Variables at an interval scale of measurement are, typically, those such as temperature (except in the Kelvin scale) or historical time. The values of these variables are quantifiably different, meaning they have a fixed quantifiable difference. Yet, this level of measurement has no real "zero" value, but if a "zero" value exists it is purely arbitrary.

For example, the Celsius scale is an interval measure. Each degree of variation is fixed and means the same along the scale. The scale has a "zero" value, but this is an arbitrary "zero", set as for convenience by Celsius himself. Another interval scale for temperature is the Fahrenheit scale, which, again, has fixed degrees of variation along this scale and another arbitrary "zero" value. Because the difference in values is fixed for both scales, transforming Celsius degrees into Fahrenheit degrees, and vice versa, is easy and dependable.

Another interval scale is historical time. Many Western countries use an arbitrary "year zero" set around 2000 years ago for measuring historical times. Yet other ways of counting historical time exists, such as the Muslim calendar, the Hebrew calendar, etc.

The values of variables at an interval scale are normally quantified numerically, although the actual quantification may vary depending on the scale used (eg, Celsius degrees and Fahrenheit degrees, or historical time according to a particular calendar).

We may use the concept of "interval variable" (or "interval-level variable") to refer to those variables which use an interval scale of measurement. This is "handy" although not fully correct. That is, the quality of being "interval" refers to the scale not to the variable. That is, it is possible to transform the values of a "quantitative" variable into an interval scale, and, although less likely, to treat the values of an interval scale as if they were at a "higher" level of measurement.

For example, we may treat an ordinal variable (satisfaction, or intelligence) as an interval one, analyzing the numerical data using statistics more appropriate to the latter than to the former.

# Properties

- Interval variables vary quantitatively in a fixed and dependable manner.
- If required, the most appropriate "average" for interval variables is the mean.

# Want to know more?

**Wikipedia - Level of measurement**- You can learn a bit more about levels of measurement in Wikipedia.