This essay will focus on the problems that arise in both intrapersonal and interpersonal ordinal measurement – that is in comparing a person’s health states at different times or in comparing the health states of different persons. The paper is concerned exclusively with health, its relations to well being and preferences, and how the health states of individuals should be compared. nSection 1 explains why health measurements are really evaluations, and section 2 argues that there are many ways to evaluate health states. Section 3 explores the reasons in favor of the social choice model, while section 4 argues that the well-known problems that preferences may reflect false beliefs and cognitive deficiencies are particularly serious in the context of evaluating health. Section 5 asks what harms may result in practice if one evaluates health in terms of preferences. Section 6 argues that health states should not be evaluated in terms of their consequences for well-being and suggests that they should instead be evaluated in terms of their consequences for comfort and capabilities.