Glossary of Statistical Terms: Definition Rating scale
A rating scale is used to record rank-based statements of respondents in a survey. Often, rating scales are composed of steps 1 to 7, but generally the number of steps can vary. Usually, we label the two ends of a scale with descriptions such as "1: I'm very interested" to "7: I'm not interested at all". Alternatively, all the steps of a scale are labeled; this is not essential, though. When looking at rating scales with an uneven number of steps, the mean value usually represents the answer "do not know/undecided".
Strictly speaking, most rating scales are ordinal scales. The intervals between the individual steps are not always the same size. It is not unusual in statistics, however, to define these rating scales as interval scales anyway. Same-sized intervals between individual steps are assumed; therefore a mean is sometimes defined for rating scales as well.
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.