Measurement scale.
Place each possible value taken by the data, and because of the different characteristics of the data, there are different types of scales. Nominal scale, when a data identifies a label for an element. In this way the order of the labels are meaningless, the purpose of this type of data is to classify observations. Ordinal scale, when the data shows the properties of the nominal data, but also makes sense of the order. Interval scale, the data has the properties of the ordinal data, but in turn the separation between the variables makes sense. Reason scale, the data has all the properties of the interval data, and the ratio between them makes sense.