Figure 5

Individual modelling in SSFs. SSF estimates computed at the individual level can be further analysed with common statistical packages to make inferences about the effects of additional covariates on habitat selection. In example a), age of monitored animals are plotted on the x-axis, while individual selection coefficients β estimated with SSFs (say selection for roads) are plotted on the y-axis. In this case, there is a clear increase in the avoidance of roads in older individuals, and this pattern can be analysed with a linear regression, a generalized linear model, or a generalized additive model. In example b), selection coefficients estimated with SSFs (say selection for open areas) are plotted for females with or without offspring. The effect of offspring on selection for open areas by mothers can be tested with an independent sample t-test, for instance, or using generalized linear models if other covariates are available (say the age of the female).