Applications of Multidimensional Scaling to Research in the Health Professions

Publisher Website
Google Scholar
Abstract
Multidimensional scaling (MDS) is a procedure for transforming similarity data such as correlations into a graph or map. Although MDS is similar to factor analysis in function and purpose, it is more flexible and often produces results that are easier to interpret. This article discusses MDS and its potential use for research and evaluation in the health-related professions. The first half of the article presents an overview of MDS, the types of data that can be used as input, and issues related to interpretation of results. The second half describes various applications of MDS, including the development of a working taxonomy of nursing specialty practice, the identification of the dimensions underlying physician performance, and the classification of different types of psychotherapies examined in the metaanalysisperformedby Smith and Glass (1977). The article concludes with a brief discussion on how MDS relates to factor analysis, discriminant analysis, and cluster analysis.

This publication has 26 references indexed in Scilit: