Agriculture Reference
In-Depth Information
Example: Purity.
The H-value of heterogeneity represented by twenty different purity analyses (tests)
from a chaffy seed lot consisting of, for example, 300 individual containers can be illustrated by the follow-
ing example (the same will apply to germination test results):
1. The test results: 99.4, 99.9, 99.6, 98.7, 99.5, 99.6, 98.9, 98.9, 99.0, 99.3, 98.5, 99.4, 99.6, 99.5,
99.1, 98.2, 99.2, 99.9, 99.8, 99.5.
2. The mean of 20 (N) different purity test results, X = 99.275.
3. The acceptable variance (v) is calculated as follows:
v = [
¯
•
(100 -
¯
)]
/
n
•
1.2 = [99.275 (100 - 99.275)/1000]
•
1.2 = 0.0864.
For this example, f = 1.2 for purity of chaffy seeds (from Table 13.18) .
4. The actual (observed) variance is calculated as follows:
V = [NΣX
2
-(ΣX)
2
]/[N(N-1)] = 20(99.4
2
+...+ 99.5
2
) - (99.4 +...+ 99.5)
2
/20(20-1) = 0.2125
5. The heterogeneity value (H) is calculated by:
H = (V/W) - f = (0.2125/0.0864) -1.2 = 1.26
According to Table 13.19 (Table 2D, ISTA rules, 2010), the critical H-value based on results from
20 samples taken from lots with more than 50 bags is 1.09. The calculated heterogeneity (H-value) of the
example above is greater than 1.09, so the heterogeneity is considered excessive.
Other seed counts.
Heterogeneity among test results for other seeds is illustrated by the following
example for a non-chaffy seed lot of 300 bags from which 20 were sampled and tested for other seeds:
1. The test results are: 5, 8, 10, 20, 25, 9, 10, 11, 18, 16, 19, 22, 4, 13, 20, 30, 27, 12, 15, 18.
2. The mean number of other seeds in 20(N) samples is: x = (5 + 8 + ...+ 18)/20 = 15.6.
3. The acceptable variance (W) is calculated by the formula below.
W =
× ¯
•
f = (15.6)
•
(1.4) = 21.84
4. The actual (observed) variance is calculated by:
V = [20 (5
2
+ 8
2
+... + 18
2
) - (5 + 8 + ... + 18)
2
] / [20(20-1)]
=
51.621
5. The heterogeneity is calculated as:
H = (V/W) - f = (51.621/21.84) - 1.4 = 0.964
This H-value is then compared with the critical H-values in Table 13.19 (Table 2D in the ISTA rules,
2010). The heterogeneity is less than the critical H-value of 2.00, so the lot is considered to be homogenous,
or does not exhibit excessive heterogeneity.
