Special Topics in Survey Methodology: Statistical Methods for Analysis of Complex Samples in Public Health
Mathematical statistics course at the Master’s level (e.g., UMD STAT/SURV 420, EPIB 652, or equivalent.) If you are unsure about your qualifications for the course, please contact the instructors.
With large health surveys becoming increasingly available for public use, there is a growing demand to analyze data from surveys to address scientific and programmatic questions. This course provides statistical techniques for use in survey analysis, making health surveys accessible to statisticians, biostatisticians, epidemiologists, and health researchers. The course is organized by important designs commonly employed in public health studies. Topics include cross-sectional studies, cohort studies and population based case-control studies, and special topics in population-based studies, such as health disparities, genetic studies with complex designs, graphs with survey data, ROC curves, population attributable risk/fraction, etc. Theory and methods of survey analysis along with Monte Carlo simulation results and real-world applications using up-to-date information from the literature will be presented and discussed. Derivation of formulae will be presented wherever necessary to explain some of the advanced topics. The course emphasizes data analysis via modern computer methods and R freeware packages that are introduced and used throughout the course.