cr210

Statistical Methods And Experimental Design I

CR210

The following course content has been agreed upon by the relevant departments, and is to be taught for the Departments of Animal Science and Crop Science, in the first semester of year two,

COURSE ORGANIZATION

Lectures: 39 x 1hr

Practicals: 13 x 3 hr

Grading:

Continuous assessment: 40%

Assignments
Practicals
Quizzes

Final Examination: 60%

One (closed book) hour written paper.

COURSE TEXT BOOKS:

Until such times when textbooks are available at affordable prices, the text suggested will be “Experimental Design” by Sr. Jane Canhao. Other texts are available and students may
buy them for reference if they so wish.

COURSE CONTENT:

1. Multiple Regression

1.1 First Order With At Least Two independent Variables

Model, applications and meaning of the regression parameters
Estimation of the model parameters: Least Squares method
Checking the model error assumptions
F Test for the regression relation
Inferences about sets of parameters: Reduced and full model approach
Inference concerning the mean response and the response
Transformations to achieve linearity, normality and constant variance
Multiple regression using MINITAB, SAS and/or GENSTAT

1.2 Polynomial Regression With One independent Variable

Model, applications and meaning of the regression parameters
Estimation of the model parameters: Least squares method.
Checking the model error assumptions
F Test for the regression relation
Inferences about sets of model parameters: Reduced and full model approach
Inferences concerning the mean response and the response
Transformations to achieve linearity, normality, and constant variance.
Polynomial regression using MINITAB or SAS and/or GENSTAT

2. Analysis of variance for some Standard Experimental Designs

2.1 Introduction to Analysis of Variance

One Way ANQVA model; applications, meaning of model parameters.
Logic behind ANOVA: Par titioning the Total Variation into Between Sample Variation(Treatment variation) and Within Sample Varialion (Random variation
Checking model error assumptions
F Test for equality of Population/Treatment/Factor Level means.
Multiple comparisons of Population/Treatment/Factor I evel means: LSD, Scheffe, Tuckey, Bonniferoni
ANOVA using SAS and/or GENSTAT.

2.2 Principles of Experimental Design

Experiment, treatments, experimental units, response
Objectives of replication, randomisation and blocking

2.3 Compietely Randomised Design

Objectives of the experiment; Design (objectives, assumptions, Randomisation, advantages and disadvantages)
Model; Analysis of variance
Multiple comparisons
ANOVA using SAS and/or GEl (STAT

2.4 Completely Randomised Block Design

Objectives of the experiment; Design (objectives, assumptions, Randomisation, advantages and disadvantages)
Model; Analysis of variance
Multiple comparisons
ANOVA using SAS and/or GENSTAT

2.5 Latin Square Design

Objectives of the experiment; Design (objectives, assumptions, Randomisation, advantages and disadvantages)
Model; Analysis of variance
Multip(e comparisons
ANOVA using SAS and/or GENSTAT

2.6 Nested Designs

Objectives of the experiment; Design (objectives, assumptions, Randomisation, advantages and disadvantages)
Model; Analysis of variance
Multiple comparisons
ANQVA using SAS and/or GENSTAT

2.7 Complete Factorial Experiments (at most 3 factors) using Compietelp Randomised, Complete Randomised Block, Latin Square and Nested Designs

Objectives of the experiment; Design (objectives, assumptions, Randomisation, advantages and disadvantages)
Model; Analysis of variance
Multiple comparisons
ANQVA using SAS and/or GENSTAT

3. Analysis of Covariance

3.1 A Completely Randomised Design with One Covariate

Model, applications, meaning of model parameters
Logic behind ANCOVA: Reduction of error variability-
Checking model error assumptions
Multipte comparisons of Population/Treatment/Factor Level means: LSD, Scheffe, Tuckey, Bonniferoni
AflCOVA using SAS and/or GEllSTAT

3.2 A Completely Randomised Block Design with One Covariate

Model, Analysis of covariance, multiple comparisons
ANCOVA using SAS and/or GENSTAT

4. Introduction to Sample Surveys:

4.1 Principles of Sample surveys

Uses and advantages over censuses
Principal Steps in a sample survey

4.2 Simple Random Sampling

Design of the sample survey
Regression estimators
Categorical data analysis

4.3 Stratified Random Sampling.

Design of the sample survey
Inferences about population means, lotais and ratios
Regression estimators
Categorical data analysis

5. Introduction To Time Series Analysis:

(Optional Input from the Dept. of Agricultural Economics)

Back to Crop Sciences Menu Back to Top of Page

Other Menus:
Agricultural Economics
Agricultural Engineering
Animal Science
Soil Sciences

Home