Agriculture Reference
In-Depth Information
ANOVA table for 3
4 split plot experiment
Table value of
F
SOV
d.f.
MS
MS
F
-ratio
p ¼
0.05
p ¼
0.01
Replication
3
1
¼
2
0.035
0.018
0.545
6.94
18.00
3
1
¼
2
Main Plot Factor(P)
45.207
22.603
704.519
6.94
18.00
Error I
(3
1) (3
1)
¼
4
0.128
0.032
Subplot factor(V)
(4
1)
¼
3
23.992
7.997
464.366
3.16
5.09
Interaction (P
V)
(3
1) (4
1)
¼
6
6.198
1.033
59.978
2.66
4.01
Error II
3 (4
1) (3
1)
¼
18
0.310
0.017
3.3.4
1
¼
35
Total
75.87
'
3. Standard error for difference between two tillage
means at the same or different level of subplot
Thus, second tillage is the best, which is
significantly superior to the other two tillages.
The two methods of tillage, namely, T
3
and T
1
,
are statistically at par. So far as the effect of
variety is concerned, maximum yield of ginger
is obtained from variety V
2
followed by V
3
,
V
1
,andV
4
. All the varieties are significantly
different from each other with respect to yield
of ginger. The interaction of the method of
tillage and the varieties has significantly dif-
ferent effects from each other; the interac-
tion effects which are not significantly
different have been put under the same lines.
Variety in combination with tillage T2V
2
has
produced significantly higher yield than any
other combination, followed by P
2
V
3
,P
2
V
4
,
andsoon.
The above analysis can also be done using
SAS statistical software. The following few slides
present the data entry, command syntax, and the
output for the same:
r
2
½ðn
1
Þ
ErMS-II
þ
ErMS-I
rn
treatment
¼
¼
r
2
½ð
4
1
Þ
0
:
017
þ
0
:
032
¼
0
:
1386, but the
3
:
4
ratio of the treatment mean difference and the
above SE does not follow
t
distribution, and the
t
approximate value of
is given by
t ¼
t
1
ErMS-I
þ t
2
ðn
1
Þ
ErMS-II
ErMS-II
¼
þðn
Þ
ErMS-I
1
ð
2
:
776
Þð
0
:
032
Þþð
2
:
101
Þð
4
1
Þ
0
:
017
¼
ð
0
:
032
Þþð
4
1
Þð
0
:
017
Þ
2
:
361, where
t
1
¼ t
0.025,4
value and
t
2
¼
t
0.025,18
value
and
the
corresponding
CD value
could
be LSD
ð
0
:
05
Þ
¼
SE
d
tð
cal
Þ¼
0
:
1386
2
:
361
¼
0
:
327.
Table of mean comparison
Average
LSD(0.05)
Tillage
T
2
10.700
0.2027
T
3
8.383
10.12.2 Strip Plot Design
T
1
8.267
Variety
V
2
10.444
0.129
In split plot design, if both the factors require a
large plot like that in the main plot factor, then it
may not be possible in split plot design. Strip
plot design is the solution. In agricultural field
experimentation with factors like different
methods of irrigation, different methods of pest
control, different methods of mulching, differ-
ent methods of plowing, etc., require larger plot
size for convenience of management. In
experiments involving these factors, strip plot
design may become very useful. In strip plot
design each replication is divided into the num-
ber of horizontal rows and the number of verti-
cal columns equals to the number of levels of
V
3
9.122
V
1
8.533
V
4
8.367
T V
T
2
V
2
12.467
0.227
T
2
V
3
11.033
and 0.324
T
2
V
4
9.800
T
3
V
2
9.667
T
2
V
1
9.500
T
1
V
2
9.200
T
1
V
1
8.600
T
3
V
3
8.367
T
3
V
4
8.000
T
1
V
3
7.967
T
3
V
1
7.500
