Clustered Data Designs and Applications

Catalog description

Data collected within clusters are not generally independent and analysis strategies are needed to accommodate this construct. Focus will be on identifying clustered design structures, such as: patients within clinics and measurements over time on the same patient. The course will include design, sample size, and power implications for clustered studies and mechanisms for the analysis and estimation of the factors of interest, including the ICC components. Prerequisites: BIOSTAT 201, BIOSTAT 202, BIOSTAT 204, BIOSTAT 205, or permission of the Director of Graduate Studies Credits: 2