Measurement scale
A measurement scale classifies a variable as nominal, ordinal, interval or ratio. The level determines which statistical operations are valid: calculating a mean is meaningless for nominal data, while most financial figures are ratio-scale.
See it move
Loading infographic...
Measurement scales form a ladder of increasing power. Nominal data are categories only, such as industry sector. Ordinal data add ranking, such as a credit rating grade, but the gaps are unequal. Interval data have equal gaps with no true zero, like degrees Celsius. Ratio data add a true zero, like revenue, so the full range of arithmetic — means, ratios, regression — becomes valid.
Where it fits
TopicFoundations: Data, Populations & SamplingCoreTopicDescriptive StatisticsCoreSubjectData Analysis & StatisticsCore