Table 3.

Model selection results from conditional logistic regression of the effects of snow conditions, prey species, and forest attributes on the likelihood of path use by Canada lynx (Lynx canadensis) and bobcat (Lynx rufus)

Species	Predictors	K	AICc	ΔAICc	AICc weight	R 2
Lynx	Snow + Hare + Forest	6	32.09	0	0.539	0.496
	Snow + Hare + Alternative Prey	7	34.43	2.34	0.167	0.499
	Hare + Forest	4	34.91	2.82	0.132	0.315
	Snow + Hare + Alternative Prey + Forest	9	35.07	2.98	0.122	0.609
	Hare + Alternative Prey + Forest	7	37.25	5.16	0.041	0.431
Bobcat	Snow + Hare + Forest	6	29.96	0	0.969	0.566
	Snow + Hare + Alternative prey + Forest	9	39.86	9.90	0.025	0.577
	Snow + Hare + Alternative prey	7	41.26	11.30	0.005	0.384
	Hare + Forest	4	43.14	13.18	0.001	0.099
	Hare + Alternative prey + Forest	7	49.16	19.20	0.000	0.166

K is the number of model parameters. AICc is the Akaike information criterion corrected for small sample size, ΔAICc is the difference in AICc between each model and the top model, AICc weight indicates the likelihood of the model being the best model given the overall model set, and R ² is McFadden's pseudo R ². Snow includes snow depth and snow hardness; hare is snowshoe hare; alternative prey includes deer, squirrel, and grouse; and forest includes coniferous forest and immature forest.