This paper describes a methodology, called Profile Analysis via Multidimensional Scaling (PAMS), designed to identify major patterns of variables and to study the relationships into which those patterns enter. Patterns of variables are here called profiles. The PAMS procedure has been adapted to characteristics of the National Assessment of Educational Progress (NAEP) data. Statistical Package for the Social Sciences (SPSS) and Statistical Analysis System (SAS) templates to implement the methodology have been or are being developed. Adaptations of the general PAMS approach for the study of course-taking patterns are described. Multidimensional scaling is being used to identify patterns of course-taking in mathematics and science, and preliminary results from these analyses are reported. The direction of future research is outlined. Four tables and 12 figures illustrate the analyses. (Contains 20 references.) (Author/SLD)