It is important to understand types of data when planning a usability study. In this blog, we will cover 4 essential types of data in usability: Nominal, Ordinal, Interval, and Ratio data.

Each type has its own pros and cons, by exploring their characteristics we can understand what are we dealing with, and what can or can’t be done with each data type. So, without any further delation let’s dive into the first type.

Nominal Data

Nominal data is easy to understand, nominal data is simply an unordered set of categories for example different operating systems for mobile users.

• iOS

• Android

• Samsung One UI

• MIUI

The easiest way to recognize nominal data is that there is no ranking or order between categories, the only characteristic is that they are different. Although nominal data seems simple, and in a way primitive data type when it comes to mathematical characteristics, it is necessary in many usability studies when we want to gather data such as geographic location, gender, or task success (binary).

Ordinal Data

Because it has ranking ordinal data is a level higher than nominal data. The trick with ordinal data is that the ranking between categories doesn’t necessarily need to be equal. For example, think about educational levels.

• Elementary school

• High school

• Bachelor’s degree

• Master’s degree

• Ph.D.

We can clearly see the order in this list, high school comes after elementary school because it is a higher level of education, but it’s not clear what are the values between levels.

Interval Data

Interval data has ranking but it also has equal intervals between categories. These two characteristics make interval data susceptible to comparison of magnitudes of values. Simply put, you can calculate the difference between two values and get a meaningful result.

However, interval data has it’s one limitation, it lacks a true zero point. This means that a value of zero on the scale does not represent the complete absence of the attribute being measured. Instead, it's an arbitrary point on the scale.

If a user rates usability on a scale of 0 to 10 with 0 that doesn’t mean the complete absence of usability, it's just the lowest point on the scale.

Ratio Data

Ratio data unlike interval data has an absolute zero, which means that 0 really means 0 (absence of the measured attribute). This makes ratio data the highest level of measurement, it is a data type with the most mathematical properties. Ratio data allows you to say that one user in the usability test was twice as fast as another. A classic example of ratio data is competition time, 0 seconds left means no time remaining.

Summary

• Nominal data consists of an unordered set of categories

• Ordinal data includes categories with a ranking, but the intervals between categories may not be equal

• Interval data includes ranking and equal intervals between categories

• Ratio data is the highest level of measurement and has an absolute zero, representing the complete absence of the measured attribute