Levels of Measurement of a Variable

Levels of Measurement of a Variable

Click on each for additional information.

Nominal

The possible values of the variable are categorical and the categories do not have a meaningful order or ranking to them. Even if the categories are given numbers (e.g. 1=brown eyes, 2=blue eyes, etc.), these numbers are arbitrary and do not denote an order.

Gender

Eye Color

School Campus

e.g., male, female, nonbinary, transgender, etc.

e.g., brown, blue, green, hazel, etc.

e.g., campus A, campus B, campus C, etc.

Ordinal

The possible values of the variable are categorical and the categories have a meaningful order or ranking to them. You can compare values in different categories by saying one is "greater/larger/higher/more significant than" the other, but you can not measure differences between them. You also cannot compare values within the same category.

Likert Scale

Age Ranges

Size of School

e.g., strongly disagree, disagree, neutral, agree, etc.

e.g., 0-17 years, 18-35, 36-49, 50 and older

e.g., small school, medium school, large school, etc.

Interval

The possible values of the variable are numerical, not categorical. There are equal distances between adjacent values and you can calculate differences between values. However, there is no meaningful zero value. If a value of 0 exists, it does not indicate the absence of the quantity being measured. Also, if multiple scales exist, the value of 0 may vary between the scales.

Temperature (F or C)

Calendar Year

IQ Score

Ratio

The possible values of the variable are numerical, not categorical. There are equal distances between adjacent values and you can calculate differences between values. There is a meaningful zero. The value of 0 indicates the absence of the quantity being measured. The 0 value is equivalent across different scales.

Age

Annual Income

Test Score

Note: If the values of an interval or ratio variable are grouped into categories (e.g., age ranges, temperature ranges, income ranges, letter grade on a test, etc.), it becomes an ordinal variable.

Created by: Lisa R. Scavone