Table 3.
Significant differences between families and the different IADFs' types frequency.
| IADF type | Family i vs. family j | Differences of family LSM estimates (log scale) | P | OR |
|---|---|---|---|---|
| E-type | 2514 vs. 2818 | 1.057 | 0.018 | 2.877 |
| 2817 vs. 2818 | 0.961 | 0.015 | 2.615 | |
| 2860 vs. 2818 | 0.944 | 0.014 | 2.701 | |
| 2864 vs. 2818 | 0.942 | 0.021 | 2.564 | |
| E+-type | 2549 vs. 2818 | 0.890 | 0.008 | 2.436 |
| 2817 vs. 2818 | 0.892 | 0.007 | 2.441 | |
| 2860 vs. 2818 | 0.958 | 0.005 | 2.608 | |
| 3076 vs. 2818 | 0.741 | 0.034 | 2.098 | |
| 3078 vs. 2818 | 0.711 | 0.031 | 2.036 | |
| 2860 vs. 2864 | 0.435 | 0.044 | 1.545 | |
| L-type | 2860 vs. 2514 | 0.574 | 0.044 | 1.776 |
| 2860 vs. 2818 | 0.799 | 0.013 | 2.222 |
P, p-value; OR, odds ratio. OR estimates are obtained by exponentiation of the differences of Family Least Squares Means (LSM) estimates. If OR > 1 Family i has OR higher odds of IADF presence than Family j.