A model for test scores called the profile analysis via multidimensional scaling (PAMS) model is described. The model reparameterizes the linear latent variable model in such a way that the latent variables can be interpreted in terms of profile patterns, rather than factors. The model can serve as the basis for exploratory multidimensional scaling analyses to identify major patterns in test scores; or it can serve as the basis for confirmatory structural equations analyses designed to test hypotheses about profile patterns and their associations with other variables. An exploratory multidimensional scaling analysis is described with particular emphasis on the interpretation of the model parameters in terms of latent profile patterns. The analysis is illustrated with vocational interest data. Finally, the exploratory multidimensional scaling analysis is compared to some alternative methods for identifying major profile patterns in data. (Contains 3 tables, 6 figures, and 23 references.) (Author)