Data Types — Qualitative, Quantitative, Discrete & Continuous

Data types in statistics tell you what a variable represents. If the values are labels such as "red" or "biology," the data is qualitative. If the values are numerical amounts, the data is quantitative. Then, for quantitative data, you usually ask one more question: is it a discrete count or a continuous measurement?

This matters because the data type affects which graphs, summaries, and models make sense. A mean can help with heights, but not with eye color.

Qualitative vs quantitative data

Qualitative data means categories

Qualitative data describes qualities, groups, or labels rather than numerical amounts. Examples include car color, blood type, and country.

This kind of data is also often called categorical data.

Quantitative data means numerical amounts

Quantitative data records a numerical amount. The number is not just a label; it represents how much, how many, or how far.

Examples include age, height, test score, and number of pets.

Discrete vs continuous data

Discrete data comes from counting

Discrete data is quantitative data that usually comes from counting. The values jump from one allowable value to another instead of filling an entire interval.

The number of students in a class is discrete because you count whole students. Under an ordinary counting model, values such as $24.5$ students do not make sense.

Continuous data comes from measuring

Continuous data is quantitative data that usually comes from measuring. In principle, the value can be recorded with finer and finer precision, depending on the measuring tool and context.

Height, time, and temperature are standard examples. A person's height might be written as $170$ cm, $170.2$ cm, or $170.24$ cm depending on the precision you use.

Worked example: classifying student data

Suppose a school records these four variables for each student:

• homeroom
• number of siblings
• travel time to school
• favorite subject

Here is how to classify them.

Homeroom is qualitative because it is a group label.

Number of siblings is quantitative and discrete because it is a count: $0, 1, 2, 3,$ and so on.

Travel time to school is quantitative and continuous because it is measured. You might round it to the nearest minute, but the underlying variable can vary more finely than that.

Favorite subject is qualitative because it names a category, not an amount.

This example shows the main decision path. First ask "label or amount?" If it is an amount, ask "count or measurement?"

How to tell which data type you have

Use this rule of thumb:

1. If averaging the values would be meaningless, the data is probably qualitative.
2. If averaging would make sense, the data is probably quantitative.
3. If the quantitative values come from counting separate items, they are usually discrete.
4. If they come from measuring on a scale, they are usually continuous.

This is a practical shortcut, not a formal proof. The context of the variable still matters.

Common mistakes with data types in statistics

Treating numeric codes as real quantities

If survey answers are coded as $1$, $2$, and $3$, those numbers may still stand for categories rather than actual amounts. A number in the data does not automatically make the variable quantitative.

Assuming every whole-number value is discrete

A recorded measurement can appear as a whole number only because it was rounded. For example, weights listed as $68$, $72$, and $75$ kilograms are still continuous data if weight was measured rather than counted.

Mixing up the variable with the way it is stored

Travel time rounded to the nearest minute is often stored as whole numbers, but the variable itself is still continuous. The recording format does not always change the underlying type.

Where these data types are used in statistics

The classification matters whenever you choose a graph, summary, or statistical method.

For qualitative data, bar charts and frequency tables are common. For quantitative data, histograms, box plots, means, medians, and standard deviations may be useful.

The discrete-versus-continuous split also matters when you choose a probability model. Some models are built for counts, while others are built for measurements on a continuum.

Try your own version

Take five variables from everyday life, such as shoe size, ZIP code, temperature, number of emails, or hair color, and classify each one. If a case feels ambiguous, state the condition that decides it, such as whether the value is a label, a count, or a measurement.

If you want to go one step further, explore another case by asking which graph or summary makes sense for each variable and which one does not.