Description:

Predictive statistical models such as linear regression and analysis of variance and covariance are basic workhorses of modern empirical research. The simplest and most commonly used versions of these models are designed for the analysis of independent observations. When multiple observations are gathered on the same units or units are sampled in groups, observations are no longer independent and methods that take into account correlations among dependent observations are generally required in order to make valid inferences. This webinar will provide an introductory overview of methods for analysis of correlated data, beginning with simple tests of equality of means and touching on more extensive methods for the predictive analysis of data from repeated measurements and clustered or hierarchical/multilevel data.  

Prerequisites: Participants should have a basic background in statistics including exposure to linear regression and analysis of variance models. 

Registration Required.