Multidimensional Scaling

Multidimensional Scaling

Multidimensional scaling (MDS) can be considered to be an alternative to factor analysis. In general, the goal of the analysis is to detect meaningful underlying dimensions that allow the researcher to explain observed similarities or dissimilarities (distances) between the investigated objects. In factor analysis, the similarities between objects (e.g., variables) are expressed in the correlation matrix. With MDS, you can analyze any kind of similarity or dissimilarity matrix, in addition to correlation matrices.

The final step of the analysis is the interpretation of dimensions. The actual alignments of the axes from the MDS analysis are random, and can be revolved in any direction. A first step is to generate scatterplots of the objects in the different two-dimensional planes. Three-dimensional solutions can also be demonstrated graphically; but, their analysis is a bit complex. In addition to "meaningful dimensions," one can also look for clusters of points or particular arrays and configurations.