Table 4.
Model selection for time and tie width effects on recapture and survival probability in forest and urban great tit males, specifically testing for a difference in slopes in survival rate in the two habitats. Models are ranked according to AICc values
| Model | AICc | Delta AICc | AICc Weights | Model Likelihood | Num. Par | Deviance |
|---|---|---|---|---|---|---|
| Phi(g + t + tie + group * tie) p(g * t + tie) | 1314.10 | 0.0 | 0.98 | 1 | 40 | 1229.76 |
| Phi(g + t + tie) p(g * t + tie) | 1321.70 | 7.6 | 0.02 | 0.023 | 39 | 1239.57 |
| Phi(g + t + tie + group * tie) p(g * t) | 1328.14 | 14.0 | 0.00 | 0.001 | 38 | 1248.23 |
| Phi(g + t + tie) p(g * t) | 1337.08 | 23.0 | 0.00 | 0 | 37 | 1259.38 |
Phi, survival probability; p, recapture probability; time (t), parameters are allowed to change between capture occasions; tie, width of the black breast band; g, group, referring to forest and urban birds; +, only main factors included in the model; *, main factors and its interaction included in the model.
A better fit of the interaction (*) model than the additive model (+) indicates a difference in slopes.