Lecture 9 Farha Hassan

Used to compare the means of two or more populations It was developed by R. A. Fisher

PROBLEM 1 The following is the sales revenue of three salesmen taken on five occasions. Test at 5% level whether there is any difference between the performances of three salesmen.

Observation s
1 2 3

A
5 6 8

B
3 5 2

C
10 13 7

4
5

1
5

10
0

13
17

STEP 1 Frame the hypothesis H0: There is no difference in the performance of the salesmen. HA: There is difference in the performance of the salesmen. STEP 2 Reproduce the table with Column Total

Observation
1 2 3

A
5 6 8

B
3 5 2

C
10 13 7

4
5 TOTAL SS (Treatment)

1
5 25

10
0 20

13
17 60

STEP: 3 Reproduce the table with sum of squares

Observati A ons 1 2 3 4 5 TOTAL 5 6 8 1 5 25

Ya 2 25 36 64 1 25 151

B 3 5 2 10 0 20

Yb 2 9 25 4 100 0 138

C 10 13 7 13 17 60

Yc 2 100 169 49 169 289 776

SS

(Total)

STEP: 4
Use the formula and find out the following: Correction Factor = Y2/N

(A + B + C )2 / N
SS (Total) = Ya 2 + Yb 2 + Yc 2 _ Y2/ N

SS (Treatment) = A/ CN + B/ CN + C/ CN - Y2/ N
SS (Error) = SS (Total) - SS (Treatment)

STEP : 5
PREPARE ANOVA TABLE
Sources of Variances SS (Treat) SS(Error) SS DF C-1 N-C Mean Computed F Tabulated F

SS(Total)

N-1

ANOVA TABLE Sources of Variances SS (Treat) SS(Error) SS 190 140 DF 2 12 Mean 95 11.67 Computed F Tabulated F 8.14 3.88

SS(Total)

330

14

STEP : 6 Compare tabulated value and calculated value. If the Calculated value is Less, Accept the Null Hypothesis and vv..

STEP: 7 Conclude.

Tabulated value is 3.88 And calculated value is 8.14 As calculated value is more than the tabulated value, reject the null hypothesis. Hence we can conclude that , there is difference in the performance of the salesmen.

Problem 2 Set up an analysis of variance table for the following per acre production data for three varieties of wheat, each grown on 4 plots and state if the variety differences are significant.

ANOVA TABLE

PROBLEM 3
A company has deputed A, B, C and D to four different training programmes. Check whether there is any significant difference between the training programmes of A, B, C, and D at 5% level of significance.
Employee s 1 2 3 A 80 90 96 B 70 60 55 C 65 50 58 D 90 89 85

4
5

85
70

85
90

55
40

95
80

ANOVA TABLE Sources of Variances SS (Treat) SS(Error) SS 3570 1798.8 DF 3 16 Mean 1190 112.42 Computed F Tabulated F 10.585 3.23

SS(Total)

5368.8

19

Problem 4 A company has appointed four salesmen, A, B, C and D and has observed their sales in three seasons. Perform ANOVA to check whether there is any significant difference between sales of salesmen during different seasons.
Seasons Summer Winter Rainy A 36 28 26 B 36 29 28 C 21 31 D 35 32

29 29

ANOVA TABLE Sources of Variances SS (Treat) SS(Error) SS DF Mean Computed F Tabulated F

SS(Total)

Ms Farha Hassan Lecture: 10

Problem 1 A company has appointed four salesmen, A, B, C and D and has observed their sales in three seasons. Perform ANOVA to check whether there is any significant difference between sales of salesmen or between sales during different seasons. Salesmen Season s Summe r Winter Rainy A 36 28 26 B 36 29 28 C 21 31 29 D 35 32 29

Step: 1 Frame the Hypothesis

Hypothesis w.r.t Seasons Ho : There is no difference in the means of sales in seasons. H1: Sale is different in at least one of the seasons. Hypothesis w.r.t Sales Ho : There is no difference in the mean of sales by salesmen. H1: Sale of at least one of the salesmen is different.

Seasons Summer Winter

A 36 28

B 36 29

C 21 31

D 35 32

Total 128 120

Rainy
Total

26
90

28
93

29
81

29
96

112
360

Step: 3 Find out Correction factor T2/N 3602/12 = 10800 Step: 4 Find out the following: Sum of squares between treatment( Salesmen) Sum of squares between blocks ( Seasons) Total sum of squares

SS (Treatment) (Salesman) = 902/3+932/3+812/3+962/3 10800 = 42 SS (Blocks) (Seasons) = 1282/4+1202/4+1122/4 10800 = 32 SS(Total) = 362+362+212+352+282 ..+292 10800 = 210 SS (Error) = 210 - ( 42 + 32 ) = 136

Step: 5
Prepare Anova table Sources S.S of Variance SS (Treatment
)

D.F

Mean

Computed F

Tabulated F

C-1

SS (Blocks)
SS (Error) SS (Total)

R-1
(C-1) ( r-1)

N-1

Sources of Variances SS(treatment) Between samples

SS

DF

Mean

Computed

F 42 (c-1) 4-1 3 42/3 14 14/22.67 0.62

Table value of F 4.76

SS (blocks) between seasons

32

(r-1) 3-1 2
(c-1) x (r1) 3x 2 6 n-1 12-1 11

32/2 16
136/6 22.67

16/22.67 0.71

5.14

136

210

STEP : 6 Compare tabulated value and calculated value. If the Calculated value is Less, Accept the Null Hypothesis and vv..

STEP: 7 Conclude.

Tabulated F (3,6,.05) = 4.76 and F (2,6,.05) = 5.14 In case of the Salesmen, calculated F is less than table value, hence we accept Ho. In case of the Seasons, calculated F is less than table value, hence we accept Ho. Thus, we can conclude that there is no difference either between the sales of the Salesmen, or between the sales of the Seasons.

Problem 2
Set up an analysis of variance table for the following two-way design results:
Per Acre Production Data of Wheat (in metric tonnes)

Also state whether variety differences are significant at 5% level.

Problem 3
Test the following observation with Two Way Anova

Salesme n Area P A 10 B 15 C 14 D 20

Q
R S

20
30 40

18
16 12

16
18 20

16
18 14

Sources of Variances SS(treatment) Between samples

SS

DF

Mean

Computed

F 229.18 (c-1) 3 76.39 1.539

Table value of F 3.86

SS (blocks) between seasons

112.19

(r-1) 3
(c-1) x (r1) 9 n-1 15

37.39

0.753

3.86

446.56

49.61

787.94

Problem 4
Test the following observations in Two Way Anova

TEACHI STUDEN TS
PERFOR MANCE IN EXAM

NG ME

THODS

A
P Q R S 20 16 18 15

B
8 10 12 16

C
14 16 18 20

D
20 16 18 14

Sources of Variances SS(treatment) Between samples

SS

DF

Mean

Computed

F 93.69 (c-1) 31.23 3 3.51

Table value of F 3.86

SS (blocks) between seasons

9.68

(r-1) 3
(c-1) x (r1) 9 n-1 11

0.363
3.229

3.86

80.06

8.895

183.43