DEPARTMENT OF MATHEMATICS

NAME OF THE SUBJECT : STATISTICS &

NUMERICAL METHODS

SUBJECT CODE : MA8452

REGULATION : 2017

UNIT – II : DESIGN OF EXPERIMENTS

 UNIT –II - DESIGN OF EXPERIMENTS
NOTES
Analysis of Variance :
Analysis of Variance is a statistical method used to test the difference between 2 or more means. In
short it is ANOVA
Uses of ANOVA:
• To test the homogeneity of several mean
• It is now frequently used intesting the linearity of the fitted regression line or in the
significance of the correlation ratio
Assumptions of ANOVA
• The sample observations are independent
• The environmental effects are additive in nature
• Sample observation are coming from normal distribution/population
Experimental error:
Factors beyond the control of the experiment are known as experimental error
Aim of the design of experiment:
Aim is to control the extraneous variables so that the result could be attributed only to the
experimental variables
Basic principles of design of experiments:
• Randomization
• Replication
• Local Control
Three essential steps to plan Design of experiment:
To plan an experiment the following three are essential
• A Statement of the objective. Statement should clearly mention the hypothesis to be tested
• A description of the experiment. Description should include the type of experimental material,
size of the experiment and the number of replications.
• The outline of the method of analysis. The outline of the method consists of analysis of
variance
Completely randomized design:
In Completely randomized design the treatments are given to the experimental units by a procedure of
random allocation. It is used when the units are homogeneous.
ANOVA table for One Way classification (CRD)

Source of Sum of Degrees of

Mean Square F- Ratio
Variation Squares freedom

Between SSC
SSC K-1 MSC =
Samples K −1 MSC
FC =
MSE
Within SSE
SSE N-K MSE =
Samples N−K
Two-way classification or Randomized Block Design (RBD)
When data are classified according to two factors one classification is taken column wise and the other
row wise. Such a classification is called two-way classification
 ANOVA table for Randomized Block Design

Source of Sum of Degrees of

Mean Square F- Ratio
Variation Squares freedom
Column SSC MSC
SSC c-1 MSC = FC =
Treatment c −1 MSE
Row SSR MSR
SSR r-1 MSC = FR =
Treatments r −1 MSE
Error (or) SSE
SSE (r-1) (c-1) MSE =
Residual (r − 1)(c − 1)

ANOVA table for Latin Square Design

Source of Sum of Degrees of

Mean Square F- Ratio
Variation Squares freedom
Column SSC MSC
SSC n-1 MSC = FC =
Treatments n −1 MSE
Row SSR MSR
SSR n-1 MSR = FR =
Treatments n −1 MSE
Between SST MSK
SST n-1 MSK = FK =
Treatments n −1 MSE
Error (or) SSE
SSE (n-1) (n-2) MSE =
Residual (n − 1)(n − 2)
ANOVA table for 22 factorial design

Source of Sum of Degree of

Mean Square F- Ratio
Variation Squares freedom
SS A MS A
A SS A 1 MS A = FA =
d.f SS E
SS B MS B
B SS B 1 MS B = FB =
d.f SS E
SS AB MS AB
AB SS AB 1 MS AB = FAB =
d.f SS E
Error (or) SS
SSE 4(r –1) MS E = E
Residual d.f
PROBLEMS:
1. The following table shows the lives in hours of 4batches of electric bulbs. [2015]
1 1610 1610 1650 1680 1700 1720 1800
2 1580 1620 1620 1700 1750
3 1460 1550 1600 1620 1640 1740 1820
4 1510 1520 1530 1570 1600 1680
Perform an analysis of variance of these data and show that a significance tet dose not
reject their homogeneity
Solution:
We subtract 1640 from the given values and workout with the new values of x ij

Batc Ti 2
Ti ni
lives of bulbs ni
hes

1 -30 -30 10 40 60 80 160 - 290 7 12014

2 -60 0 0 60 110 - - - 110 5 2420

3 -180 -90 -40 -20 0 20 100 180 -30 8 113

4 -130 -120 -110 -70 -40 40 - - -430 6 30817

Total -60 26 45364

N = 26 T = 98
T2
C.F = = 369.39
N
T2
TSS = ∑∑ x ij2 − = 1950.62
N

(∑ Ti )2T2
SSC = = 452.25
−
ni N
SSE = TSS − SSC = 1498.36

Source of Sum of Degree of

Mean Square F- Ratio
Variation Squares freedom

Between
SSC = 452.25 h-1=3 MSC = 150.75
Column
15075
= 2.21
6811
Error SSE = 1498.36 N-h=22 MSE = 68.11

Total TSS = 1950.62 N-1 = 25

From the table F0.05 (v1 = 3, v 2 = 22) = 3.05

Calculated F < Tabulated F
Conclution: Hence we accept H 0 the lives of 4 batches of bulbs do not differ significantly.
2. As head of the department of a consumers research organization you have the responsibility of
testing and comparing life times of 4 brands of electric bulbs.suppose you test the life time of 3
electric bulbs each of 4 brands,the data is given below,each entry representing the life time of an
electric bulb,measured in hundreds of hours.

A B C D
20 25 24 23
19 23 20 20
21 21 22 20
Solution:
H0: Here the population means are equal .
H1 : The population mean are not equal.

X1 X2 X3 X4 X12 X22 X32 X42

20 25 24 23 400 625 576 529
19 23 20 20 361 529 400 400
21 21 22 20 441 441 484 400
TOTAL 60 69 66 63 1202 1595 1460 1329

N = Total No of Observations = 12 T= Grand Total = 258

( Grand total )2
Correction Factor = = 5547
Total No of Observatio ns
2 2 2 T2
TSS = ∑ X 1 + ∑ X 2 + ∑ X 3 − = 39
N
(∑ X ) (∑ X ) (∑ X )
1
2
2
2
3
2
T2
( N1 = No of element in each column )
SSC = + + − = 15
N1 N1 N1 N
SSE = TSS – SSC = 39 - 15 = 24
ANOVA TABLE

Source of Sum of Degree of

Mean Square F- Ratio
Variation Squares freedom

Between SSC
SSC=39 C-1=4-1=3 MSC = =5 MSC
Samples C −1 FC =
MSE
Within SSE =1.67
SSE=15 N-C=12-4=8 MSE = =3
Samples N−C

Cal FC = 1.67 & Tab FC (3,8)=4.07

Conclusion : Cal FC < Tab FC ⇒ Hence we accept H0
3. The accompanying data results from an experiment comparing the degree of soiling for fabric
co-polymerized with the three different mixtures of methacrcylic acid. Analysis is the given
classification
Mixture 1 0.56 1.12 0.90 1.07 0.94
Mixture 2 0.72 0.69 0.87 0.78 0.91
 Mixture 3 0.62 1.08 1.07 0.99 0.93
Solution:
H0: The true average degree of soiling is identical for 3 mixtures.
H1 : The true average degree of soiling is not identical for 3 mixtures.
We shift the origin

X1 X2 X3 TOTAL X12 X22 X32

0.56 0.72 0.62 1.9 0.3136 0.5184 0.3844
1.12 0.69 1.08 2.89 1.2544 0.4761 1.1664
Total 0.90 0.87 1.07 2.84 0.8100 0.7569 1.1449
1.07 0.78 0.99 2.84 1.1449 0.6084 0.9801
0.94 0.91 0.93 2.78 0.8836 0.8281 0.8649
4.59 3.97 4.69 4.4065 3.1879 4.5407

N = Total No of Observations = 15
T= Grand Total = 13.25
( Grand total )2
Correction Factor = = 11.7042
Total No of Observatio ns
2 2 2 T2
TSS = ∑ X 1 + ∑ X 2 + ∑ X 3 − = 0.4309
N
(∑ X ) (∑ X ) (∑ X )
1
2
2
2
3
2
T2
( N1 = No of element in each column )
SSC = + + − = 0.0608
N1 N1 N1 N

SSE = TSS – SSC = 0.4309 – 0.0608 = 0.3701

ANOVA TABLE

Source of Sum of Degree of

Mean Square F- Ratio
Variation Squares freedom
SSC
Between MSC = =0.030
SSC=0.0608 C-1=3-1=2 C −1
Samples MSE
4 FC =
MSC
SSE
Within MSE = =0.30 =10.144
SSE=0.3701 N-C=15-3=12 N−C
Samples
84
Cal FC = 10.144 & Tab FC (12,2)=19.41
Conclusion : Cal FC < Tab FC ⇒ Hence we accept H0
4. Analyse the following RBD and find the conclusion
Treatment T1 T2 T3 T4
s
B1 12 14 20 22

Blocks
B2 17 27 19 15
B3 15 14 17 12
B4 18 16 22 12
B5 19 15 20 14
Solution:
H0: There is no significant difference between blocks and treatment
H1 : There is no significant difference between blocks and treatment
We subtract 15 from the given value
T1 T2 T3 T4 Total=Ti [Ti2]/k ΣXij2

B1 -3 -1 5 7 8 16 84

B2 2 12 4 0 18 81 164

B3 0 -1 2 -3 -2 1 14

B4 4 0 5 -1 8 16 42

B5 4 0 5 -1 8 16 42

Total=Tj 6 11 23 0 40 130 372

[Tj2]/h 7.2 24.2 105.8 0 137.2

2
∑y ij 38 147 119 68 372

N = 20
T=Grand Total = 40 ;
(Grand total ) 2 ( 40) 2
Correction Factor = =
Total No of Observations 20
2 T2
TSS = ∑ ∑ X ij − = 292
N
2

SSC =
∑T J
− C.F = 57.2
h

T2 2
SSR = ∑ ∑ Yij −
N
= 50
SSE = TSS – SSC – SSR = 184.8
ANOVA Table
 Source of Sum of Degree of Mean
F- Ratio FTab Ratio
Variation Squares freedom Square
Between
Rows SSR = 50 h - 1= 3 MSR=12.5 FR = 1.232 F5%(12,4) =
(Blocks) 5.91
Between FC= 1.238 F5%(3, 12) =
Columns MSC = 3.49
SSC= 57.2 k – 1=4
(Treatmen 19.07
ts)
(h – 1)( k –
SSE =
Residual 1) MSE = 15.4
184.8
= 12
Total 292

Conclusion : Cal FC < Tab FC and Cal FR < Tab FR ⇒ hence the difference between the blocks and
that treatments are not significant

5. Consider the results given in the following table for an experiment involving 6 treatments
in 4 randomized blocks. The treatments are indicated by numbers with in the paranthesis.

1 (1) (3) (2) (4) (5) (6)

24.7 27.7 20.6 16.2 16.2 24.9
2 (3) (2) (1) (4) (6) (5)
22.7 28.8 27.3 15 22.5 17
3 (6) (4) (1) (3) (2) (5)
26.3 19.6 38.5 36.8 39.5 15.4
4 (5) (2) (1) (4) (3) (6)
17.7 31 28.5 14.1 34.9 22.9

Test whether the treatments differ significantly [ (F0.05 (3,15)=5.42, F0.05 (5,15)=4.5]

Solution:
We subtract the origin to 25 and workout with new values of Xij
Total
1 2 3 4 5 6 [Ti]/k ΣXij2
=Ti

2.7 -8.8 181.6

1 -0.3 -404 -8.8 -0.1 -19.7 64.68
3

-10 -2.5 195.2

2 22.3 3.8 -2.3 -8 -16.7 46.48
7
 11.8 -9.6 113.5 654.7
3 13.5 14.5 -5.4 1.3 26.1
4 5

9.9 -7.3 324.1

4 3.5 6 -10.9 -2.4 -1.2 0.24
2

Total 224.9 1355.

19 19.9 22.1 -35.1 -33.7 -3.7 -11.5
=Tj 4 77

[Tj2]/ 122.1 283.9 906.6

90.25 99 3.8 3.42
h 0 2 9

T=Grand Total = -11.5 ;

(Grand total ) 2 (−11.5) 2
Correction Factor = =
Total No of Observations 24

SSR =
∑T i
− C.F = 224.94 −
(− 11)2 = 219.43
k 24
2

SSC =
∑T j
− C.F = 906.69 −
(− 11)
2
= 901.18
h 24

SSE = TSS – SSC – SSR = 229.65

ANOVA Table

Source of
Sum of Degree of Mean
Variatio F- Ratio FTab Ratio
Squares freedom Square
n
Between 4.78 F5%(3, 15) =
Rows SSR=219.43 h - 1= 3 MSR=73.14 11.75 5.42
(Blocks) F5%(5,15) =
Between 4.5
Columns MSC
SSC=901.18 k – 1=5
(Treatme =180.24
nts)
(h – 1)( k –
MSE
Residual SSE=229.65 1)
=15.31
=15
Total 1350.26
 Conclusion : Cal FC < Tab FC and Cal FR > Tab FR ⇒ There is no significant difference
between the blocks and there is significant difference between the Treatments.

6. Three Varieties A,B,C of a crop are tested in a randomized block design with four
replications. The plot yield in pounds are as follows

A 6 C 5 A 8 B 9
C 8 A 4 B 6 C 9
B 7 B 6 C 10 A 6
Analyze the experimental yield and state your conclusion
Solution:

The table can be given as

I II III IV
A 6 4 8 6
B 7 6 6 9
C 8 5 10 9
We shift the origin Xij = xij – 6; h = 3; k = 4; N = 12
Total=Ti
I II III IV [Ti*2]/k ΣX*ij2
*
A 0 -2 2 0 0 0 8
B 1 0 0 3 4 4 10
C 2 -1 4 3 8 16 30
Total=T*j 3 -3 6 6 12 20 48
[T*j2]/h 3 3 12 12 30

T=Grand Total = 12
(Grand total ) 2 (12) 2
Correction Factor = = = 12
Total No of Observations 12
TSS = ∑∑ X ij2 − C.F = 48 − 12 = 36
i j

SSR =
∑T i*
− C.F = 20 − 12 = 8
k
2

SSC =
∑T *j
− C.F = 30 − 12 = 12
h
SSE = TSS – SSC – SSR = 36-8-18 =10
ANOVA Table
Source of Sum of Degree of Mean
F- Ratio FTab Ratio
Variation Squares freedom Square
Between
Rows SSR=8 h - 1= 2 MSR=4
(Workers)
 Between FR = 2.4 F5%(2, 6)
Columns SSC=18 k – 1=3 MSC = 6
(Machine)
FC = 3.6 =5.14

F5%(3, 6)
(h – 1)( k – MSE =
Residual SSE = 10
1) = 6 5/3
=4.76

Total 36

Conclusion : Cal FC < Tab FC and Cal FR < Tab FR ⇒ There is no significant difference
between the crops and no significant difference between the plots
7. An experiment was designed to study the performances of 4 different detergents for
cleaning fuel injectors. The following “cleanliness” readings were obtained with specially
designed experiment for 12 tanks of gas distributed over 3 different models of engines:

ENGINES I II III

A 45 43 51
B 47 46 52
DETERGENTS
C 48 50 55
D 42 37 49
Perform the ANOVA and test at 0.01 level of significance whether there are difference in
the detergents or in the engines.
Solution :
We shift the origin Xij = xij – 50; h = 4; k = 3; N = 12

I II III Total=Ti [Ti*2]/k ΣX*ij2

*
A -5 -7 1 -11 40.33 75
B -3 -4 2 -5 8.33 29
C -2 0 5 3 3 29
D -8 -13 -1 -22 161.33 234
Total=T* -18 -24 7 -35 212.99 367
j
[T*j2]/h 81 144 12.25 237.25

(Grand total ) 2 (−35) 2

T=Grand Total = -35 , Correction Factor = =
Total No of Observations 12

(−35) 2
TSS = ∑∑ X ij2 − C.F = 367 − = 264.92
i j 12
2

SSR =
∑T i*
− C.F = 212.99 −
(−35) 2
= 110.91
k 12
 2

SSC =
∑T *j
− C.F = 237.25 −
(−35) 2
= 135.17
h 12

SSE = TSS – SSC – SSR = 264.92-110.91-135.17 = 18.

ANOVA Table

Source of Sum of Degree of Mean F- F Tab

Variation Squares freedom Square Ratio Ratio
Between Rows
(DETERGENT SSR=110.91 h - 1= 3 MSR=36.97 FR = F5%(3, 6)
S) 11.774 = 4.76
Between F5%(2, 6)
SSC=135.1 MSC =
Columns k – 1=2 FC = = 5.14
7 7.585
(ENGINES) 21.52
(h – 1)( k –
Residual SSE= 18.84 MSE = 3.14
1) = 6
Total 264.92
Conclusion :
Cal FC > Tab FC and Cal FR > Tab FR ⇒ There is significant difference between the
DETERGENTS and significant difference between the ENGINES

8. A set of data involving four “ four tropical feed stuffs A,B,C,D” tried on 20 chicks is given
below. All the twenty chicks are treated alike in all respects except the feeding treatments and
each feeding treatment is given to 5 chicks. Analyze the data ( Apr/May 2017)

A 55 49 42 21 52
B 61 112 30 89 63
C 42 97 81 95 92
D 169 137 169 85 154
Solution:
H0 : There is no significant difference between column means as well as row means

H1 : There is no significant difference between column means as well as row means

X1 X2 X3 X4 X5 Total X21 X22 X23 X24 X25

A 5 -1 -8 -29 2 -31 25 1 64 841 4
B 11 62 -20 39 13 105 121 3844 400 1521 169
C -8 47 31 45 42 157 64 2209 961 2025 1764
D 119 87 119 35 104 464 14161 7569 14161 1225 10816
Total 127 195 122 90 161 695 14371 13623 15586 5612 12753

N = 20 T = 695
T2
C.F = = 24151.25
N
 2 2 2 T2
TSS = ∑ X 1 + ∑ X 2 + ∑ X 3 ∑ X 42 + ∑ X 52 − = 37793.75
N
(∑ X ) (∑ X ) (∑ X ) (∑ X ) (∑ X )
1
2
2
2
3
2
4
2
5
2
T2
SSC = + + + + − = 1613.50
N1 N1 N1 N1 N1 N
( N1 = No of element in each column )
(∑ Y ) (∑ Y ) (∑ Y ) (∑ Y ) (∑ Y )
1
2
2
2
T2
3
2
4
2
5
2

SSR = + + + + = 26234.95 −
N2 N2 N2 N2 N2 N
( N 2 = No of element in each row )
SSE = TSS – SSC-SSR = 37793.75 – 1613.5 – 26234.95 = =9945.3

ANOVA TABLE
S.V DF SS MSS F cal F
tab
MSC
Column c-1= 5-1 SSC FC = = 2.055
=4
SSC=1613.5 MSC = = 403 .375 MSE 3.26
treatment C −1
MCR
Between SSR FC = = 10.552
r-1=4-1=3 SSR=26234.95 MSR = = 8744.98 MSE 3.49
Row R −1
N-c- SSE
Error r+1=12
SSE=9945.3 MSE =
12
= 828 .775

Conclusion: Cal Fc < Tab Fc , Accept H0

Cal FR > Tab FR , Reject H0
9. Three varieties of coal were analysed by 4 chemists and the ash content is given below. Perform
an ANOVA Table

Chemists
A B C D
I 8 5 5 7
COAL II 7 6 4 4
III 3 6 5 4
Solution:

Chemists
A B C D TOT
I 8 5 5 7 25
COAL II 7 6 4 4 21
III 3 6 5 4 18
TOT 18 17 14 15 64

N = 12
 T = 64
T2
C.F = = 341.33
N
2 2 2 T2
TSS = ∑ X 1 + ∑ X 2 + ∑ X 3 ∑ X 42 − = 24.67
N
2 2 2

SSC =
(∑ X ) + (∑ X ) + (∑ X )
1 2 3
−
T2
= 3.34
N1 N1 N1 N
( N1 = No of element in each column )
2 2 2 2

SSR =
(∑Y ) + (∑Y ) + (∑Y ) + (∑Y )
1 T2
2
= 6.17
3 4
−
N2 N2 N2 N2 N
( N 2 = No of element in each row )
SSE = TSS – SSC - SSR = 24.67 – 3.34 – 6.17 = 15.16

ANOVA TABLE
S.V DF SS MSS F cal F
tab
Column C-1= 4-1 SSC MSE
=3
SSC=3.34 MSC = = 1.11 FC = = 2.28 3.49
treatment C −1 MSC
Between R-1=3- SSR MCR
1=2
SSR=6.17 MSR = = 3.09 FC = = 1.22 3.26
Row R −1 MSE
N-c- SSE
Error R+1=6
SSE=15.16 MSE =
12
= 2.53

Conclusion: Cal Fc > Tab Fc , Reject H0

Cal FR > Tab FR , Reject H0
10. The following is the latin square of a design when 4 varieties of seed are being tested. Set
up the analysis of variance table and state your conclusion. You can carry out the suitable
change of origin and scale
A 110 B 100 C 130 D 120
C 120 D 130 A 110 B 110
D 120 C 100 B 110 A 120
B 100 A 140 D 100 C 120
Solution:
Subtracting 100 and dividing by 10
Total=Ti [Ti*2]/n ΣX*ij2
1 2 3 4
*

1 A1 B0 C3 D2 6 9 14
2 C2 D3 A1 B1 7 12.25 15
3 D2 C0 B1 A2 5 6.25 9
4 B0 A4 D0 C2 6 9 20
Total=T*j 5 7 5 7 24 36.5 58
[T*j2]/n 6.25 12.25 6.25 12.25 37
∑ yij2 9 25 11 13 58
 Letters Total=TK [TK2]/n
A 1 1 2 4 8 16
B 0 1 1 0 2 1
C 3 2 0 2 7 12.25
D 2 3 2 0 7 12.25
Total 24 41.5

T2 1 T2
Q = ∑ ∑ Yij2 −
N
= 22 Q1 =
n
∑ i N = 0.5
T 2
−

1 T2
Q2 =
n
∑ j N =1
T 2
−

1 T2
Q3 =
n
∑ K N = 5.5
T 2
−

Q 4 = Q − Q1 − Q 2 − Q3 = 15

Source of Sum of Degree of Mean FTab Ratio

F- Ratio
Variation Squares freedom Square ( 5% level)
Between
0.5 3 0.167 FR(6,36)=8.9
Rows FR

Between =14.97
1 3 0.333 4
Columns
Between FC
5.5 3 1.833 Fc(6,3)=8.94
Letters
=7.508
FL(6 ,3)=8.94
Residual 15 6 2.5 FL
=1.364
Total 22 15
Conclusion:
Cal FR > Tab FR , Cal FC < Tab FC , Cal FL < Tab FL There is a significant difference
between rows and no significant difference between column and also between letters.
11. A Company wants to produce cars for its own use. It has to select the make of the car out
of the four makes A,B,C,D available in the market. For this he tries 4 cars of each make by
assigning the cars to 4 drivers to run on 4 different routes. The efficiency of the cars is
measured in terms of time in hours.
Analyse the experinment data and draw conclusion (F0.05 (3,5)=5.41).
18(C) 12(D) 16(A) 20(B)
26(D) 34(A) 25(B) 31(C)
15(B) 22(C) 10(D) 28(A)
30(A) 20(B) 15(C) 9(D)

Solution:
We subtract 20 from the given value and workout with new value of Xij
1 2 3 4 Total=Ti [Ti2]/n ΣXij2
1 C D A B -14 49 84
-2 -8 -4 0
2 D A B C 36 324 378
6 14 5 11
3 B C D A -5 6.25 193
-5 2 -10 8
4 A B C D -6 9 246
10 0 -5 -11
Total=Tj 9 8 -14 8 11 388.25 901
[Tj2]/n 20.25 16 49 16 101.25
ΣXi2 165 264 166 306 901

Letters Total=TK [TK2]/n

A -4 14 8 10 28 196
B 0 5 -5 0 0 0
C -2 11 2 -5 6 9
D -8 6 -10 -11 -23 132.25
Total 11 337.25

(Grand total ) 2 (11) 2

T=Grand Total = 11 ; Correction Factor = =
Total No of Observations 16

(11) 2
TSS = ∑∑ X ij2 − C.F = 901 − = 893.438
i j 16
2

SSR =
∑T i*
− C.F = 388.25 −
(11) 2
= 380.688
n 16
2

SSC =
∑T j
− C.F = 101.25 −
(11) 2
= 93.688
n 16
 2

SSL =
∑T K
− C.F = 337.25 −
(11) 2
= 329.688
n 16

SSE = TSS – SSC – SSR-SSL = 89.374

Source of Sum of Degree of Mean FTab Ratio

F- Ratio
Variation Squares freedom Square ( 5% level)
Between MSR=126.89
SSR=380.688 n - 1= 3 FR(3,
Rows 6 FR =

Between MSC 8.519

SSC=93.688 n - 1= 3 6)=4.76
Columns =31.229
Between SSL = MSL=109.89 F
n - 1= 3 C Fc(3, 6)=4
Letters 329.688 6
=2.096
.76

FL
(n – 1)(n – MSE = FL(3, 6)=4
Residual SSE= 89.374 =7.378
2) = 6 14.896
.76

Total 893.438
Conclusion :
Cal FC < Tab FC , Cal FL > Tab FL and Cal FR > Tab FR ⇒ There is significant difference
between the rows , no significant difference between the column and significant difference
between the letters
12. A variable trial was conducted on wheat with 4 varieties in a Latin square Design. The plan of
the experiment and the per plot yield are given below:

C 25 B 23 A 20 D 20
A 19 D 19 C 21 B 18
B 19 A 14 D 17 C 20
D 17 C 20 B 21 A 15
Solution:
Subtract 20 from all the items
X1 X2 X3 X4 Total X21 X22 X23 X24
Y1 5 3 0 0 8 25 9 0 0
Y2 -1 -1 1 -2 -3 1 1 1 4
Y3 -1 -6 -3 0 10 1 36 9 0
Y4 -3 0 1 -5 -7 9 0 1 25
Total 0 -4 -1 -7 -12 36 46 11 29

H0 : There is no significant difference between rows, columns & treatments.

H1 : There is significant difference between rows, columns & treatments.
 N = 16 T = -12
T2 2 2 2 T2
C.F = =9 TSS = ∑ X 1 + ∑ X 2 + ∑ X 3 ∑X 2
4 + ∑ X 52 − = 113
N N

(∑ X ) (∑ X ) (∑ X ) (∑ X ) (∑ X )
1
2
2
2
3
2
4
2
5 T2
2

SSC = + + + + = 7.5
−
N1 N1 N1 N1 N1 N
( N1 = No of element in each column )
(∑ Y ) (∑ Y ) (∑ Y ) (∑ Y ) (∑ Y )
1
2
2
2
3
2
4
2
5
2
T2
SSR = + + + + − = 46.5
N2 N2 N2 N2 N2 N

( N 2 = No of element in each row )

SSK:

T
A 0 -1 -6 -5 -12
B 3 -2 -1 1 1
C 5 1 0 0 6
D 0 -1 -3 -3 -7

SSK =
(− 12)2 + (1)2 + (6)2 + (− 7 )2
T2
= 48.5 −
4 4 4 4 N
SSE = TSS – SSC - SSR = 113-7.5-46.5-48.5 = 10.5

ANOVA TABLE
S.V DF SS MSS F cal F
tab
Column SSC MSC
k-1=3 SSC=7.5 MSC = = 2 .5 FC = = 1.43 4.76
treatment K −1 MSE
Between SSR MCR
k-1=3 SSR=46.5 MSR = = 15.5 FR = = 8.86 4.76
Row K −1 MSE
Between SSK MSK
k-1=3 SSK=48.5 MSK = = 16.17 FT = = 9.24 4.76
Treatment K −1 MSE
SSE
(k-1)(k-2) MSE = = 1.75
Error SSE=10.5 ( K − 1)( K − 2)
=6

Conclusion: Cal Fc < Tab Fc

Cal FR > Tab FR
Cal FT > Tab FT
There is significant difference between treatment and rows but there is no significant difference between
columns.
13. Analyse 22 factorial experiment for the following table

Block Treatment
(1) kp k p
I
64 6 25 30
 k (1) kp P
II
14 75 33 50
kp p k (1)
IIII
17 41 12 76
p k (1) kp
IV
25 33 75 10
Solution:
Treatme
I II III IV
nt
(l) 64 75 76 75
(k) 25 14 12 33
(p) 30 50 41 25
(kp) 6 33 17 10

We shift the origin Xij = xij – 37;

Total=Ti [Ti*2]/n ΣX*ij2
Treatment I II III IV
*

(l) 27 38 39 38 142 5041 5138

(k) -12 -23 -25 -4 -64 1024 1314
(p) 7 13 4 -12 12 36 378

(kp) -31 -4 -20 -27 -82 1681 2106

Total=T*j -9 24 -2 -5 8 7782 8936
[T*j2]/n 20.25 144 1 6.25 171.5

T=Grand Total = 8: N=16

(Grand total ) 2 (8) 2
Correction Factor = = =4
Total No of Observations 16
TSS = ∑∑ X ij2 − C.F = 8936 − 4 = 8932
i j

SSR =
∑T i*
− C.F = 7782 − 4 = 7778
n
2

SSC =
∑ T* j
− C.F = 171.5 − 4 = 167.5
n
SSE = TSS – SSC – SSR = 8932 – 7778 – 167.5 = 986.5
[k] = [kp] – [p] + [k] – [1] =-300 ; [p] = [kp] + [p] - [k] – [1] =-148

[kp] = [kp] – [p] - [k] + [1] =126

Sk = [k]2 /4r = 5625; Sp = [p]2 /4r = 1369; Skp = [kp]2 /4r = 992.2
ANOVA Table

Source
Sum of Degree of Mean
of F- Ratio FTab Ratio
Squares freedom Square
Variatio
 n

F5%(1, 9) =
K 5625 1 5625
6.99
P 1369 1 1369 Fk = 51.32
F5%(1, 9) =
Fp = 12.49
Kp 992.25 1 992.25
6.99
Fkp = 9.05
F5%(1, 9) =
Error 986.5 9 109.6
6.99

Conclusion : Cal Fk > Tab Fk , Cal Fp > Tab Fp and Cal Fkp > Tab Fkp ⇒ There is significant
difference between the treatments.
14. Given the following observation for the 2 factors A & B at two levels compute (i) the main effect
(ii) make an analysis of variance.

Treatment
Replication I Replication II Replication III
Combination
(1) 10 14 9
A 21 19 23
B 17 15 16
AB 20 24 25
Solution:
H0 : No difference in the Mean effect.
H1 :Tthe is a difference in the Mean effect.
We code the data by subtracting 20

Treatment Replication Total X21 X22 X23

(l) -10 -6 -11 -27 100 36 121

(a) 1 -1 3 3 1 1 9
(b) -3 -5 -4 -12 9 25 16
(ab) 0 4 5 9 0 16 25
Total -27 110 78 171
T= Grand Total = -27 N =12
(Grand total ) 2 (−27) 2
Correction Factor = = = 60.75
Total No of Observations 12
A Contract = a + ab – b – (1) = 3+9+-(12)-(-27) = 51
B Contract = b + ab – a – (1) = -12+9-3-(-27) = 21
A Contract = (1) + ab – a – b = - 27+9-3-(-12) = -9
(i) Main effects of A = A Contract / 2n = 51/6 = 8.5
Main effects of B = B Contract / 2n = 21/6 = 3.5
Main effects of AB = AB Contract / 2n = -9 / 6 = -1.5
 2 2 2 T2
TSS = ∑ X 1 + ∑ X 2 + ∑ X 3 − = 110 + 78 + 171 − 60.75 = 298.25
N
2
( A contract) 2 ( 51)
SSA = = = 216.75
4n 12
2
( B contract)2 ( 21)
SSB = = = 36.75
4n 12
2
( AB contract)2 ( −9 )
SSAB = = = 6.75
4n 12
SSE = TSS – SSA – SSB - SSAB = 298.25 – 216.75 – 36.75 – 6.75 = 38
ANOVA Table
Degree
Source of Sum of Mean
of F- Ratio FTab Ratio
Variation Squares Square
freedom
MSA=216.
A SSA=216.75 1
75
FA = 45.63 F5%(1, 8) = 5.32
MSB=36.7
B SSB=36.75 1
5 FB = 7.74 F5%(1, 8) = 5.32
MSAB=6.7 FAB = 1.42 F5%(1, 8) = 5.32
AB SSAB=6.75 1
5

Error SSE=38 4(n-1)=8 MSE=4.75

Total TSS=298.25 4n-1=11

Conclusion : Cal FA > Tab FA , Reject H0

Cal FB > Tab FB Reject H0
Cal FAB < Tab FAB Accept H0
15. The following are the number of mistakes made in 5 successive days by 4 technicians
working for a photographic laboratory. Test whether the difference among the four
sample means can be attributed to chance at α = 0.01 ..
Technician I II III IV
Day 1 6 14 10 9
Day 2 14 9 12 12
Day 3 10 12 7 8
Day 4 8 10 15 10
Day 5 11 14 11 11
Solution:
H0: There is no significant difference between the technicians
H1 : Significant difference between the technicians
We shift the origin
X1 X2 X3 X4 TOTAL X12 X22 X32 X42
Total -4 4 0 -1 -1 16 16 0 1
4 -1 2 2 7 16 1 4 4
 0 2 -3 -2 -3 0 4 9 4
-2 0 5 0 3 4 0 25 0
1 4 1 1 7 1 16 1 1
-1 9 5 0 13 37 37 39 10
N= Total No of Observations = 20 T=Grand Total = 13
(Grand total )2
Correction Factor = = 8.45
Total No of Observations
2 2 2 2
TSS = ∑X 1 + ∑X 2 + ∑X 3 + ∑X 4 − C.F = 37 + 37 + 39 + 10 − 8.45 = 114.55

SSC =
( ∑X 1)
2

+
( ∑X 2)
2

+
( ∑X 3)
2

− C .F =
(−1) 2 (9) 2 (5) 2
+ + + 0 − 8.45 = 12.95
N1 N1 N1 5 5 5

SSE = TSS – SSC = 114.55-12.95= 101.6

ANOVA Table

Source of Sum of Degree of

Mean Square F- Ratio
Variation Squares freedom

Between SSC
SSC=12.95 C-1= 4-1=3 MSC = =4.317 MSC
Samples C −1 FC =
MSE
Within SSE =1.471
SSE=101.6 N-C=20-4=16 MSE = =6.35
Samples N−C

Cal FC = 1.471 & Tab FC (16,3)=5.29

Conclusion : Cal FC < Tab FC ⇒ There is no significance difference between the
technicians
16. The following data represent the number of units of production per day turned out by
different workers using 4 different types of machines. [May/June-2013]
A B C D
Machine
type
1 44 38 47 36
2 46 40 52 43
Workers 3 34 36 44 32
4 43 38 46 33
5 38 42 49 39
(1) Test whether the five men differ with respect to mean productivity and
(2) Test whether the mean productivity is the same for the four different machine types.
Solution:
H0: There is no significant difference between the Machine types and no significant
difference between the Workers
H1 : Significant difference between the Machine types and no significant difference
between the Workers
We shift the origin Xij = xij – 46; h = 5; k = 4; N = 20
A B C D Total=Ti [Ti*2]/k ΣX*ij2
*
1 -2 -8 1 -10 -19 90.25 169
2 0 -6 6 -3 -3 2.25 81
3 -12 -10 -2 -14 -38 361 444
4 -3 -8 0 -13 -24 144 242
5 -8 -4 3 -7 -16 64 138
Total -25 -36 8 -47 -100 661.5 1074
=T*j
[T*j2]/h 125 259.2 12.8 441.8 838.8
T=Grand Total = -100
(Grand total ) 2 (−100) 2
Correction Factor = = = 500
Total No of Observations 20
TSS = ∑∑ X ij2 − C.F = 1074 − 500 = 574
i j

SSR =
∑T i*
− C.F = 661.5 − 500 = 161.5
k
2

SSC =
∑T *j
− C.F = 838.8 − 500 = 338.8
h

SSE = TSS – SSC – SSR = 574 – 161.5 – 338.8 = 73.7

ANOVA Table
Source of
Sum of Degree of Mean
Variatio F- Ratio FTab Ratio
Squares freedom Square
n
Between
Rows SSR=161. MSR = FR = 6.574 F5%(4, 12) =
h - 1= 4
(Workers 5 40.375 3.26
)
Between
Columns SSC=338. MSC =
k – 1=3 FC =
(Machine 8 112.933 F5%(3, 12) =
) 18.388 3.59
SSE = (h – 1)( k – MSE
Residual
73.7 1) = 12 = 6.1417
Total 1074
Conclusion : Cal FC < Tab FC and Cal FR < Tab FR ⇒ There is no significant difference
between the Machine types and no significant difference between the Workers