The main aim of experimental design is the observed variance in a particular variable is partitioned into components attributable to different sources of variation. It provides a statistical test of whether or not the means of several groups are equal. Comparisons of mean squares, along with an F-test allow testing of a nested sequence of models.

After completing this Course, student should

Describe or explain the variation of information under condition that are hypothesized to reflect the variation,

Include the establishment of validity, reliability and reliability and replicability

Achieving appropriate levels of statistical power and sensitivity

Understand comparative design are Important to choose between alternatives with narrow scope, suitable for initial comparison

Experimental design has long enjoyed the status of being the most used statistical technique in statistical, psychological as well as other field of research. It is probably the most useful technique in the field of statistical inference.

Mathematical relationship which relates changes in a given response to changes in one or more factors may learn from statistical model.

Design of experiment  is computationally elegant and relatively robust against violations of its assumptions

Fractional Factorial and Main Effect Plan: Review of factorial experiment, confounding, fractional replication and related plan. Construction of plan with factors at 2 levels. Orthogonal arrays of strength 3 with factors at 2 levels. Orthogonal main effect plans factors at 3 and other levels. Mixed factorial experiment, Orthogonal main effect plans of size 2 ´ Sr. Analysis of orthogonal plans.

Weighing Design: Complete block design as weighing design. Two pan weighing design from BIB design. Two associate PBIB designs as one pan weighing design. Weighing design from truncated BIB design. Efficiency.

Lattice Design: Balanced lattices. Partially balance lattices. Rectangular lattices. Cubic lattices. Lattice squares- description. Statistical analysis with different replications.

Multivariate analysis of variance (MANOVA): Introduction. Omnibus MANOVA tests.  Analysis and interpreting MANOVA. Causal models underlying MANOVA. Complex design.

Nested Design: Introduction, two stage nested design, three stage nested design.

Response Surface Design: Introduction, first order design, second order design, method of steepest ascent, difference between response surface design and usual design.

1)

Cohran and Cox (2000): Experimental Design, John Wiley, N.Y. [Cox with R]

2)

Federer, W.T. 1955. Experimental design, Mcmillan, New York.

3)

Fisher, R.A.(1995): Design of Experiment, Hafner, N.Y. [Gupta, Lindman]

4)

John and Quenouille(1977): Experiments: Design and Analysis, 2nd ed., Charles Griffin, London. [John]

5)

Kutner, M. H., Nachtsheim, C. J., Neter, J. and Li, W. (2005). Applied Linear Statistical Models. 5th Edn., McGraw-Hill, Irwin. [Solution]

6)

Montegomery, D.C. (2005): Design and Analysis of Experiment, John Wiley, N.Y. [Dean]

7)

Steel, R.G.D and J.H. Torrie. 1980. Principles and procedures of statistics, 2nd edition, McGraw-Hill Book Co. New York. [Lawson]