When Torgerson's multidimensional scaling model is used in conjunction with the method of tetrads, derived coordinates are based on data which is assumed to be distributed normally. The object of this study was to determine the amount of error contained in derived coordinates when the normality assumption is violated. Torgerson coordinates were derived from various cases of nonnormally distributed data. Derived coordinates were then compared for accuracy to true coordinate values, which were known in each case. The Torgerson model produced highly accurate coordinates in all cases. Since a wide range of distributions was considered, the results appear to be quite generalizable. (Author)