Environmental Engineering Reference
In-Depth Information
the "Pred R-Squared" of 0.9626 is good agreement with the "Adj R-Squared"
values of 0.9958 . The F-value of 76.91 confirm that the the model is
significant. And also "Prob > F" less than 0.05 shows that the model terms are
significant. The linear mixture components, AB, AC are significant model
terms. The values greater than 0.1 indicate the model terms are not significant.
As required for the model ,the "Lack of Fit F-value" 0.56 shows not significant
relative to the pure error .
Table 29 shows the ANOVA for mixture quadratic model for response
density and Table 30 shows the ANOVA for mixture special cubic model for
response cadmium leaching. The ANOVA for mixture reduced cubic model
for response lead leaching is shown in Table 31. As shown in Table 30, the
'F'-value of 15.08 implies the model is significant. Linear Mixture omponents
are significant model terms and values greater than 0.1 indicate the model
terms are not significant. The "Lack of Fit F-value" of 376.19 shows the Lack
of Fit is significant. But significant lack of fit is bad for the model to fit.
However, as shown in Table 24 the model summary statistics for response
cadmium leaching, the "Pred R-Squared" 0.7601 is in reasonable agreement
with the "Adj R-Squared" of 0.8667.
Table 27. ANOVA for mixture quadratic
model for response water absorption
Source
Sum of
Square
DF
Mean
Squares
F Value
Prob >
F
Model
82.31
5
16.46
56.23
<
0.0001
significant
Linear
Mixture
74.63
2
37.32
127.46
<
0.0001
AB
3.55
1
3.55
12.12
0.0083
AC
0.020
1
0.020
0.067
0.8026
BC
1.221E-
004
1
1.221E-
004
4.172E-
004
0.9842
Residual
2.34
8
0.29
Lack of Fit
0.99
4
0.25
0.73
0.6151
not
significant
Pure Error
1.35
4
0.34
Cor Total
84.65
13
Search WWH ::
Custom Search