Skip to main content

Advertisement

Table 2 Results of model averaging across the highest ranked (ΔAIC c  < 2) GLMMs for each response variable (from A to D)

From: Home-range use patterns and movements of the Siberian flying squirrel in urban forests: Effects of habitat composition and connectivity

Model A: length of burst B: speed of burst C: length of nightly track D: number of nest sites
  Estimate SE w Estimate SE w Estimate SE w Estimate SE w
Parameters
Intercept 4.265 0.449   0.897 0.341   −8.484 1.135   −2.441 0.856  
Sexa       1.00    1.00    
   Males 1.083 0.372 1.00 0.640 0.231   1.573 0.486     
Montha       1.00    1.00    
   April −0.390 0.362 1.00 −0.150 0.222   0.565 0.346     
   May 0.201 0.331   0.099 0.203   1.390 0.330     
   June 0.717 0.327   0.383 0.201   1.652 0.325     
   July 0.584 0.335   0.312 0.206   1.908 0.336     
   August 0.774 0.336   0.381 0.207   1.931 0.339     
   September 0.476 0.377   0.257 0.231   1.422 0.374     
H1 (suitable) 0.726 0.484 0.41 −0.038 0.290 0.16 0.057 0.820 0.27 −0.453 0.552 0.22
H2 (movement) 0.700 0.529 0.78 0.293 0.192 0.71 −0.916 0.492 0.63 0.425 0.467 0.24
H3 (urban) 1.295 0.514 1.00 0.748 0.241 1.00 0.852 0.619 0.56    
Sexa*Montha    1.00    1.00    1.00    
   Males: April 0.191 0.425   0.075 0.261   −0.187 0.427     
   Males: May −0.819 0.400   −0.506 0.245   −1.093 0.420     
   Males: June −1.114 0.407   −0.719 0.250   −1.284 0.429     
   Males: July −0.239 0.410   −0.077 0.252   −0.710 0.432     
   Males: August −0.400 0.413   −0.278 0.254   −0.949 0.435     
   Males: September −0.736 0.454   −0.535 0.279   −1.113 0.477     
Sexa*Habitat
   Males: H1        1.813 0.965 0.14    
   Males: H2 0.307 0.500 0.23 0.184 0.309 0.07       
   Males: H3     0.390 0.353 0.21 −1.540 0.704 0.56    
HR           0.239 0.054 1.00
D(burst)     −0.095 0.071 0.35       
D(night)        1.375 0.125 1.00    
E(sites)           0.388 0.156 1.00
  1. Model averaged parameter estimate, associated standard error (SE) and variate weight (w) is provided for each explanatory variable that was chosen among the highest ranked models. Bolded value indicates that 95 % confidence intervals of a parameter estimate does not include zero
  2. aDenotes categorical variable where the effect of the first category (Sex: Females or Month: March) was set at the reference level. The variables are defined in the methods section and the complete list of models used in model averaging is provided in Additional file 4