Research Methodology and Experimentation
BIOMET 7000WT
30th of June to 11th of July, 2003

Faculty of Sciences

Subject Co-ordinator: Trevor Hancock

Co-ordinating Department: BiometricsSA

Timetable: Please visit The University of Adelaide Online Timetable. Lectures and tutorials coincide with Agricultural Experimentation (BIOMET 3000WT), and students may also attend lectures and tutorials for Biometry (BIOMET 2000WT).

Lectures will be held on Mondays to Fridays at 9am to 9:50am, 10:10am to 11am, 2pm to 2:50pm, and 3:10pm to 4pm in Lecture Room 3.

Tutorials will be held on Mondays to Fridays at 11am to 1pm, and 4pm to 5pm in the PC Computing Laboratory 2.

General Outline

The design and analysis of experiments and surveys is fundamental to developments in agricultural, biological, environmental, aquatic, food and wine sciences. Proper design of such studies will ensure that meaningful analysis is possible, and so this subject covers the principles of design of experiments and of surveys. In addition methods of analysis, such as regression and analysis of variance, are also covered.

The ideas and methods are combined with case studies. This demonstrates the vital role of careful design and informative analysis in the "Scientific Method" as practised in most disciplines of science.

The subject uses the statistical package, GenStat for Windows, and the spreadsheet Excel.

Aim of subject

To acquaint the student with a variety of statistical methods and to introduce them to the statistical computing package, GenStat for Windows.

By the end of this course students should possess an appreciation of the basic concepts of experimental design.

Further they should be able to decide which method, of those included in this course, is most appropriate for analysing a particular set of data and be able to carry out that analysis using GenStat (Excel or a hand-calculator).

Content

• Introduction to the computing environment and GENSTAT for WINDOWS
• Simple linear regression (2 examples of regression)
• Regression through the origin (Calibration problem)
• Analysis of residuals
• Polynomial regression
• Multiple linear regression
• Sampling - Simple random, Systematic, Stratified, Multi-stage, Sampling plan
• Designed experiments - Analysis of Variance
  • Completely randomized design (CRD) - (or one-way ANOVA(No Blocking))
  • Theory of least squares estimates
  • Randomized Complete Block Design (RCBD) - concepts of blocking and randomization
  • Two-way Classification
  • Factorial Designs
  • Split Plot Designs
  • Latin Squares and Graeco-Latin Squares;
  • Analysis of Covariance;
• Review of design - Randomization of treatments;
• Nonparametric statistics - (central tendency and dispersion)
  • Mann-Whitney U Test
  • Wilcoxon Matched Pairs Signed - Rank Test
  • Kruskal Wallis Analysis of Variance
  • Friedman Two-way Analysis of Variance
  • Spearman's Rank Correlation Coefficient
• Multiple Comparisons, Simultaneous Confidence Intervals, Linear Comparisons and Orthogonal Polynomials;
• Distinct regression lines versus parallel regression lines;
• Transformations;
• Analysis of categorical data - Contingency tables, Fisher's Exact Test ;
• Generalized Linear Models - Poisson & binomial error structures;
• Probit Analysis;
• Depending on time available some other topics may be covered;

Throughout the course a selection of research projects will be discussed to illustrate important steps involved, i.e. aim, design, collection and analysis and interpretation of data.

Text Books

There is no set text book.

Reference Books

• Cochran, W.G. and Cox, G.M.(1957) Experimental Designs. (Wiley)
• Colquhoun, D.(1971) Lectures in biostatistics. (Clarendon)
• Cox, D.R.(1958) Planning of Experiments. (Wiley)
• Draper, N.R. and Smith, H.(1998) Applied Regression Analysis (3rd Ed.). (Wiley)
• Lane, P.W. and Payne, R.W.(1998) GENSTAT for WINDOWS: an Introductory Course (4th Ed). (Lawes Agricultural Trust - Rothamsted Experimental Station)
• McConway, K.J., Jones, M.C. and Taylor, P.C.(1999) Statistical Modelling using GENSTAT. (Arnold)
• Mead, R. and Curnow, R.C.(1983) Statistical methods in agriculture and experimental biology. (Chapman and Hall)
• Moore, D.S. and McCabe, G.P.(1998) Introduction to the Practice of Statistics (3rd Ed). (W.H. Freeman & Company).
• Siegel, S. and Castellan, N.J.(1988) Nonparametric statistics for the behavioural sciences (2nd Ed). (McGraw-Hill)
• Snedecor, G.W. and Cochran, W.G.(1980) Statistical methods (7th Ed). (The Iowa State University Press).
• Sokal, R.R. and Rohlf, F.J.(1995) Biometry (3rd Ed). (W.H. Freeman & Company)
• Steel, R.G.D. and Torrie, J.H.(1960) Principles and Procedures of Statistics. (McGraw-Hill).

Assessment

The assessment in this subject is as follows:

• Individual Assignment: 40%
• Final Exam: 60%

Tutorial exercises will be handed out. Students are expected to work on the exercises during the tutorial period. Any student who does not attend at least 75% of the tutorials may be precluded from the examination.

The Individual Assignment, The Individual Assignment, which forms a further 40% of the assessment, will involve the design, collection or generation of data, and analysis of results for an experiment of the individual student's choice. The final written report including Genstat analysis is due by 5:00pm on Friday the 18th of July 2003.

All work covered in the course will be examined in early August and this examination will constitute the other 60% of the final assessment. This Final Exam is an "Open Book" exam which means you can (in addition to pens, pencils, rulers, calculators etc) take your Lecture Notes, Hand-outs, and tables into the exam room. However you MUST NOT TAKE into the exam the Tutorial Exercises and Solutions handed out during the semester or your solutions to those weekly tutorials or your solutions to previous Exam Papers.

The students overall performance will be recorded as a scaled mark out of 100 and results summarized using the following grading scheme :

• Higher Distinction - 85% or greater
• Distinction - 75% to 84%
• Credit - 65% to 74%
• Pass - 50% to 64%
• Conceded Pass - 45% to 49%
• Fail - less than 45%

Students can arrange with the Subject Co-ordinator to review their examination paper. Any unresolved discussion concerning the assessment will be referred to the Examiner's Committee for this subject.

• Special Needs: Students with special needs are requested to inform the subject co-ordinator, during the first week of the semester, so that appropriate arrangements are expedited.
• General Rules and Plagiarism: For general rules and procedures see the "Student Guide and Timetable" and give particular attention to the item on plagiarism. Copying of another student or author's work without proper acknowledgement is not accepted in this subject.

Examination Information

The date, time and location of the Final Examination will be posted in the Foyer of the Undergraduate Teaching Wing at the Waite Institute. Students are required to check these for themselves.

For past Examination Papers visit http://www.library.adelaide.edu.au/eres/Biometry/

Supplementary Examination Policy

Revised procedures for Supplementary Assessment were approved by Academic Board on 28 July 1999. Accordingly Faculty Board at its meeting on 21 February 2000 approved the procedures below for adoption in all subjects taught by departments of the Faculty.

1. The offer of supplementary assessment is solely at the discretion of the examiner(s) and Head of the relevant department. No student is automatically entitled to supplementary assessment and the University is under no compulsion to offer supplementary assessment in any form.
2. For each subject a committee of the examiner(s) and the Head of Department (or nominee) will consider the final results. The ultimate decision lies with the Head of Department.
3. Where a student has not met the requirements to pass a subject, any form of supplementary assessment (including an exam) can be used which is judged to be fair and equitable for the student and subject concerned.
4. If a student applies for supplementary assessment based on medical or compassionate grounds the Examiners Committee will consider the student's overall performance in the subject. If justified, an appropriate adjustment will be made to the final mark or some supplementary assessment set.

Top