03 · What You Need to Know
When Does Qualitative Understanding Need Quantitative Evidence?
Rich Understanding Does Not Automatically Tell Us How Common Something Is
Qualitative research can provide detailed understanding of how people experience a phenomenon, how they interpret events, which processes shape their behavior, and how context influences what happens.
Those strengths do not automatically provide population prevalence estimates.
Suppose interviews with university students identify several forms of anxiety about generative AI: concern about inaccurate information, academic-integrity accusations, privacy, dependency, and uncertainty about institutional rules. The study may provide compelling evidence that these concerns exist and illuminate how students experience them.
It does not necessarily tell us what percentage of all students experience each concern.
If that prevalence matters to theory, institutional planning, policy, intervention design, or another consequential decision, the absence of quantitative estimation may constitute a gap.
A Quantitative Gap Is About Missing Knowledge, Not a Missing Method
The same principle that applies to qualitative gaps applies in reverse.
Weak justification
Previous studies are qualitative, so quantitative research is needed.
Stronger justification
Existing qualitative studies identify recurring barriers to technology adoption, but their prevalence and relative distribution across the target student population remain unknown.
The second statement explains what researchers still need to know and why another type of evidence is useful.
A methodological imbalance may alert you to a possible gap. It does not become a substantive research gap until you identify the unresolved question.
Quantitative Research Can Estimate How a Phenomenon Is Distributed
Qualitative studies often use purposive sampling to obtain information-rich cases rather than samples designed for statistical population inference. This can be entirely appropriate for their aims.
When the research question shifts toward population distribution, a different sampling logic may be needed.
| What qualitative research may establish |
Potential quantitative follow-up question |
| Students describe several barriers to AI adoption |
How prevalent is each barrier in the target student population? |
| Teachers describe experiences of workload associated with a new platform |
How much additional workload is reported across the broader workforce? |
| Patients identify different reasons for discontinuing a service |
How frequently is each reason reported? |
| Employees describe variation in psychological safety |
How is psychological safety distributed across teams or organizational units? |
| Participants identify a possible process linking two phenomena |
Is the hypothesized relationship observable across a larger sample, and how strong is it? |
Quantification can therefore extend qualitative understanding without implying that the qualitative evidence was incomplete for its original purpose.
Quantitative Evidence Can Test Relationships Suggested by Qualitative Research
Qualitative inquiry may generate hypotheses about how constructs relate. Participants might consistently describe a process in which unclear institutional policies reduce trust, which in turn affects willingness to use a technology.
A subsequent quantitative study could operationalize those constructs and investigate whether the proposed relationships are supported across a larger sample.
That does not transform the qualitative finding into a hypothesis that must be "proved." Rather, qualitative evidence can inform a testable model whose statistical relationships, effect sizes, uncertainty, and boundary conditions require separate evidence.
Quantitative Research Can Test Whether Groups Differ
Qualitative studies may suggest that experiences differ among groups. Perhaps working students describe time constraints more frequently or differently than full-time students. A quantitative study can examine whether such differences are detectable and estimate their magnitude.
This is where a quantitative gap can overlap with a missing-subgroup gap.
However, group comparisons should be theoretically or practically justified. The fact that a dataset contains demographic categories does not mean every possible comparison deserves hypothesis testing.
Quantitative Evidence Can Estimate Magnitude, Not Merely Statistical Significance
A useful quantitative follow-up should not be reduced to asking whether a P value falls below a conventional threshold.
If qualitative research suggests that a particular factor matters, quantitative research can estimate the magnitude of its association or effect and the uncertainty surrounding that estimate. Confidence intervals and other appropriate measures of uncertainty can help researchers judge how compatible the data are with substantively different effect sizes.
This is often more informative than converting a rich qualitative insight into a binary hypothesis and asking whether it is "significant."
Qualitative Themes Are Not Automatically Variables Ready for a Survey
A common mistake is to take themes from interviews, convert each into a questionnaire item, and immediately begin statistical analysis.
Qualitative concepts may need substantial conceptual clarification before they can be operationalized quantitatively. Researchers should determine what the construct means, whether existing measures already capture it, how items represent its domain, and what evidence is needed to support score interpretation.
If existing instruments are inadequate, the problem may intersect with poor measurement. Developing a questionnaire is a measurement project, not merely a formatting exercise in which interview quotations acquire Likert scales.
Watch Out
Do not assume that frequency of a qualitative theme within an interview sample provides a population prevalence estimate. Qualitative sampling and analysis are generally designed for purposes other than statistical estimation unless the study explicitly uses a design that supports such inference.
Quantitative Research Can Evaluate an Intervention Suggested by Qualitative Findings
Qualitative research may identify a recurring problem and suggest a plausible intervention. For example, interviews may reveal that students avoid an online support service because its purpose is unclear and they fear being judged for using it.
Researchers might use those findings to redesign communication and service delivery. A subsequent quantitative evaluation could then estimate whether the intervention changes awareness, use, or another relevant outcome compared with an appropriate counterfactual or baseline.
The qualitative evidence contributes to intervention development. The quantitative evidence addresses whether and to what extent the intervention produces the intended outcome.
Quantitative Evidence Is Not Automatically More Generalizable
A survey of 2,000 convenience-sampled participants does not automatically provide stronger population inference than a carefully designed qualitative study simply because the sample is larger.
Generalizability depends on the target population, sampling process, participation, measurement, study design, and intended inference. Large samples can produce very precise estimates of a biased quantity.
If the quantitative gap concerns prevalence or population parameters, sampling design deserves particular attention. A study should be capable of producing the type of inference used to justify it.
Quantitative Research Is Not a “Next Stage” That Makes Qualitative Findings Scientific
It is misleading to frame qualitative research as preliminary exploration that becomes legitimate only after quantitative confirmation. Qualitative and quantitative methodologies can produce different forms of rigorous knowledge.
A qualitative study may answer its research question completely without any quantitative follow-up. Quantification becomes warranted when another important question arises that requires estimation, comparison, prediction, effect evaluation, or another form of numerical evidence.
This distinction prevents methodological hierarchy from masquerading as research-gap analysis.
Mixed Methods Can Connect Discovery With Estimation
Some research problems genuinely benefit from both forms of evidence. An exploratory sequential mixed-methods design, for example, may begin qualitatively to identify concepts or develop an understanding of the phenomenon, followed by a quantitative phase designed to examine their distribution or relationships more broadly.
Other mixed-methods designs may begin quantitatively and use qualitative inquiry to explain results, or collect both forms of evidence concurrently.
The NIH Office of Behavioral and Social Sciences Research emphasizes that mixed methods involves intentional integration of quantitative and qualitative approaches. The value lies in what the integration contributes to the research question, not simply in conducting two studies side by side.
A Quantitative Gap Can Exist Even When Many Qualitative Studies Agree
Consistency across qualitative studies can strengthen confidence that a phenomenon or interpretation recurs across studied contexts. It still does not automatically provide estimates of population prevalence, effect magnitude, predictive performance, or statistical relationships.
For example, 20 qualitative studies may repeatedly identify workload as a barrier to technology adoption. That is substantial evidence about the importance and nature of workload in participants' accounts. It does not tell us whether workload is reported by 15%, 50%, or 90% of the target population, nor how strongly it predicts adoption relative to other factors.
If those quantities matter, quantitative uncertainty remains.
How Do You Establish That Quantitative Evidence Is Actually Missing?
Do not assume a literature is qualitative merely because the most influential papers you encountered use interviews. Search for surveys, cohort studies, experiments, trials, administrative datasets, quantitative observational studies, measurement studies, and other relevant designs.
Systematic and scoping reviews can help reveal the methodological composition of a literature. Then determine which quantitative questions have already been addressed and whether the available evidence is sufficiently strong.
A defensible gap statement might say, "Qualitative studies consistently identify uncertainty about institutional AI policies as an important influence on students' decisions, but the prevalence of this concern and its association with AI use across the broader student population remain unclear."
That is more useful than claiming simply that few quantitative studies exist.
