This course focuses on multivariate research design, statistical methods and data analysis. Emphasis will be placed on providing a fundamental understanding of the multivariate quantitative methodological techniques used in the empirical social, behavioural, and educational sciences. Emphasis is placed on encouraging you to learn to communicate in the language of multivariate data analysis. Communication is at the core of this course. Computation is meant to aid in communication.
Emphasis will be placed on data analysis techniques. An awareness of the common pitfalls and misconceptions of the various techniques and the fundamental assumptions you need to make to apply these methods will be emphasized. The multivariate general linear model and other multivariate methods will be discussed both from a geometric and matrix algebraic formulation.
Topics covered include:
- Why do we need multivariate statistics in our research repertoire — or do we? Along the way, a brief review of some univariate and bivariate statistics, including multiple regression modeling.
- Some rudiments of matrix algebra in the context of statistical analysis and some fundamental statistical theorems and results for multivariate analysis.
- Analysis of experimental data for designs involving completely between or within factors, mixed or split-plot designs, nested designs, and fixed versus random factors. Repeated measures ANOVA.
- Analysis of covariance.
- Brief discussion of canonical correlation.
- Multivariate analysis of variance and covariance (MANOVA/MANCOVA), step-down analysis, post-hocs to MANOVA.
- Discriminant Function analysis.
- Brief discussion of principal components and factor analysis.
Prerequisites: Successful completion of EPSE 592 and EPSE 596 or equivalent graduate-level courses involving regression and the analysis of experimental data. It will be very helpful to have taken (or be taking) a course in regression because some of those concepts will be built on in this class. If you are unsure whether your background is sufficient, please contact the instructor.