Additional Graduate Courses
COURSE TITLE: Applied Multivariate Statistical Methods for Public Health
COURSE NUMBER: PBHL B653
B653 is an introductory applied multivariate statistics course designed specifically for graduate students with a PhD major in epidemiology (or advanced masters epidemiology students). The course can also be taken by other non-statistician majors, for example, PhD students in other medical sciences and health care professionals. Students are expected to have taken two previous courses in statistics (introductory and intermediate) covering up through t-test, ANOVA, ANCOVA, linear regression, and logistic regression. This course builds on the material taught in the two prerequisite public health statistics courses, PBHL P551 and PBHL P652. The overall objective of the B653 course is to use public health examples while introducing classic multivariate statistical techniques. The course will focus on applications using the SAS software. Very little attention will be given to matrix algebra. Instead, greater importance will be placed on conceptual understanding and interpretations. Basic bivariate statistics, data screening (e.g., missing data, outliers, assumptions, multi-collinearity), and regression will be reviewed. The following classic multivariate techniques will be covered: canonical correlation, MANOVA, MANCOVA, discriminant analysis, principal components analysis, exploratory factor analysis, confirmatory factor analysis, and structural equation modeling (SEM). Two special topics will be introduced but not tested over: (1) mixed linear models for repeated measures analysis and multi-level modeling of clustered data; and (2) analysis of sample survey data, obtained from complex sampling designs, using the SAS SURVEY procedures with sampling weights.