Glossary of Statistical Terms: Definition Interval scale
In interval scales, the location parameters mode, median and mean can be calculated. An interval scale can always be divided into equal portion scales. For example, you can use a ruler to measure length or a stopwatch to measure a time interval.
Another typical example is the Celsius scale, which is divided into one hundred equal degree sections between the freezing and boiling points of water. Therefore, unlike with nominal and ordinal scales, the intervals between the individual data can be calculated with. The difference in temperature between 10°C and 20°C is the same distance as between 40°C and 50°C. However, other than ratio scales, interval scales lack a natural zero point. In the case of the Celsius temperature scale '0°C' is an arbitrarily chosen zero point (whereas the Kelvin temperature scale which begins at the absolute zero point with 0K (-273.15°C/-459.67°F). Multiplications with an interval scale are therefore not useful. While it is true for the Kelvin ratio scale that 100K defines double the heat energy in a mass as 50K, this is not true for 50°C/100°C.
Another type of interval scale is a rating scale. With rating scales it is usually assumed that the distances between the single expressions of evaluation (score) are equal. There is however no natural zero point. Thus, we can calculate an average of a rating scale. However, it is not possible to make the statement that the score 8 (on a scale of 1 to 11) is twice as high as the score 4.
Please note that the definitions in our statistics encyclopedia are simplified explanations of terms. Our goal is to make the definitions accessible for a broad audience; thus it is possible that some definitions do not adhere entirely to scientific standards.