Abstract: The models and methods available to the social scientist are now more numerous and more complex than ever before. One of the most complex family of models presently available is non-metric multidimensional analysis. In multidimensional analysis the investigator seeks a spatial representation of a set of data. Ideally the rank-order of the interpoint distances in the multidimensional space should be identical to the rank order of the variables they represent. The emphasis on non-metric models has been due to a self-conscious attempt by social scientists to accommodate cruder levels of observation in their conceptualization and measurement procedures. The algorithms for multidimensional analysis all use iterative procedures which take some initial spatial representation of the data, assess the degree of fit with a criterion, and then change the configuration of points in ways designed to improve the fit. This iterative process is continued until some arbitrary cutoff is reached.