Course Description
Discriminant analysis, multivariate analysis of variance, canonical correlation analysis, and cluster analysis. Relating research questions to methods, conducting computer analyses, interpreting computer printouts, and critiquing analysis reports.
Athena Title
MULVAR MTHDS IN ED
Prerequisite
ERSH 8320
Semester Course Offered
Offered every year.
Grading System
A - F (Traditional)
Course Objectives
1. To acquire an "overall picture" of the myriad multivariate analyses. 2. To know when and why a multivariate analysis should be conducted. 3. To appreciate the concepts of "linear composite" and "latent structure" and the roles they play in multivariate (multi-variable?) analyses and interpretations thereof. 4. To become familiar with various computer output for at least four types of multivariate analyses. 5. To know how to conduct some multivariate analyses using one or more of the statistical packages. 6. To be able to interpret the results of some multivariate analyses. 7. To appreciate some of the research questions that may be addressed via multivariate analyses. 8. To appreciate the desirability of "relating" the design, analysis, and results of a study to a substantive theory. 9. To be able to apply a multivariate analysis to a substantive research problem. 10. To be able to critically read a published article in which multivariate results are reported. 11. To extend and refine your "philosophy" of data analysis.
Topical Outline
Overview of Multivariate Methods Predictive Discriminant Analysis (PDA) I Classification Multivariate Normal Rules Classification Results PDA II Hit Rate Estimation Effectiveness of Classification Rules Selecting and Ordering Predictors Two-Group Classification Nonnormal Rules Reporting Results of a PDA Applications of PDA Multivariate Aanlysis of Variance Group Separation Assessing Effects Descriptive Discriminant Analysis (DDA) Describing Effects Selecting and Ordering Variables Reporting Results of a DDA Applications of DDA Cluster Analysis
Syllabus
Public CV