Multivariate Data Analysis Chapter 10 - Multidimensional Scaling多元数据分析10章-多维标度.pptVIP

Multivariate Data AnalysisChapter 10 - Multidimensional Scaling Chapter 10 What Is Multidimensional Scaling? A Simplified Look at How Multidimensional Scaling Works Comparing MDS to Other Interdependence Techniques Individual As the Unit of Analysis Lack of a Variate A Decision Framework for Perceptual Mapping Stage 1: Objectives of Multidimensional Scaling Key Decisions in Setting Objectives Identification of All Relevant Objects to Be Evaluated Similarities Versus Preference Data Aggregate Versus Disaggregate Analysis A Decision Framework for Perceptual Mapping (Cont.) Stage 2: Research Design of Multidimensional Scaling Selection of Either a Decompositional (Attribute-free) or Compositional (Attribute-based) Approach Decompositional or Attribute-free Approach Compositional or Attribute-based Approach Selecting Between Compositional and Decompositional Techniques Objects: Their Number and Selection Nonmetric Versus Metric Methods Collection of Similarity or Preference Data Similarities Data Comparison of Paired Objects Confusion Data Derived Measures Collecting Preference Data Direct Ranking Paired Comparisons Preference Data Versus Similarity Data A Decision Framework for Perceptual Mapping (Cont.) Stage 3: Assumptions of Multidimensional Scaling Analysis Stage 4: Deriving the MDS Solution and Assessing Overall Fit Determining an Objects Position in the Perceptual Map Selecting the Dimensionality of the Perceptual Map Incorporating Preferences into Multidimensional Scaling Ideal Points Positioning the Ideal Point Internal Analysis External Analysis Vector Versus Point Representations Summary Decision Framework for Perceptual Mapping (Cont.) Stage 5: Interpreting the MDS Results Identifying the Dimensions Subjective Procedures Objective Procedures Selecting Between Subjective and Objective Procedures Stage 6: Validating the MDS Results Correspondence Analysis A Simple Example of Correspondence Analysis Calcu