Quantitative research measures phenomena numerically: how many, how often, how much. Qualitative research interprets meaning: why people behave as they do, what an experience means, how a process unfolds. The choice between them depends on your research question: if you need to know the scale of something, use quantitative methods; if you need to understand it in depth, use qualitative. Most real-world research questions benefit from both.
The core difference at a glance
| Dimension | Quantitative research | Qualitative research |
|---|---|---|
| Goal | Measure, test, generalise | Understand, interpret, explore |
| Data type | Numbers, ratings, counts | Text, observations, images, audio |
| Sample size | Large (100s to 1,000s+) | Small to medium (10-50 typical) |
| Data collection | Surveys, experiments, sensors | Interviews, focus groups, observation, documents |
| Analysis | Statistical (regression, t-tests, ANOVA) | Thematic, content analysis, discourse analysis |
| Output | Estimates, correlations, effect sizes | Themes, frameworks, narratives |
| Strengths | Generalisable, replicable, precise | Rich, contextual, discovers the unexpected |
| Limitations | Misses meaning and context; assumes you know what to measure | Limited generalisability; time-intensive |
| Typical time to complete | Days to weeks (survey data collection) | Weeks to months (interviewing and analysis) |
When is quantitative research the right choice?
Quantitative research is most appropriate when:
You have a hypothesis to test. Quantitative methods are designed for hypothesis testing. If you want to know whether a new onboarding flow increases 30-day retention, a controlled experiment (or at minimum a comparative analysis) gives you a defensible answer. An interview study cannot.
You need to estimate scale or prevalence. "40% of our users experience this problem" is a claim that requires a representative sample and a quantitative instrument. Twelve qualitative interviews cannot support that claim.
You need statistical power to detect small effects. If the difference between two outcomes is small, you need a large sample to detect it reliably. Qualitative research, with its small samples, cannot identify effects that are small in aggregate even if significant in impact.
The phenomenon can be reduced to predefined categories without losing what matters. A satisfaction scale works for overall satisfaction because the concept is simple enough to rate on a five-point scale. It does not work for understanding what satisfaction means or what drives it.
When is qualitative research the right choice?
Qualitative research is most appropriate when:
You do not yet know what the important variables are. Before you can design a survey, you need to understand the range of possible experiences and responses. Qualitative research is the appropriate starting point for phenomena that are underexplored or poorly understood.
The research question asks "how" or "why." How do employees experience remote work? Why do customers churn? What makes expert consulting advice credible? These questions require interpretation of meaning, not measurement of frequency.
Context is analytically important. What a phenomenon means varies by context: the same behaviour means different things in different organisational cultures, national settings, or life stages. Qualitative research preserves and analyses context in ways quantitative research abstracts away.
The topic is sensitive or complex. Experiences of discrimination, mental health, bereavement, or ethical conflict require a research instrument that is flexible, relationship-based, and participant-led. A fixed survey scale is inadequate for this kind of material.
The false choice between them
The quantitative/qualitative distinction is often presented as a binary: pick one and commit. This is rarely the right framing for real research questions.
Most important questions in social science, business, and public policy have a quantitative dimension (how widespread? how large an effect? how many people?) and a qualitative dimension (what does it mean? why does it happen? what are people actually experiencing?). Treating these as competing approaches is less useful than treating them as complementary tools that answer different parts of the same question.
The formal answer to this is mixed methods research: research designs that integrate both strands deliberately. But even without a formal mixed methods design, most good research draws on both. A quantitative study needs qualitative understanding to design the right survey items. A qualitative study needs quantitative context to situate its findings.
The 5 main differences, explained with examples
1. Research question
Quantitative: "What percentage of customers would recommend this product, and does this differ by segment?"
Qualitative: "What experiences lead customers to recommend or not recommend this product, and how does the way they frame their recommendation decision vary by their relationship with the product?"
The quantitative question produces a number with confidence intervals. The qualitative question produces an understanding of the decision process. Both are valuable; neither is a substitute for the other.
2. Data collection
Quantitative data collection uses standardised instruments: the same survey, the same experiment, the same measurement applied to everyone. This is how you get data that is comparable across respondents and generalisable across populations.
Qualitative data collection is adaptive: the interviewer follows the participant's lead, pursuing interesting threads and setting aside questions that are not relevant to this particular person's experience. This is how you get depth and discovery.
3. Sample size and sampling logic
Quantitative research uses random or representative sampling to ensure that findings from the sample can be generalised to a population. Sample sizes are determined by statistical power: how large a sample do you need to detect an effect of the size you expect?
Qualitative research uses purposive or theoretical sampling: choose participants who have the experience you are studying, who represent the range of perspectives you want to capture, or who are most likely to challenge your emerging analysis. Sample sizes are determined by theoretical saturation: keep going until you are no longer finding new themes. For academic research, this typically means 15-25 participants for a well-defined research question. For more on this, see qualitative research sample size.
4. Analysis
Quantitative analysis applies statistical tests to numbers: means, distributions, correlations, regressions, significance tests. The goal is to establish patterns that are unlikely to have occurred by chance.
Qualitative analysis interprets meaning in text, images, or observations: coding, theming, contextualising. The goal is to develop a credible, evidence-grounded account of what is happening and why. The main approaches are thematic analysis, grounded theory, content analysis, and discourse analysis.
5. What you can claim from findings
Quantitative findings support generalisable claims: "X% of Y population does Z, with a margin of error of ±N percentage points."
Qualitative findings support contextual claims: "In this study of N participants selected for X characteristic, the analysis identified the following patterns, which are likely to be applicable to similar contexts for the following reasons."
This distinction matters when presenting findings to stakeholders who expect statistical evidence. Qualitative findings are not less rigorous than quantitative ones; they support different kinds of claims. See presenting qualitative research findings to executives for practical guidance on navigating this.
A decision framework for choosing
When facing the choice between qualitative and quantitative research, three questions help:
1. What is the state of knowledge about this topic? If the phenomenon is well understood and the relevant variables are known, quantitative research can test and measure efficiently. If the phenomenon is underexplored, qualitative research is the appropriate starting point.
2. What kind of claim do you need to make? If you need to say "X% of people do Y," you need quantitative research with a representative sample. If you need to say "here is how and why people experience Y," you need qualitative research.
3. What is the cost of being wrong in each direction? Quantitative research risks measuring the wrong things precisely. Qualitative research risks finding patterns that do not generalise. Which type of error is more costly for your specific decision?
What AI is changing about this distinction
Recommended reading
The renaissance of qualitative research
The traditional quantitative/qualitative trade-off was partly about cost: quantitative research could be scaled cheaply (add another row to the survey dataset), while qualitative research could not (adding another interview means another 60+ hours of researcher time for data collection and analysis).
AI-assisted qualitative analysis is changing the cost structure. What previously required weeks of analyst time for a 25-interview dataset can now be done in hours. This has two effects on the quantitative/qualitative choice.
First, qualitative research now competes on cost at scales previously associated only with quantitative methods. A structured analysis of 150 interview transcripts, unthinkable without months of researcher time a decade ago, is now achievable within a standard project timeline. The sample sizes that make qualitative findings statistically interesting are within reach.
Second, mixed methods designs that were previously too expensive to run are now viable. When qualitative analysis no longer consumes 40-50% of a project's research time, the remaining budget can support a quantitative strand. Teams that previously had to choose between qualitative and quantitative can now run both.
This does not make quantitative research obsolete. Statistical testing, experimental design, and representative sampling remain essential for the kinds of claims that require them. But the argument that qualitative research is inherently limited by scale (that it is inherently a method for small, exploratory questions) is weaker than it was. For more on how this plays out, see strengths and weaknesses of qualitative research.
The new landscape looks less like a binary choice and more like a spectrum: from pure qualitative interpretation of small purposive samples, through AI-assisted qualitative analysis at scale, through mixed methods integration, to pure quantitative measurement. The expanded middle of that spectrum, AI-assisted qualitative research with samples of 50-200 participants, is where much of the interesting methodological development is happening.
Frequently asked questions
Is qualitative research less scientific than quantitative research?
No. Qualitative research meets the criteria for scientific inquiry: systematic data collection, transparent analytical procedures, evidence-grounded claims, and peer review. Rigour looks different from quantitative rigour (it is assessed through credibility, transferability, and confirmability rather than reliability and validity), but it is not lesser. Many of the most important findings in social science, psychology, organisational behaviour, and public health came from qualitative research.
Can qualitative data be made quantitative?
Yes, through coding and quantification: assigning numerical values to coded categories and analysing frequencies and cross-tabulations. This is the basis of content analysis and of the qualitative component in mixed methods research. For more on this, see how to quantify qualitative data.
When should I use a mixed methods design?
Mixed methods is appropriate when your research question has both a "how many" and a "why" component; when you want to test whether qualitative findings hold at scale; when you want to explain quantitative patterns through qualitative investigation; or when you want to triangulate findings across different data sources to increase confidence. For more, see mixed methods research.
What is the most common mistake when choosing between qualitative and quantitative?
Choosing based on what is familiar rather than what is appropriate. Quantitatively trained researchers default to surveys even when the question requires interpretation. Qualitatively trained researchers resist quantification even when scale matters. The question to start with is always: what kind of claim does my research question require? The method follows from that.
What is the difference between qualitative and quantitative data?
Qualitative data is information that takes the form of words, images, or observed behaviour: interview transcripts, field notes, documents, recordings. Quantitative data is information that takes the form of numbers: ratings, counts, measurements. For a more detailed explanation of qualitative data types, see qualitative data examples.
Looking to add rigour and scale to your qualitative research without the traditional time cost? Try Skimle for free and see how AI-assisted analysis works alongside your existing methods.
Related reading: What is qualitative research? Methods, types and when to use it | Mixed methods research: 4 designs, examples and when to use each | Strengths and weaknesses of qualitative research
About the authors
Henri Schildt is a Professor of Strategy at Aalto University School of Business and co-founder of Skimle. He has published over a dozen peer-reviewed articles using qualitative methods, including work in Academy of Management Journal, Organisation Science, and Strategic Management Journal. His research focuses on organisational strategy, innovation, and qualitative methodology. Google Scholar profile
Olli Salo is a former Partner at McKinsey & Company where he spent 18 years helping clients understand the markets and themselves, develop winning strategies and improve their operating models. He has done over 1000 client interviews and published over 10 articles on McKinsey.com and beyond. LinkedIn profile
