interviews or observations. Researchers need a defensible strategy for selecting participants, cases, documents, or settings; procedures appropriate to the phenomenon; a systematic approach to analysis; and a transparent connection between evidence and interpretation.

Methodological literature in the health sciences, for example, identifies qualitative research as particularly useful for investigating the nature of phenomena, understanding why something is or is not observed, examining complex interventions, and investigating aspects of experience and context that quantitative measures may not capture adequately.

Qualitative rigor does not need to imitate quantitative rigor

A common mistake is to judge qualitative research entirely through criteria developed for quantitative designs. Some concerns are shared across methodologies, including transparency, appropriate evidence, logical reasoning, and attention to alternative interpretations. Their operational expression, however, may differ substantially.

Depending on the qualitative methodology, researchers may need to justify sampling or case selection, document analytical decisions, consider reflexivity, examine discrepant or negative cases, provide sufficient contextual information, demonstrate how interpretations were developed from the data, and make clear the boundaries of their claims.

These practices should not be treated as a universal checklist for every qualitative study. Ethnography, grounded theory, phenomenological approaches, qualitative case studies, narrative inquiry, and other traditions have different purposes and methodological commitments. Rigor needs to be evaluated in relation to the approach actually being used.

Quantification does not automatically make research more scientific

The reverse misconception is equally important. A study does not become rigorous merely because researchers assign numerical codes to responses and run statistical tests.

Numbers can be generated from poorly defined constructs, biased samples, unreliable measurements, inappropriate statistical models, or data that cannot support the conclusions being drawn. A precisely calculated estimate of the wrong quantity remains the wrong quantity.

This is why research does not necessarily require statistics and can be entirely qualitative. Statistical analysis is a tool for answering certain questions, not a certification mechanism for scientific legitimacy.

Qualitative research can be empirical

Qualitative research is sometimes incorrectly contrasted with empirical research. Yet interviews, observations, documents, recordings, artifacts, and fieldwork can all provide evidence grounded in observation or experience.

A qualitative study may therefore be empirical even though its evidence is not primarily numerical. Whether a particular form of inquiry should be considered empirical depends on what evidence it uses and how the term is understood in the relevant discipline. The distinction is explored more fully in the question of what makes research empirical.

Qualitative and quantitative approaches can also be combined

Researchers do not always need to choose one approach to the exclusion of the other. When the research question genuinely requires both forms of evidence, qualitative and quantitative methods may be integrated within a mixed-methods design.

For example, a researcher might quantify patterns of student withdrawal from an online course and then investigate how students experienced the circumstances surrounding withdrawal. The numerical evidence can establish patterns, while qualitative evidence can illuminate processes and meanings that the numerical pattern alone does not explain.

Using both approaches is not automatically superior. Combining methods adds methodological demands and should be justified by what the research question requires, not by the assumption that more methods necessarily mean better research.

Investigating Why Faculty Members Hesitate to Use Generative AI

The study does not become less scientific because its central findings cannot be reduced meaningfully to a mean score. Its purpose is to understand reasoning, meaning, and context rather than estimate how frequently each position occurs in an entire population.

At the same time, the researchers should not make claims their design cannot support. An in-depth qualitative sample cannot simply be treated as though it were a probability sample producing population prevalence estimates. Scientific rigor includes knowing what your evidence cannot tell you.

Common Misconceptions About Qualitative and Scientific Research

Choose the Form of Evidence Your Question Actually Requires

If you are deciding between qualitative and quantitative approaches, do not begin by asking which one looks more scientific. Begin with what you need to know.

A quantitative design may be appropriate when the question concerns magnitude, frequency, numerical differences, distributions, prediction, or relationships that can be meaningfully represented quantitatively. A qualitative design may be appropriate when the question requires close investigation of meaning, experience, process, practice, interpretation, or context.

The standard should be methodological fit and defensible inference. Research may be scientific without being experimental, and it may likewise be scientific without being quantitative. Neither experiments nor numbers should function as decorative badges of scientific status.

Before Choosing Qualitative or Quantitative Research

Questions About Qualitative Research and Scientific Inquiry

Numbers Are One Form of Scientific Evidence, Not the Definition of Science

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.