Types of Data: Qualitative vs. Quantitative

In statistics, data is our raw material. But just like a chef can’t treat a piece of chicken the same way they treat a carrot, a statistician cannot treat all data the same way.

If you try to calculate the “average” of a list of postal codes, the math will technically work, but the answer will be completely meaningless. Postal codes are just labels disguised as numbers.

To choose the correct statistical tool, you first have to know what kind of data you are holding. All data in the universe can be split into two massive categories: Qualitative and Quantitative.

1. Qualitative Data (Categorical)

Qualitative data describes qualities or characteristics. It groups things into categories. You cannot do traditional math (like addition or division) on qualitative data.

We can break this down into two sub-types:

Nominal Data (No Order)

Nominal data consists of categories that have no logical order or ranking. They are just names or labels.

• Examples: Eye colour (Blue, Brown, Green), Blood type (A, B, O), or Car brands (Toyota, Ford, Honda).
• The Rule: Is “Blue” mathematically greater than “Brown”? No. They are just different.

Ordinal Data (Ordered)

Ordinal data also consists of categories, but these categories have a strict, logical order.

• Examples: Customer satisfaction ratings (Poor, Fair, Good, Excellent), T-shirt sizes (Small, Medium, Large), or Education level (High School, Bachelor’s, Master’s).
• The Rule: You know that “Excellent” is better than “Good,” but you don’t know exactly how much better it is. The distance between the categories isn’t perfectly measurable.

2. Quantitative Data (Numerical)

Quantitative data is all about numbers. It represents a measurable quantity. This is the data you can safely add, subtract, multiply, and divide.

Quantitative data is also split into two sub-types based on how it is counted:

Discrete Data (Counted)

Discrete data consists of whole numbers. It represents things that can be counted individually. You cannot have half of a discrete unit.

• Examples: The number of children in a family (1, 2, or 3), the number of cars in a parking lot, or the number of heads you get when flipping a coin.
• The Rule: You cannot have 2.5 children or 4.1 cars. The data must be a distinct, whole integer.

Continuous Data (Measured)

Continuous data can take on any value within a range, including infinitely small fractions and decimals. It represents things that are measured rather than counted.

• Examples: Height (175.5 cm), Weight (68.2 kg), Temperature (28.6 degrees Celcius), or Time (4.33 seconds).
• The Rule: Continuous data is only limited by the precision of your measuring tool. A stopwatch can measure 4 seconds, 4.3 seconds, or 4.332 seconds.

The Ultimate Cheat Sheet

If you ever get confused about a dataset, use this quick reference guide:

Main Category	Sub-Type	The Core Question	Example
Qualitative (Words/Labels)	Nominal	Is there a logical order? (No)	Hair Colour
	Ordinal	Is there a logical order? (Yes)	Movie Ratings (1 to 5 stars)
Quantitative (Numbers)	Discrete	Can I have a fraction of it? (No)	Number of Employees
	Continuous	Can I have a fraction of it? (Yes)	Exact Salary in Rupees and Paise (or Dollars and cents)

Understanding these four types of data is crucial. The type of data you have dictates exactly which charts you can draw and which statistical tests you are allowed to run.