Back in 1981, Milton Lodge, a political scientist, published a book which gave researchers from the social sciences a crash course in using the method of magnitude estimation (ME), which he claimed to yield ratio scale type data, and which had, up to that point, been mostly used in psychophysics. Lodge was confident that this method would vastly improve the measurement of attitudes and opinions which was then dominated by the method of Likert-scale type questionnaires (asking participants to give judgments on an n-point scale), which produced only data of nominal or ordinal scale type. In his book, Lodge issued a warning against premature adaptations of the ME method, but also against premature rejection of its usage:

“If people can make only nominal or ordinal judgments on a dimension, then overspecifying the data as having interval or ratio properties will misrepresent the relations truly implied by their judgments. If, on the other hand, people are capable of making ratio judgments about the intensity of some social attributes, but only interval or ordinal judgments are recorded, the underspecificity will result in the loss of both information and proper access to powerful analytical tools.”(Lodge 1981: 30)