The type of measurement scale used for generating scores, which determines the permissible transformations that can be performed on the scores without changing their relationship to the attributes being measured. For example, it is meaningless to calculate the mean of a collection of telephone numbers, because the level of measurement involved in the numbers (a nominal scale) renders the operations of addition and division meaningless. Level of measurement is not a property of the data themselves, but of the data and their interpretation considered together: the same scores can represent different levels depending on their interpretation. Among the most important measurement levels, with separate entries, are the following, arranged from lowest to highest level of measurement: nominal scale, ordinal scale, interval scale, ratio scale, absolute scale. Another, at the same level as the interval scale, is the log-interval scale. See also measurement, scale (1).