Qualitative data is information you can describe with words rather than numbers.
Quantitative data is information that uses numbers.
One type of data isn’t better than the other.
To conduct thorough research, you need both. Here are the key differences:
| Qualitative Data | Quantitative Data | |
|---|---|---|
| Type of Data | Descriptive and conceptual. Involves words, images, and observations. | Numerical and measurable. Involves counts, percentages, and statistical values. |
| When to Use | To explore underlying reasons, motivations, and meanings. | To quantify results or to identify patterns. |
| Data Collection Methods | Open-ended methods such as interviews, focus groups, observations, and open-ended surveys. | Structured methods such as closed-ended surveys, experiments, analytics, and tests. |
| Sample Size | Typically small and non-representative. Focuses on depth over breadth. | Typically large and representative. Focuses on breadth and generalizability. |
| Methods of Analysis | Thematic analysis, content analysis, narrative analysis, subjective interpretation. | Statistical analysis, mathematical modeling, objective interpretation. |
| Flexibility | Very flexible, can change course as needed to explore any surprises that come up. | Less flexible, the structured data requires planning and is difficult to change once collection has begun. |
| Reliability | Often unreliable and prone to bias. | More reliable if the data collection is structured properly. |
| Examples | Transcripts from interviews discussing user experiences and notes from observing customer interactions. | Website traffic analytics showing number of visitors per day, survey results indicating percentage of customers satisfied with a product. |
#1. The Type of Data
The single biggest difference between quantitative and qualitative data is that one deals with numbers, and the other deals with concepts and ideas.
The words “qualitative” and “quantitative” are really similar, which can make it hard to keep track of which one is which. I like to think of them this way:
- Quantitative = quantity = numbers-related data
- Qualitative = quality = descriptive data
Qualitative data—the descriptive one—usually involves written or spoken words, images, or even objects. It’s collected in all sorts of ways: video recordings, interviews, open-ended survey responses, and field notes, for example.
I like how researcher James W. Crick defines qualitative research in a 2021 issue of the Journal of Strategic Marketing: “Qualitative research is designed to generate in-depth and subjective findings to build theory.”
In other words, qualitative research helps you learn more about a topic—usually from a primary, or firsthand, source—so you can form ideas about what it means. This type of data is often rich in detail, and its interpretation can vary depending on who’s analyzing it.
Here’s what I mean: if you ask five different people to observe how 60 kittens behave when presented with a hamster wheel, you’ll get five different versions of the same event.
Quantitative data, on the other hand, is all about numbers and statistics. There’s no wiggle room when it comes to interpretation. In our kitten scenario, quantitative data might show us that of the 60 kittens presented with a hamster wheel, 40 pawed at it, 5 jumped inside and started spinning, and 15 ignored it completely.
There’s no ifs, ands, or buts about the numbers. They just are.
#2. When to Use Each Type of Data
You should use both quantitative and quantitative data to make decisions for your business.
Quantitative data helps you get to the what. Qualitative data unearths the why.
Quantitative data collects surface information, like numbers. Qualitative data dives deep beneath these same numbers and fleshes out the nuances there.
Research projects can often benefit from both types of data, which is why you’ll see the term “mixed-method” research in peer-reviewed journals. The term “mixed-method” refers to using both quantitative and qualitative methods in a study.
So, maybe you’re diving into original research. Or maybe you’re looking at other peoples’ studies to make an important business decision. In either case, you can use both quantitative and qualitative data to guide you.
Imagine you want to start a company that makes hamster wheels for cats. You run that kitten experiment, only to learn that most kittens aren’t all that interested in the hamster wheel. That’s what your quantitative data seems to say. Of the 60 kittens who participated in the study, only 5 hopped into the wheel.
But 40 of the kittens pawed at the wheel. According to your quantitative data, these 40 kittens touched the wheel but did not get inside.
This is where your qualitative data comes into play. Why did these 40 kittens touch the wheel but stop exploring it? You turn to the researchers’ observations. Since there were five different researchers, you have five sets of detailed notes to study.
From these observations, you learn that many of the kittens seemed frightened when the wheel moved after they pawed it. They grew suspicious of the structure, meowing and circling it, agitated.
One researcher noted that the kittens seemed desperate to enjoy the wheel, but they didn’t seem to feel it was safe.
So your idea isn’t a flop, exactly.
It just needs tweaking.
According to your quantitative data, 75% of the kittens studied either touched or actively participated in the hamster wheel. Your qualitative data suggests more kittens would have jumped into the wheel if it hadn’t moved so easily when they pawed at it.
You decide to make your kitten wheel sturdier and try the whole test again with a new set of kittens. Hopefully, this time a higher percentage of your feline participants will hop in and enjoy the fun.
This is a very simplistic and fictional example of how a mixed-method approach can help you make important choices for your business.
#3. Data You Have Access To
When you can swing it, you should look at both qualitative and quantitative data before you make any big decisions.
But this is where we come to another big difference between quantitative vs. qualitative data: it’s a lot easier to source qualitative data than quantitative data.
Why? Because it’s easy to run a survey, host a focus group, or conduct a round of interviews. All you have to do is hop on SurveyMonkey or Zoom and you’re on your way to gathering original qualitative data.
And yes, you can get some quantitative data here. If you run a survey and 45 customers respond, you can collect demographic data and yes/no answers for that pool of 45 respondents.
But this is a relatively small sample size. (More on why this matters in a moment.)
To tell you anything meaningful, quantitative data must achieve statistical significance.
If it’s been a while since your college statistics class, here’s a refresh: statistical significance is a measuring stick. It tells you whether the results you get are due to a specific cause or if they can be attributed to random chance.
To achieve statistical significance in a study, you have to be really careful to set the study up the right way and with a meaningful sample size.
This doesn’t mean it’s impossible to get quantitative data. But unless you have someone on your team who knows all about null hypotheses and p-values and statistical analysis, you might need to outsource quantitative research.
Plenty of businesses do this, but it’s pricey.
When you’re just starting out or you’re strapped for cash, qualitative data can get you valuable information—quickly and without gouging your wallet.
#4. Big vs. Small Sample Size
Another reason qualitative data is more accessible? It requires a smaller sample size to achieve meaningful results.
Even one person’s perspective brings value to a research project—ever heard of a case study?
The sweet spot depends on the purpose of the study, but for qualitative market research, somewhere between 10-40 respondents is a good number.
Any more than that and you risk reaching saturation. That’s when you keep getting results that echo each other and add nothing new to the research.
Quantitative data needs enough respondents to reach statistical significance without veering into saturation territory.
The ideal sample size number is usually higher than it is for qualitative data. But as with qualitative data, there’s no single, magic number. It all depends on statistical values like confidence level, population size, and margin of error.
Because it often requires a larger sample size, quantitative research can be more difficult for the average person to do on their own.
#5. Methods of Analysis
Running a study is just the first part of conducting qualitative and quantitative research.
After you’ve collected data, you have to study it. Find themes, patterns, consistencies, inconsistencies. Interpret and organize the numbers or survey responses or interview recordings. Tidy it all up into something you can draw conclusions from and apply to various situations.
This is called data analysis, and it’s done in completely different ways for qualitative vs. quantitative data.
For qualitative data, analysis includes:
- Data prep: Make all your qualitative data easy to access and read. This could mean organizing survey results by date, or transcribing interviews, or putting photographs into a slideshow format.
- Coding: No, not that kind. Think color coding, like you did for your notes in school. Assign colors or codes to specific attributes that make sense for your study—green for positive emotions, for instance, and red for angry emotions. Then code each of your responses.
- Thematic analysis: Organize your codes into themes and sub-themes, looking for the meaning—and relationships—within each one.
- Content analysis: Quantify the number of times certain words or concepts appear in your data. If this sounds suspiciously like quantitative research to you, it is. Sort of. It’s looking at qualitative data with a quantitative eye to identify any recurring themes or patterns.
- Narrative analysis: Look for similar stories and experiences and group them together. Study them and draw inferences from what they say.
- Interpret and document: As you organize and analyze your qualitative data, decide what the findings mean for you and your project.
You can often do qualitative data analysis manually or with tools like NVivo and ATLAS.ti. These tools help you organize, code, and analyze your subjective qualitative data.
Quantitative data analysis is a lot less subjective. Here’s how it generally goes:
- Data cleaning: Remove all inconsistencies and inaccuracies from your data. Check for duplicates, incorrect formatting (mistakenly writing a 1.00 value as 10.1, for example), and incomplete numbers.
- Summarize data with descriptive statistics: Use mean, median, mode, range, and standard deviation to summarize your data.
- Interpret the data with inferential statistics: This is where it gets more complicated. Instead of simply summarizing stats, you’ll now use complicated mathematical and statistical formulas and tests—t-tests, chi-square tests, analysis of variance (ANOVA), and correlation, for starters—to assign meaning to your data.
Researchers generally use sophisticated data analysis tools like RapidMiner and Tableau to help them do this work.
#6. Flexibility
Quantitative research tends to be less flexible than qualitative research. It relies on structured data collection methods, which researchers must set up well before the study begins.
This rigid structure is part of what makes quantitative data so reliable. But the downside here is that once you start the study, it’s hard to change anything without negatively affecting the results. If something unexpected comes up—or if new questions arise—researchers can’t easily change the scope of the study.
Qualitative research is a lot more flexible. This is why qualitative data can go deeper than quantitative data. If you’re interviewing someone and an interesting, unexpected topic comes up, you can immediately explore it.
Other qualitative research methods offer flexibility, too. Most big survey software brands allow you to build flexible surveys using branching and skip logic. These features let you customize which questions respondents see based on the answers they give.
This flexibility is unheard of in quantitative research. But even though it’s as flexible as an Olympic gymnast, qualitative data can be less reliable—and harder to validate.
#7. Reliability and Validity
Quantitative data is more reliable than qualitative data. Numbers can’t be massaged to fit a certain bias. If you replicate the study—in other words, run the exact same quantitative study two or more times—you should get nearly identical results each time. The same goes if another set of researchers runs the same study using the same methods.
This is what gives quantitative data that reliability factor.
There are a few key benefits here. First, reliable data means you can confidently make generalizations that apply to a larger population. It also means the data is valid and accurately measures whatever it is you’re trying to measure.
And finally, reliable data is trustworthy. Big industries like healthcare, marketing, and education frequently use quantitative data to make life-or-death decisions. The more reliable and trustworthy the data, the more confident these decision-makers can be when it’s time to make critical choices.
Unlike quantitative data, qualitative data isn’t overtly reliable. It’s not easy to replicate. If you send out the same qualitative survey on two separate occasions, you’ll get a new mix of responses. Your interpretations of the data might look different, too.
There’s still incredible value in qualitative data, of course—and there are ways to make sure the data is valid. These include:
- Member checking: Circling back with survey, interview, or focus group respondents to make sure you accurately summarized and interpreted their feedback.
- Triangulation: Using multiple data sources, methods, or researchers to cross-check and corroborate findings.
- Peer debriefing: Showing the data to peers—other researchers—so they can review the research process and its findings and provide feedback on both.
Whether you’re dealing with qualitative or quantitative data, transparency, accuracy, and validity are crucial. Focus on sourcing (or conducting) quantitative research that’s easy to replicate and qualitative research that’s been peer-reviewed.
With rock-solid data like this, you can make critical business decisions with confidence.
