Practical No: 7

Name of Experiment:
Analysis Using Two-Way ANOVA (Without Replication)
Question:
The following table gives the result of the experiment on four varieties of a crop in 5 blocks plot.
Blocks
Block I Block II Block III Block IV Block V
A 32 B 33 D 30 A 35 C 36
B 34 C 34 C 35 C 32 D 29
C 31 A 34 B 36 B 37 A 37
D 29 D 26 A 33 D 28 B 35
 Analyze the above result to test whether there is significant difference between yields of four
varieties.

Working Expression:
Hypothesis setting:

Null hypothesis (𝐻0𝑇): 𝜇𝐴 = 𝜇𝐵 = 𝜇𝐶 = 𝜇𝐷 i.e there is no significant difference between yields of

varieties.
Null hypothesis (𝐻0𝐵) : 𝜇𝐼 = 𝜇𝐼𝐼 = 𝜇𝐼𝐼𝐼 = 𝜇𝐼𝑉= 𝜇𝑉 i.e there is no significant difference between
blocks.

Alternative hypothesis (𝐻1𝑇) : At least one varieties of yields is different to other.

Alternative hypothesis (𝐻1𝐵) : There is significant difference between blocks.

Level of significance (𝛼) : 0.05

UNIANOVA Values BY Block Treatments

/METHOD=SSTYPE(2)
/INTERCEPT=EXCLUDE
/SAVE=PRED RESID
/POSTHOC=Block Treatments(TUKEY LSD)
/PLOT=PROFILE(Block*Treatments) TYPE=LINE ERRORBAR=NO MEANREFERENCE=NO YAXIS=AUTO
/PRINT DESCRIPTIVE LOF
/PLOT=RESIDUALS
/CRITERIA=ALPHA(.05)
/DESIGN=Block Treatments.

Univariate Analysis of Variance

Between-Subjects Factors
Value Label N
There are five 1 Block 1 4
blocks(b1,b2,b3,b4,b5) 2 Block 2 4
3 Block 3 4
4 Block 4 4
5 Block 5 4
There are four 1 A 5
treatments(a,b,c,d) 2 B 5
3 C 5
4 D 5
 Descriptive Statistics
Dependent Variable: Values
There are five There are four
blocks(b1,b2,b3,b4,b5) treatments(a,b,c,d) Mean Std. Deviation N
Block 1 A 32.00 . 1
B 34.00 . 1
C 31.00 . 1
D 29.00 . 1
Total 31.50 2.082 4
Block 2 A 34.00 . 1
B 33.00 . 1
C 34.00 . 1
D 26.00 . 1
Total 31.75 3.862 4
Block 3 A 33.00 . 1
B 36.00 . 1
C 35.00 . 1
D 30.00 . 1
Total 33.50 2.646 4
Block 4 A 35.00 . 1
B 37.00 . 1
C 32.00 . 1
D 28.00 . 1
Total 33.00 3.916 4
Block 5 A 37.00 . 1
B 35.00 . 1
C 36.00 . 1
D 29.00 . 1
Total 34.25 3.594 4
Total A 34.20 1.924 5
B 35.00 1.581 5
C 33.60 2.074 5
D 28.40 1.517 5
Total 32.80 3.122 20

Tests of Between-Subjects Effects

Dependent Variable: Values
Type II Sum of
Source Squares df Mean Square F Sig.
Model 21672.500 a
8 2709.062 1101.992 .000
Block 21.700 4 5.425 2.207 .130
Treatments 134.000 3 44.667 18.169 .000
Error 29.500 12 2.458
Total 21702.000 20

a. R Squared = .999 (Adjusted R Squared = .998)

Lack of Fit Tests

Dependent Variable: Values
Source Sum of Squares df Mean Square F Sig.
Lack of Fit 29.500 12 2.458 . .
Pure Error .000 0 .

Post Hoc Tests

There are five blocks(b1,b2,b3,b4,b5)

Multiple Comparisons
Dependent Variable: Values
95%
Confidence
Mean Interval
(I) There are five (J) There are five Difference (I- Std. Lower
blocks(b1,b2,b3,b4,b5) blocks(b1,b2,b3,b4,b5) J) Error Sig. Bound
Tukey Block 1 Block 2 -.25 1.109 .999 -3.78
HSD Block 3 -2.00 1.109 .415 -5.53
Block 4 -1.50 1.109 .666 -5.03
Block 5 -2.75 1.109 .160 -6.28
Block 2 Block 1 .25 1.109 .999 -3.28
Block 3 -1.75 1.109 .536 -5.28
Block 4 -1.25 1.109 .790 -4.78
Block 5 -2.50 1.109 .225 -6.03
Block 3 Block 1 2.00 1.109 .415 -1.53
Block 2 1.75 1.109 .536 -1.78
Block 4 .50 1.109 .990 -3.03
Block 5 -.75 1.109 .958 -4.28
Block 4 Block 1 1.50 1.109 .666 -2.03
Block 2 1.25 1.109 .790 -2.28
Block 3 -.50 1.109 .990 -4.03
Block 5 -1.25 1.109 .790 -4.78
Block 5 Block 1 2.75 1.109 .160 -.78
Block 2 2.50 1.109 .225 -1.03
Block 3 .75 1.109 .958 -2.78
 Block 4 1.25 1.109 .790 -2.28
LSD Block 1 Block 2 -.25 1.109 .825 -2.67
Block 3 -2.00 1.109 .096 -4.42
Block 4 -1.50 1.109 .201 -3.92
Block 5 -2.75 *
1.109 .029 -5.17
Block 2 Block 1 .25 1.109 .825 -2.17
Block 3 -1.75 1.109 .140 -4.17
Block 4 -1.25 1.109 .282 -3.67
Block 5 -2.50 *
1.109 .044 -4.92
Block 3 Block 1 2.00 1.109 .096 -.42
Block 2 1.75 1.109 .140 -.67
Block 4 .50 1.109 .660 -1.92
Block 5 -.75 1.109 .512 -3.17
Block 4 Block 1 1.50 1.109 .201 -.92
Block 2 1.25 1.109 .282 -1.17
Block 3 -.50 1.109 .660 -2.92
Block 5 -1.25 1.109 .282 -3.67
Block 5 Block 1 2.75* 1.109 .029 .33
Block 2 2.50 *
1.109 .044 .08
Block 3 .75 1.109 .512 -1.67
Block 4 1.25 1.109 .282 -1.17

Multiple Comparisons
Dependent Variable: Values
95% Confidence
(I) There are five (J) There are five Interval
blocks(b1,b2,b3,b4,b5) blocks(b1,b2,b3,b4,b5) Upper Bound
Tukey HSD Block 1 Block 2 3.28
Block 3 1.53
Block 4 2.03
Block 5 .78
Block 2 Block 1 3.78
Block 3 1.78
Block 4 2.28
Block 5 1.03
Block 3 Block 1 5.53
Block 2 5.28
Block 4 4.03
Block 5 2.78
Block 4 Block 1 5.03
Block 2 4.78
 Block 3 3.03
Block 5 2.28
Block 5 Block 1 6.28
Block 2 6.03
Block 3 4.28
Block 4 4.78
LSD Block 1 Block 2 2.17
Block 3 .42
Block 4 .92
Block 5 -.33
Block 2 Block 1 2.67
Block 3 .67
Block 4 1.17
Block 5 -.08
Block 3 Block 1 4.42
Block 2 4.17
Block 4 2.92
Block 5 1.67
Block 4 Block 1 3.92
Block 2 3.67
Block 3 1.92
Block 5 1.17
Block 5 Block 1 5.17
Block 2 4.92
Block 3 3.17
Block 4 3.67

Based on observed means.

The error term is Mean Square(Error) = 2.458.
*. The mean difference is significant at the .05 level.

Homogeneous Subsets
Values
There are five Subset
blocks(b1,b2,b3,b4,b5) N 1
Tukey HSD a,b
Block 1 4 31.50
Block 2 4 31.75
Block 4 4 33.00
Block 3 4 33.50
Block 5 4 34.25
Sig. .160
Means for groups in homogeneous subsets are displayed.
Based on observed means.
The error term is Mean Square(Error) = 2.458.
a. Uses Harmonic Mean Sample Size = 4.000.
b. Alpha = .05.

There are four treatments(a,b,c,d)

Multiple Comparisons
Dependent Variable: Values
(I) There are four (J) There are four Mean Difference
treatments(a,b,c,d) treatments(a,b,c,d) (I-J) Std. Error
Tukey HSD A B -.80 .992
C .60 .992
D 5.80 *
.992
B A .80 .992
C 1.40 .992
D 6.60 *
.992
C A -.60 .992
B -1.40 .992
D 5.20* .992
D A -5.80 *
.992
B -6.60 *
.992
C -5.20 *
.992
LSD A B -.80 .992
C .60 .992
D 5.80 *
.992
B A .80 .992
C 1.40 .992
D 6.60 *
.992
C A -.60 .992
B -1.40 .992
D 5.20* .992
D A -5.80 *
.992
B -6.60 *
.992
C -5.20 *
.992
 Multiple Comparisons
Dependent Variable: Values
95% Confidence
(I) There are four (J) There are four Interval
treatments(a,b,c,d) treatments(a,b,c,d) Sig. Lower Bound
Tukey HSD A B .850 -3.74
C .929 -2.34
D .000 2.86
B A .850 -2.14
C .516 -1.54
D .000 3.66
C A .929 -3.54
B .516 -4.34
D .001 2.26
D A .000 -8.74
B .000 -9.54
C .001 -8.14
LSD A B .436 -2.96
C .556 -1.56
D .000 3.64
B A .436 -1.36
C .183 -.76
D .000 4.44
C A .556 -2.76
B .183 -3.56
D .000 3.04
D A .000 -7.96
B .000 -8.76
C .000 -7.36

Multiple Comparisons
Dependent Variable: Values
95% Confidence
(I) There are four (J) There are four Interval
treatments(a,b,c,d) treatments(a,b,c,d) Upper Bound
Tukey HSD A B 2.14
C 3.54
D 8.74
B A 3.74
C 4.34
D 9.54
 C A 2.34
B 1.54
D 8.14
D A -2.86
B -3.66
C -2.26
LSD A B 1.36
C 2.76
D 7.96
B A 2.96
C 3.56
D 8.76
C A 1.56
B .76
D 7.36
D A -3.64
B -4.44
C -3.04

Based on observed means.

The error term is Mean Square(Error) = 2.458.
*. The mean difference is significant at the .05 level.

Homogeneous Subsets
Values
There are four Subset
treatments(a,b,c,d) N 1 2
Tukey HSD a,b
D 5 28.40
C 5 33.60
A 5 34.20
B 5 35.00
Sig. 1.000 .516

Means for groups in homogeneous subsets are displayed.

Based on observed means.
The error term is Mean Square(Error) = 2.458.
a. Uses Harmonic Mean Sample Size = 5.000.
b. Alpha = .05.
Profile Plots

Decision:-
Here, for treatment 2p = 0.000 which is less than the value of 0.05. Using p-value approach
we reject the null hypothesis i.e 𝐻0𝑇.
Similarly, for block 2p = 0.130 which is greater than the value of 0.05. Using p-value
approach we fail to reject the null hypothesis i.e 𝐻0𝐵.

CONCLUSION
Therefore, at least one varieties of yield are different to the other. Likewise, there is no
significant effect due to blocks.