Types of Multivariate Analysis

There are three types of multivariate analysis that you will use to find associations and correlations in your data:

• Logistic regression
• Multivariate linear regression
• Cox's proportional hazards survival analysis

Logistic regression is used when your outcome variable is categorical.

When your outcome has two categories, such as gender [male; female] or survival status [dead; alive], then you should use binary logistic regression.

If your outcome has more than two categories, you need to decide whether there is an order to the categories. If there is, such as when your outcome has categories of size [small; medium; large] or grade [A; B; C; D; E], then you should use ordinal logistic regression.

Alternatively, if your categories have no order, such as nationality [British, American, Spanish, …] or religion [Hindu, Buddhist, Jedi], then you should use a nominal logistic regression (also known as multinomial logistic regression).

Multivariate...