Agriculture Reference
In-Depth Information
Mean sum of squares is calculated by dividing
the sum of squares by the corresponding
degrees of freedom.
ANOVA table for 3
(b) Standard error for difference between
two horizontal plot
treatment means
¼
q
2ErMS
-
II
rm
and the corresponding CD
value could be
4 strip plot design
with 3 replications
6.
F
-ratios for replication and vertical factor
effects are obtained by comparing the respec-
tive mean sum of squares against the mean
sum of squares due to error I. On the other
hand, the
r
2ErMS-II
3
:
3
CD
ð
0
:
05
Þ
¼
t
2
;
error
II d
:
f
:
r
2
0
:
034
¼
2
:
447
¼
0
:
213
:
-ratios corresponding to horizontal
factor and interaction between horizontal and
vertical factors are worked out by comparing
the respective mean sum of squares against
the mean sum of squares due to error II and
error III, respectively.
7. Calculated
F
3
:
3
(c) Standard error for difference between two
vertical plot treatment means at the same
level
of
horizontal
plot
treatment
q
2
½ðn
1
Þ
ErMS
-
III
þ
ErMS
-
I
rn
¼
F
-ratios
are
compared with
q
2
ð
4
1
Þ
0
:
046
þ
0
:
019
3
:
4
tabulated value of
at appropriate level of
significance and degrees of freedom. It is
found from the above table that except
for replication effect, all other effects are
F
162.
(d) Standard error for difference between two
horizontal plot
¼
¼
0
:
treatment means at
the
same level of vertical plot
treatment
Table value of
F
F
-value
(
p ¼
0.05)
(
p ¼
0.01)
SOV
d.f.
SS
MS
Replication
2
0.062
0.031
1.6
6.94
18.00
Vertical factor (manure)
2
32.362
16.181
832.1702
6.94
18.00
Error I
4
0.078
0.019
Horizontal factor (irrig.)
3
83.212
27.737
814.0336
4.76
9.78
Error II
6
0.204
0.034
M
I
6
10.644
1.774
38.7855
3.00
4.82
Error III
12
0.549
0.046
Total
35
127.112
q
2
significant at both 5 and 1% level of
significance.
8. Once the
½ðm
1
Þ
ErMSIII
þ
ErMSII
¼
rm
q
2
½ð
3
1
Þ
0
:
046
þ
0
:
034
3
-test becomes significant, our next
task is to estimate the SEs for different types
of comparison as given below:
(a) Standard error for difference between two
vertical plot treatment means
F
¼
167.
But the ratio of the treatment difference
and the above SE in (c) and (d) does not
follow the
¼
0
:
3
distribution, and the approxi-
mate weighted
t
q
2ErMS
-
I
rn
¼
t
is calculated as follows:
and the corresponding CD value could be
fðn
Þ
t
III
gþð
t
I
Þ
1
ErMS-III
ErMS-I
r
2ErMS-I
rn
fðn
1
Þ
ErMS-III
þ
ErMS-I
g
CD
α
¼
t
2
;
error
Id
:
f
:
¼
fð
4
1
Þ
0
:
046
2
:
179
gþð
0
:
019
2
:
776
Þ
¼
2
:
248
r
2
fð
4
1
Þ
0
:
046
þ
0
:
019
g
0
:
019
¼
2
:
776
¼
0
:
1562
:
3
:
4
