STAT357 Design and Analysis of Experiments

Updated: 29 March 2012

Credit Points	6
Offering	Not offered in 2013
Intensive School(s)	None
Supervised Exam	There is no UNE Supervised Examination.
Pre-requisites	STAT261 or candidature in a postgraduate award in the School of Environmental and Rural Science or School of Science and Technology
Co-requisites	None
Restrictions	None
Notes	Two lectures and a two-hour combined tutorial/laboratory session per week for on-campus students; off-campus students must have access to the statistical package R and the Gendex design of experiment toolkit which is mounted on turing(www.designcomputing.net/gendex).
Combined Units	None
Coordinator(s)	Robert Murison (rmurison@une.edu.au)
Unit Description	This unit is concerned with the design and analysis of experiments. Topics include factorial designs, confounded and fractional replicated designs (incl Taguchi), incomplete blocks, design optimality. The block designs are analysed as mixed models. Attention is given to the construction of designs by optimal allocation of units to interested effects.
Materials	Textbook information will be displayed approximately 8 weeks prior to the commencement of the teaching period. Please note that textbook requirements may vary from one teaching period to the next.
Disclaimer	Unit information may be subject to change prior to commencement of the teaching period.

Assessment

Assessment information will be published prior to commencement of the teaching period.

Learning Outcomes (LO)

Upon completion of this unit, students will be able to:

understand the new philosophy of designing experiments, i.e. design for the experiment, do not experiment for the design;
design experiments to optimise particular comparisons;
write the statistical model for a designed experiment, analyse the data and report the results;
use the Gendex toolkit to generate experimental designs and use R to analyze the data;
realise the broad use of experiment designs in industry through practice in designing and analysing case studies; and
appreciate the theoretical methods that underpin optimum experiment designs.

Graduate Attributes (GA)

	Attribute	Taught	Assessed	Practised
1	Knowledge of a Discipline Through learning experiment design graduates will recognise the attributes and limitations of different designs.
2	Communication Skills Graduates will communicate their results with formal setting out of solutions, augmented by clear sentences which explain the results and the meanings and importances of terms in the algebra. The explanation has to be relatively free of technical jargon and where such is necessary, its meaning be fully explained. Students will develop skills in drawing pen-pictures that explain features of graphs.
3	Global Perspectives Graduates will appreciate that statistics is ubiquitous by learning how statistical techniques are generic and are driven by properties of the data rather than the application.
4	Information Literacy Graduates will be equipped with terminology so that they can recognise experimental design in diverse settings such as science, business, social science, humanities, etc.
5	Life-Long Learning Solid foundations in experimental design principles and the practice in analysing will allow graduates to confront a non-standard design problem in any discipline.
6	Problem Solving Graduates will be proficient in translating problems expressed in words to an algebraic formulation which allows analysis, performing the analysis mathematically, and expressing the solution in words in the context of the original problem.

Course and Unit Catalogue 2013