Most methods for detecting structure in sociometric data involve either continuous spatial representations (e.g. MDS) or discrete hierarchical clustering analysis (e.g. CONCOR). By producing either spatial or clustering representations, these methods can highlight only some of the theoretically interesting group structures. Correspondence analysis, in contrast, can provide either spatial or clustering representations by assigning spatial coordinates to minimize the distance between individuals linked by a sociometric relationship. These scales may then be used to identify individuals' locations in a multidimensional representation of a group's structure or to reorder the rows and columns of a sociomatrix. Unlike many other methods of sociometric analysis, the numerical methods of correspondence analysis also are well understood and the optimization of the goodness-of-fit measure allows an evaluation of a particular model of group structure.