Table 1.
Model selection of linear mixed models with telomere change in response to the variables (sex, year and age as well as the interactions between these variables) using Akaike's information criterion corrected for small sample size (AICc) and Akaike weights. d.f. is the number of parameters. ΔAICc is the difference in AICc values between the models, with the best-candidate model having the lowest AICc. The evidence ratio is the ratio of the Akaike weight of each model to that of best model, indicating the extent to which the best model is better than the model. All models with ΔAICc values of less than 10 are listed. The null model includes only the intercept and the random effect (i.e. individual ID).
| variables | d.f. | AICc | ΔAICc | Wi | evidence ratio |
|---|---|---|---|---|---|
| year | 6 | 122.1 | 0 | 0.66 | 1.0 |
| year, sex | 7 | 125.1 | 3.0 | 0.15 | 4.4 |
| null | 3 | 125.5 | 3.4 | 0.12 | 5.4 |
| sex | 4 | 128.0 | 5.9 | 0.04 | 18.9 |
| year, sex, year : sex | 10 | 129.5 | 7.4 | 0.02 | 38.8 |
| year, age | 7 | 130.6 | 8.5 | 0.01 | 66.0 |