We discuss several methodological problems that affect measurement of race and origin, and report on results of methodological research conducted to understand the impact of the questionnaire on measurements and to design better questions and procedures that will yield high quality and comparable measurements. First, we provide background for our methodological discussion by discussing the social constructs of race and Hispanic origin. We briefly describe the sources of data that we draw on, and apply them to discuss several methodological difficulties and the design solutions that have been developed to try to solve them.