Abstract
Multidimensional scaling (MDS) is a set of data-analytic tools for deriving a graphical representation of objects in a multidimensional space based on proximity relations among them. By the graphical representation, we gain intuitive understanding of the regularity governing the relationships among the objects. In this article, we introduce the basic concept and models of MDS along with its essential ingredients such as distance models, proximity data, fitting criteria, dimensionality selection. To illustrate the use of MDS, three useful MDS procedures (simple MDS, individual differences MDS, and unfolding analysis) are presented with empirical examples.
Original language | English |
---|---|
Title of host publication | International Encyclopedia of the Social & Behavioral Sciences: Second Edition |
Publisher | Elsevier Inc. |
Pages | 34-39 |
Number of pages | 6 |
ISBN (Electronic) | 9780080970875 |
ISBN (Print) | 9780080970868 |
DOIs | |
State | Published - Mar 26 2015 |
Keywords
- Euclidean distance model
- Graphical representation
- Ideal points
- Individual differences
- Multidimensional scaling
- Preference data
- Proximity data
- Unfolding analysis