This book declines to take for granted the widespread assumption that existing psychometric procedures provide scientific measurement. The currently fashionable concepts of measurement within psychology — operationalism and representationalism — are critically examined, and the classical view, that measurement is the assessment of quantity, is defended. Within this framework, it is shown how conjoint measurement can be used to test the hypothesis that variables are quantitative. This theme is developed in detail using familiar psychological examples, such as Thurstone”s law of comparative judgment, multidimensional scaling, and Coombs” theory of unfolding.