ANSC/TPSS 603
EXPERIMENTAL DESIGN (4 cr)
Syllabus
Last modified: January 4, 2016
Class web page: http://www2.hawaii.edu/~halina/603
This syllabus is for the benefit of the student and does not constitute a contract. The instructor reserves the right to change the course content or the sequence of instruction.
SHORT DESCRIPTION: Design of experiments and variance analyses in agricultural and biological research.
PREREQUISITES: NREM 310 or ZOOL 631
OFFERED: Spring Semester
OBJECTIVES:
1. To develop an understanding of experimental methods and major experimental designs.
2. To learn the proper application of experimental designs in agricultural research.
3. To learn appropriate analytic techniques for research data.
TOPICS COVERED:
1. The scientific method, Steps in conducting research
2. Review of basic concepts, student t calculator, tests of significance
3. Experimental Unit, Randomization, Completely randomized design
4. Blocking, Randomized complete block design, latin square, cross-over design, number of replicates
5. Mathematical models, estimators of population parameters
6. Planned F tests and orthogonal comparisons
7. Mean separation and multiple range tests
Duncan's multiple range test - multiply table values times square root (MSE/r)
8. Factorial treatment arrangements
Factorial example in Latin square design
9. Split plots, strip blocks and split split plots
Split plot layouts, Split plot example
10. Analysis of combined experiments
Bartlett's test for homogeneity of variance
11. Sampling
12. Correlation and regression
Regression example: Part 1 , Part 2
13. Covariance analysis , Example
14. Curvilinear relations and multiple regression
15. Failures of the assumptions, transformations
Assignment 11, Answers, Excel formulas
16. Response surface models, fractional replication
18. Incomplete blocks and lattice designs
20. Design and installation of experiments
21. Data management
23. Multiple Measurements, Principal Component Analysis, Cluster Analysis
24. Ethics in Research
On Being a Scientist: Responsible Conduct in Research
25. Review for Midterm 1, Answers, Midterm 1 Answers
Review Problems 1, 1a, Set 2, Set 3
TEXT: Mead, R., R.N. Curnow and A.M. Hasted. 2002. Statistical Methods in Agriculture and Experimental Biology. Third Edition. Chapman & Hall.
LAB TEXT: Brewbaker, J.L., 2000. Experimental Design on a Spreadsheet, Version 3.1. Provided as links in the Lab Schedule.
LAB FEES: A lab fee may be charged to cover the cost of the SAS program. This is estimated at $38.
INSTRUCTOR:
Aleksandra Stein
Department of Human Nutrition, Food and Animal Sciences
Ag Sciences 216
Phone: 956-7095
E-mail: aleksandra.stein@gmail.com
Office hours: by appointment
Teaching Assistant:
Ag Sciences 314V
E-mail: amitks@hawaii.edu
Office hours:
CLASS TIMES AND ROOMS
Lectures: MWF 5:30 a.m. to 6:20 p.m., Ag Sciences 204
Laboratory: W 1:30 to 4:20 p.m., Computer Lab, Ag Sciences 215
GENERAL CONDUCT OF COURSE
1. The course generally follows the text for much of the material; however, reading assignments will also be given in Little and Hills, 1978, Agricultural Experimentation and Gomez and Gomez, 1984, Statistical Procedures for Agricultural Research. The sections on sampling and soil heterogeneity are particularly good in Gomez and Gomez. There will be some reading assignments and references in books that are available at Hamilton Library.
2. The class and laboratory will consist of explanations and examples of designs and their analysis. Laboratory and lecture regular homework will be assigned and is due in one week unless otherwise directed. Problems will be corrected, graded and returned to you.
3. We will be using spreadsheets with Excel as well as SAS for laboratory problems. Bring a jump drive to laboratories to save your work.
4. Exams and grading:
2 Midterm exams = 40% of the grade
Assignments and labs = 20% of the grade
Proposal project = 5% of the grade
Final exam = 35% of the grade
Final: Take-home, due Wednesday, May 11, 6:30 p.m.
The exams will include both lecture and laboratory material. There will be some announced and unannounced quizzes in both sections of the class.
5. An electronic calculator is useful but not required for the course. In order for you to become proficient in using the hand calculator to solve statistical problems, you should have one that you can always use. The features on the hand calculator that may be useful for this course are the built-in programs for calculating the mean, standard deviation, sum of squares, correlation, linear regression and calculation of new points.
6. The following are recommended as supplementary references and are available at Hamilton Library.
Cochran, W. G., and G. M. Cox. 1957. Experimental Designs, 2nd Ed. John Wiley and Sons, NY. Call No. Q180.A1 C6 1957
Fleming, M. C., and J. G. Nellis. 1994. Principles of Applied Statistics. Routledge, London and New York. Call No. QA 276.12.F585
Gomez, K. A., and A. A. Gomez. 1984. Statistical Procedures for Agricultural Research, 2nd Ed. John Wiley and Sons, NY. Call No. S540.S7 G65 1984
Little, T. M. & F. J. Hills. 1978. Agricultural Experimentation: Design and Analysis. John Wiley and Sons. New York. Call Number: S540.A2 L57
Moore, D. S. and G. P. McCabe. 1993. Introduction to the Practice of Statistics, 2nd Ed. Call No. QA276.12 .M65 1993
Snedecor, G. W., and W. G. Cochran. 1980. Statistical Methods, 7th Ed. Iowa State Univ. Press. Ames, IA. Call Number: QA276.12 .S59 1980
Triola, M. F. 1995. Elementary Statistics, 6th Ed. Addison Wesley Publishing Co. Reading, MA. Call Number: QA276.12 .T76 1995
GUIDELINES FOR ASSIGNMENTS
1. Carry out tests of significance and state conclusions wherever possible.
2. Include units.
3. Label graphs and axes.
4. Do your own work, then compare answers with your fellow students. If your answers are different, try to find and fix the problem before handing in the assignment.
5. Assignments are accepted by laulima drop box, in person, or dropped off in Ag Sci 314T.
6. Assignments need not be typed, but must be neat, well-organized and easy to read. Answers that are hard to decipher will be given a 0.
7. Show your work. Marks are given for partially correct answers.
8. We will make every effort (but no promises) to return graded assignments during the next class period. Once the graded assignments are returned, late assignments will not receive any marks, but will still be corrected for educational purposes.
9. If you are unable to hand in an assignment on time, an extension may be granted providing you contact the instructor before the due date.