Table 4.

Results of the generalised linear mixed model (on the cloglog-scale) explaining the probability of engaging in digging behaviour (binomial distribution), as a function of nesting goose density, goose reproduction period (incubation versus brooding), interaction between goose density and reproduction period, as well as fox sex, reproductive status and their interaction (n = 55,434 30-s bursts of accelerometry collected on 23 fox-years)

Fixed effect	Estimate [95% CI]	SE	z-value	p-value
Intercept	− 1.35 [− 1.68, − 1.01]	0.16	− 8.31	< 0.001
Nesting goose density	0.16 [0.14, 0.18]	0.01	15.54	< 0.001
Goose reproduction period	0.04 [0.01, 0.07]	0.02	2.50	0.01
Density: period	− 0.03 [− 0.06, − 0.01]	0.01	− 2.38	0.02
Fox sex	0.59 [0.10, 1.07]	0.23	2.56	0.01
Fox reproductive status	0.03 [− 0.07, 0.12]	0.05	0.55	0.59
Sex: status	− 1.05 [− 1.16, − 0.94]	0.06	− 18.05	< 0.001

We used as reference values nesting goose density = average, period = incubation, sex = male, and reproductive status = reproductive. Nesting goose density, a proxy for goose nest density, was centered and standardised in the model. Variance and standard error were 0.21 and 0.46 for fox ID and 0.005 and 0.07 for year