Table 2.
Comparison between hybrids and nonhybrids
| Effect tested | Fixed effects | Random effects | AIC | χ2 | d.f. | P |
|---|---|---|---|---|---|---|
| F2 | ||||||
| D, T, D*T | P, P1, P2 | 2759.9 | ||||
| days (D) | T | P, P1, P2 | 8645.7 | 5875.7 | 12 | <0.001 |
| cross type (T) | D | P, P1, P2 | 2940.3 | 194.4 | 7 | <0.001 |
| days*type (D*T) | D, T | P, P1, P2 | 2924.2 | 176.4 | 6 | <0.001 |
| population (P) | D, T, D*T | P1, P2 | 3006.1 | 248.2 | 1 | <0.001 |
| parent 1 (P1) | D, T, D*T | P, P2 | 3027.9 | 270 | 1 | <0.001 |
| parent 2 (P2) | D, T, D*T | P, P1 | 2787.8 | 30.0 | 1 | <0.001 |
| F1 | ||||||
| D, T, D*T | P, P1, P2 | 234.2 | ||||
| days (D) | T | P, P1, P2 | 285.9 | 1634.2 | 12 | <0.001 |
| cross type (T) | D | P, P1, P2 | 233.8 | 13.7 | 7 | 0.06 |
| days*type (D*T) | D, T | P, P1, P2 | 232.8 | 10.6 | 6 | 0.10 |
| population (P) | D | P1, P2 | 231.8 | 0.0 | 1 | 0.99 |
| parent 1 (P1) | D | P2 | 234.9 | 5.1 | 1 | 0.02 |
| parent 2 (P2) | D | P1 | 317.1 | 87.3 | 1 | <0.001 |
Likelihood ratio tests comparing generalized linear mixed models (GLMMs) on the effects of the number of days in the experiment, cross type (inter- or intraspecific), their interaction (days*type), population (n = 1308), parental strain 1 (the same in both inter-and intraspecific crosses) and parental strain 2 on survival in deteriorating environments. The upper part of the table shows the analysis within F2 hybrids; the lower part shows F1 hybrids. Akaikes information criterion (AIC) describes the quality of fit of each model. To evaluate the significance of fixed and random effects, alternative models without the variable of interest were compared to the full model (bold) using likelihood ratio tests. If an alternative model had a significantly better fit (bold), this model was subsequently compared against further reduced models.