Table 3.
Models applied to infection success in the common-garden experiment, ranked by AIC
| Model | Fixed effect | Random effect | df | AIC | ΔAIC |
|---|---|---|---|---|---|
| Cops.m5 |
Family |
- |
17 |
6 136.0 |
0 |
| Cops.m0 |
- |
Family |
2 |
6 157.4 |
21.4 |
| Cops.m1 |
Group (1H) |
Family |
5 |
6 158.1 |
22.1 |
| Cops.m4 |
Sire |
Family |
4 |
6 159.2 |
23.2 |
| Cops.m2 |
Group (2H) |
Family |
6 |
6 159.5 |
23.5 |
| Cops.m3 |
Dam |
Family |
4 |
6 160.6 |
24.6 |
| Cops.m0B | - | - | 1 | 6 229.0 | 93.0 |
A set of generalized linear models (GLM) and generalized linear mixed effect models (GLMM) testing for the relative importance of different background covariates on the infection success. GLMs include fixed effect variables, and GLMMs contain both fixed and random effects. Response variable: Infection success status (binary) from common-garden trial, where unsuccessful individuals were lost somewhere between infection and sampling and all individuals sampled on termination day were defined as successful.
Covariates (fixed and/or random effect variables): Family = Families nr 1–17.
Group (1H) = Resistant, Sensitive, Hybrid, Unknown.
Group (2H) = Resistant, Sensitive, Hybrid SxR, Hybrid RxS, Unknown.
Sire (source strain of the father) = Resistant, Sensitive, Unknown.
Dam (source strain of the mother) = Resistant, Sensitive, Unknown.
Test results: degrees of freedom (df), AIC values and changes in AIC (ΔAIC).