PREVIOUS LECTURES DELIVERED.

EXPLORATORY

LITERATURE SURVEY EXPERT SURVEY CASE STUDY

DISCRIPTIVE AND DIAGNOSTIC

EXPERIMENTAL DESIGN

PRE EXPERIMENTAL TRUE EXPERIMENTAL QUASI EXPERIMENTAL STATISTICAL

QUASI-EXPERIMENTS: TIME SERIES DESIGNS

O1 O2 O3 O4 T O5 O6 O7 O8

Pre-observations to establish a baseline

A treatment intervention Post-observations to establish new baseline EX

MULTIPLE TIME SERIES

3

The basic differencen between this design and earlier is that with an

experimental group which is exposed to treatment, a control is introduced.

STATISTICAL DESIGN
4

Statistical design permit the researcher to measure

and eliminate the effect of extraneous variables. In statistical design BLOCKING FACTOR is introduced. It is the extraneous variable which researcher is able to isolate and eliminate its effect.

STATISTICAL DESIGN TYPES

5

1) Completely randomized design( C. R Design)

- Involves two principles:- principle of replication and -

principle of randomization. Subjects are randomly assigned to experimental treatments Simplest possible formal design One way ANOVAApplied for uncontrolled extraneous factors.

- Eg: 10 test units and two treatment A and B. we want to

give treatment to 5.Every possible group of 5 subjects are selected.

TYPES OF CRD
6

I Two group simple randomized design

- Selection of experimental and control group randomly. - Two groups are then subject to different treatments. - It does not control the extraneous variable

DIAG. FORM
7

Conti.
8

II Random Replication Design.

Extension to the two group simple randomized design, incorporation of repetition. - Extraneous variable controlling. - Each group is assigned equal no of items. - Test unit and administer of treatment is also randomly selected.
-

- EX

DIAG. FORM
9

10

STATISTICAL DESIGN
11

2) Randomized Block design ( R. B. design) - Improvement over C. R. design.

- Principle of Local control can be applied with the other two -

principles of experimental design. Subjects are first divided into groups or blocks. No of subjects in blocks = No of treatments Each treatment appears same no of time in each block. One subject in each block is randomly assigned to treatment. Use of ONE WAY ANOVA and F test

EXAMPLE
12

Eg: Suppose four different forms of a standardised test in statistics were given to each of five students (selected one from each of the five I.Q. blocks) and following are the scores which they obtained.

LATIN SQUARE DESIGN

13

Very frequently used in agricultural research. LS is used when there are two major extraneous

factors. For example if we want to judge the effect of five different varieties of fertilizers on the yield of wheat. The two extraneous factors here are fertility of land and varying seeds. Field is divided into 5*5 parts and each extraneous factor is taken at one axis

LATIN SQUARE DESIGN

14

The following is a diagrammatic form of such a design in respect of, say, five variety of cosmetic creams and the liking/variety
X1 X2 X3

A
O P B

B
P B C

C
B O F

D
C F P

E
F C O

X4
X5

C
F

F
O

P
C

O
B

B
P

The above diagram clearly shows that in a L.S. design the field is divided into as many blocks as there are varieties of fertilizers and then each block is again divided into as many parts as there are varieties of fertilizers in such a way that each of the fertilizer variety is used in each of the block (whether column-wise or row-wise) only once. The analysis of the L.S. design is very similar to the two-way ANOVA technique

EVALUATION
15

The merit of this experimental design is that it

enables differences in fertility gradients in the field to be eliminated in comparison to the effects of different varieties of fertilizers on the yield of the crop But this design suffers from one limitation, and it is that although each row and each column represents equally all fertilizer varieties, there may be considerable difference in the row and column means both up and across the field.

FACTORIAL DESIGN
16

This method is used in experiments where the effects

of varying more than one factor are to be determined This is specially useful in several economic and social phenomena where there are large number of factors affect a particular problem. Two types I simple factorial design - Effect of varying two factors on the dependent variable II Complex factorial design more than two factors.

SIMPLE FACTORIAL DESIGN

17

Two variables control variable and experimental variable. Extraneous variable to be controlled by homogeneity is called control

variable and the independent variable ,which is manipulated is called experimental group Four cells in which sample is divided. Experimental variable treat A Treat B Control Variable level1 I II level 2 III IV - Randomly assigned and means are obtained for control variable and Experimental variables. - One can examine the interaction between treatment and level. This enables researcher to evaluate the combined effect or the interaction effect of two or more variables simultaneously

EXAMPLE

18

Training Treat A Treat B 15 35 25 23 30 26.5 19 32.5

Control level I (Low) (Intelligence) level II (High)

Treat ments Control level(Intelligence)

Treatment and level are dependent on each other from graph

19

treatment

No relationship between treatment and intelligence

Control(intelligence)

II COMPLEX FACTORIAL DESIGN

Experiments with more than two factors at a time
20

involves the use of complex factorial design. Treatment and control variable both have different levels. Experimental variable Treat A Treat B level 1 level2 level 1 level 2 Control level1 I Variable level2 II III IV IV VI VII VIII

21

To determine the main effects of the experimental

variable, the researcher must necessarily combined mean of Cell I, II, III, IV Advantages of factorial design They provide equivalent accuracy Economic The determination of interaction effects is possible in case of factorial design.