Table 2.
Regression analyses to test the mediator effect of PG on floral symmetry prediction of CV of flower size. CI, confidence interval; floral symmetry (0, bilateral; 1, radial); n.s., not significant.
| predictor variable | B | s.e. | 95% CI | β | t | criterion variable |
|---|---|---|---|---|---|---|
| all species included | ||||||
| step 1 | ||||||
| floral symmetry | 0.128 | 0.041 | 0.044, 0.212 | 0.530 | 3.128** | CV of flower size |
| step 2 | ||||||
| floral symmetry | 0.665 | 0.264 | 0.122, 1.208 | 0.451 | 2.524* | PG |
| step 3 | ||||||
| PG | 0.077 | 0.027 | 0.021, 0.134 | 0.474 | 2.822** | CV of flower size |
| floral symmetry | 0.076 | 0.040 | −0.007, 0.160 | 0.317 | 1.885 (n.s.) | CV of flower size |
| two species excluded | ||||||
| step 1 | ||||||
| floral symmetry | 0.110 | 0.040 | 0.028, 0.192 | 0.502 | 2.786* | CV of flower size |
| step 2 | ||||||
| floral symmetry | 0.675 | 0.262 | 0.134, 1.216 | 0.474 | 2.581* | PG |
| step 3 | ||||||
| PG | 0.067 | 0.029 | 0.006, 0.127 | 0.432 | 2.295* | CV of flower size |
| floral symmetry | 0.065 | 0.041 | −0.020, 0.151 | 0.298 | 1.584 (n.s.) | CV of flower size |
*p < 0.05, **p < 0.01.