Levels of Measurement and Political Research: An Optimistic View

This paper argues that all measurement is theory testing. Therefore, measurement always constitutes a tentative statement about the nature of reality. Some general implications of this perspective are discussed. Alternating Least Squares, Optimal Scaling (ALSOS) is then presented as a general strategy for obtaining empirical information about the measurement properties that exist within a given data set. The great advantage of this approach is that characteristics such as the levels of measurement associated with particular variables are viewed as testable hypotheses rather than a priori assumptions. An ALSOS regression analysis is performed on data from the CPS 1992 National Election Study. The results show that several variables which are usually interpreted as pseudo-interval measures actually only provide ordinal-level information. More generally, levels of measurement are important because they affect the degree of ambiguity in researchers' interpretations of variability in empirical data.