This course presents concepts and accepted methodologies for the design and analysis of experiments and their applications in treatment comparison, experimental screening, and response optimization. These include blocked and (fractional) factorial designs, response surface methodology for related factors, and optimality criteria to compare competing designs. Concepts and methods are illustrated in various examples drawn from medical, agricultural, and industrial contexts. The last part of the course covers principles of survey sampling, where estimators of population means and proportions along with associated uncertainty measures are derived for different sampling schemes.

20 lectures (2 each week in Weeks 1-10 of Semester 2)

4 tutorials (fortnightly)

STATS5017 - Principles of Designed Experiments and Surveys (Level M)

90-minute, end-of-course examination (100%)

To provide an introduction to the statistical aspects of designing experimental studies; to introduce associated methods of statistical analysis; to discuss optimality concepts for the design of experiments; to review basic sampling techniques and their statistical analyses.

By the end of this course students will be able to:

■ Describe the difference between an experiment and an observational study - Explain the principles of randomisation, replication, blinding, and stratification, and understand how

■ they apply to practical problems.

- Understand the general theory of factorial and block designs and to find appropriate designs for specific applications. - Evaluate designs using common optimality criteria and used them to critically compare competing designs. - Apply and derive estimators of population means and their standard errors under a range of random sampling designs, including simple, stratified, and cluster sampling, and use these to construct efficient sampling strategies for practical survey contexts. - Apply theory and methods to a variety of applications.