Table 5.
Summary of multinomial logistic regression for variables characterising the different rice quality clusters. Cluster 1 is not shown in Table 5 because it is the reference cluster. Table 5 indicates the multinomial log-odds that samples represented in Cluster 2 or in Cluster 3, were compared to reference cluster 1 to calculate every unit increase or decrease in the different grain quality attributes included in the multinomial logistic regression model
| Grain quality attribute | Estimate | |||
|---|---|---|---|---|
| Cluster 2 | Cluster 3 | |||
| Intercept | −25.51 | (15.27) | 3.61 | (0.08)*** |
| AC (%) | 0.08 | (0.30) | −9.17 | (2.94)*** |
| G’trough | −0.33 | (0.19)* | −1.48 | (0.93) |
| BD | 0.06 | (0.03)** | 0.72 | (3.55) |
| S1 | −0.88 | (0.22)*** | −1.37 | (3.20) |
| tan (δ) at G’max | 0.55 | (0.14)*** | 0.21 | (11.71) |
| S3 | −2.84 | (0.84)*** | −0.12 | (0.66) |
| GT | 0.54 | (0.18)*** | 2.03 | (6.97) |
| COH | −0.24 | (0.08)*** | 0.01 | (11.97) |
| G’max | −0.15 | (0.06)*** | 0.22 | (8.68) |
| N | 70 | 27 | ||
Note: Total N = 211; AIC = 106.20; Overall classification accuracy: 93.84%
Reference category for the regression model is cluster 1 (n = 114)
Standard errors of the estimates are indicated in parentheses
* p < 0.1, ** p < 0.05, *** p < 0.01.
Goodness-of-fit statistics: Residual Deviance = 66.20; Degrees of freedom = 18
–2Log-likelihood: The intercept-only model: 405.87; The final model: 66.20; χ2 = 339.66; p < 0.01
Pseudo-R2: McFadden = 0.84; Cragg & Uhler = 0.94; Cox & Snell = 0.80