This book discusses constructing variables and making measures. It begins by outlining the qualities a number must meet before it qualifies as a measure of something. The basis is the measurement philosophy of G. Rasch. The first requirement for making good measures is good raw material. To achieve the possibility of comparisons, the data must contain the possibility of a single variable along which persons can be measured. Chapter 2 presents a set of data that must be inspected, and techniques for inspecting the data are reviewed. In chapter 3, five different models for measuring are described, each of which has been developed for a particular type of data. There are other models in the measurement literature, but these, all members of a family, are used because they meet the standards set for measurement. Chapter 4 shows how to use these models to get results, describing four different estimation procedures: PROX, PAIR, UCON, and CON. The quality control of variables is discussed in chapter 5. Chapters 6, 7, and 8 then illustrate the use of the techniques discussed with four different data sets, collected to measure drug use, fear of crime, knowledge of elementary physics, and child development respectively. (Contains 76 figures, 37 tables, and 52 references.) (SLD)