What Is Quantitative Data? Definition, Types, and Examples

What is quantitative data? Put simply, it's anything you can count or measure. The number of students in a class, how long a patient waits, or a score out of 40 are all numerical data, and you can add, average, and compare them.

Plenty of numbers in a spreadsheet aren't quantities at all. A student ID of 1042 isn't "more" than 1041. Spotting that difference early saves you from running the wrong statistical analysis later.

What is quantitative data, and how can you spot it?

Quantitative data answers two questions: how many, and how much. A count gives you something like 42 participants. A measurement gives you something like 7.5 hours of sleep or a blood pressure of 118 mmHg.

Here's a quick test. Subtract one value from another and ask whether the answer means anything. Thirty minutes of exercise minus twenty minutes leaves ten extra minutes, so exercise time is quantitative. ZIP code 90210 minus 10001 gives you 80209 of nothing, so a ZIP code is just a label made of digits.

Some numbers only look quantitative. Participant IDs, postal codes, and phone numbers are nominal data, which means they name things. Codes like "1 = male, 2 = female" are categories too. Class rank and age groups (say, 18 to 24) are ordinal data, meaning they have an order but the gaps between them aren't equal.

Likert items are the tricky ones. A single "strongly disagree to strongly agree" question is usually treated as ordinal, and Jamieson (Medical Education, 2004) argued it should be analyzed that way. When you add several related items into one score, many researchers treat the total as quantitative. Norman (Advances in Health Sciences Education, 2010) showed common parametric tests cope well with this kind of data. Either is defensible if you say which you chose.

Types of quantitative data: discrete and continuous

Discrete data come from counting. The values jump from one whole number to the next, so you can't have 2.5 hospital visits. Citations, errors on a test, and number of children are all discrete.

Continuous data come from measuring. They can take any value in a range, like 72.36 kg or 1.8 seconds. Height, time, weight, and temperature are the usual examples.

Age is recorded in whole years, yet most researchers treat it as continuous because it has so many possible values.

Levels of measurement: interval and ratio scales

Back in 1946, psychologist S. S. Stevens published four levels of measurement in the journal Science: nominal, ordinal, interval, and ratio. The first two describe categories. Interval and ratio scales describe quantitative variables.

On an interval scale, the gaps between values are equal, but zero isn't a true "none." Celsius is the textbook case. Zero degrees doesn't mean no temperature, so 20 °C isn't twice as hot as 10 °C. You can still take averages and differences.

A ratio scale has a real zero. Zero kilograms means no weight at all, so 20 kg really is double 10 kg. Income, reaction time, distance, and temperature in kelvin all work this way.

Still, treat these levels as a guide. Velleman and Wilkinson (The American Statistician, 1993) warned that picking tests by level alone can mislead.

Quantitative data examples in research

Ten variables you'll often meet in quantitative research:

1. Hospital visits in the past year (discrete, ratio)

2. Citations a paper has received (discrete, ratio)

3. Correct answers on a 40 question test (discrete, ratio)

4. Height in centimeters (continuous, ratio)

5. Reaction time in milliseconds (continuous, ratio)

6. Systolic blood pressure (continuous, ratio)

7. Hours of sleep per night (continuous, ratio)

8. Rainfall in millimeters (continuous, ratio)

9. Temperature in Celsius (continuous, interval)

10. Calendar year of birth (discrete, interval)

In a survey on sleep and exam results, hours of sleep and exam score are quantitative. Student ID is a label, and year of study is ordinal.

Ask for exact numbers whenever you can. If you collect exact ages, you can group them into brackets later. If you only collect brackets, you can never get the exact ages back.

How to collect quantitative data

Most data collection methods fall into five groups. Surveys with closed questions give counts and ratings. Experiments measure outcomes under controlled conditions. Structured observation has trained observers to count specific behaviors. Instruments such as scales, sensors, and wearables record measurements directly. Secondary data, like census figures or hospital records, were collected by someone else, so check how they defined each variable.

Whatever method you pick, decide exactly what you're measuring first. Writing an operational definition for each variable keeps every measurement consistent.

Next, decide who to sample, since that shapes how far your results generalize. If you're weighing up approaches, our comparison of stratified and cluster sampling methods explains the tradeoffs.

How to analyze quantitative data

Descriptive statistics summarize your data. For continuous data with a fairly even spread, report the mean and standard deviation, and plot a histogram or box plot. If the data are skewed, like income, the median and interquartile range (the spread of the middle half of values) give a fairer picture.

Then move to inferential statistics, which test whether a pattern is likely to hold beyond your sample. Common choices for continuous outcomes include the t test, ANOVA, Pearson correlation, and linear regression. Skewed data often suit rank based tests or Spearman correlation. For counts, Poisson and negative binomial regression are built for the job.

Your sample size and study design also matter, so run your plan past a supervisor.

Qualitative vs quantitative data

Qualitative data describe qualities in words, images, or categories. Think interview transcripts, open answers, or eye color. Quantitative data turn things into numbers.

Questions about how much, how many, whether groups differ, or whether two things are related call for numbers. Questions about why something happens, or how people experience it, call for words. Qualitative researchers code data into themes, and this thematic analysis example using interview data shows how that works step by step.

Many studies combine both, which is called mixed methods. Your choice often reflects your wider research philosophy, which our guide to positivism vs interpretivism unpacks.

Limitations and common mistakes

Numbers tell you what happened, but rarely why. They're only as accurate as the tools used to collect them. A strong correlation also doesn't prove cause and effect unless your design supports it.

Common slips include:

1. Saying "twice as much" with interval data.

2. Reporting a mean when the data are heavily skewed.

3. Leaving out units. "Sleep = 7" should read 7 hours per night.

Conclusion

So, what is quantitative data? It's any count or measurement where the numbers stand for real amounts. Run the subtraction test, work out whether you have discrete or continuous data on an interval or ratio scale, and let that guide your analysis.

FAQs

Can qualitative data become quantitative?

Partly. You can count how often a theme appears across interviews, a method called content analysis. The counts are numbers, but the meaning behind them is still qualitative, so be clear about which one you analyzed.

Is money discrete or continuous?

Technically it's discrete, because it's counted in the smallest unit, such as cents. In practice, researchers usually treat it as continuous because it has so many possible values.

Is quantitative data more reliable than qualitative data?

Not automatically. It's more standardized, but its quality depends on how carefully you measure it. The better choice is the one that fits your research question.

Can Canyam help me see how other studies measured their variables?

Yes. Canyam lets you search more than 65 million research papers and read AI summaries of their key findings, methods, and citations. Comparing methods sections in your field shows which validated measures other researchers used. It's also a practical way to answer quantitative data in your own research area.