Cattell (1944) initiated the use of the term 'ipsative' to refer to raw score transformations that center scores about the individual's mean. Ipsative scores were contrasted to the more common 'normative' scores, which are centered about the variable means. The ipsative transformation produces a set of scores with the property that the sum of the variables for each individual equals a constant. The applicability of the ipsative common factor model to sets of variables with ipsative properties which result from the measurement procedure, rather than an ipsative transformation of nonipsative variables, is discussed. Based upon the assumption that there exists a hypothetical nonipsative set of measures, then it is possible to apply the ipsative common factor model to such measures; however, there is no direct way to evaluate the measures' validity within this set of assumptions. The evaluation of the applicability of the ipsative common factor model in these cases requires (1) an examination of the differences in the response processes involved in producing data with ipsative and nonipsative properties and (2) the specification of plausible alternative models for sets of ipsative variables that represent both the latent content of the variables and their ipsative properties. There is also some utility in applying the ipsative common factor model to sets of ipsative variables, such as Kohn's measures of parental values, in order to obtain greater experience regarding the descriptive usefulness of the ipsative common factor model. (Author abstract modified)