Table 2. GLMM for the number of great tits at bird-feeders in relation to temperature, habitat (urban or rural), presence of novel object, density of great tits, number of cats, number of dogs, number of pedestrians and the interaction between habitat and presence of object as fixed effects.

Effects	Estimate	SE	z	P	Effect size
Intercept	0.399	0.324	1.233	0.218
Novel object	−0.410	0.093	−4.396	<0.0001	0.348
Temperature	−0.017	0.010	−1.769	0.077	0.140
Habitat: Urban	0.315	0.091	3.466	<0.0001	0.274
Early or late trial	0.176	0.102	1.716	0.086	0.136
Density of great tits	0.078	0.007	10.667	<0.0001	0.843
No. species of birds	0.058	0.013	4.451	<0.0001	0.352
No. cats	0.113	0.037	3.061	0.002	0.242
No. dogs	−0.124	0.023	−5.317	<0.0001	0.420
No. pedestrians	0.009	0.003	3.490	0.005	0.276
Habitat * Novel object	0.535	0.119	4.485	<0.0001	0.355

City was used as a random effect with a variance of 1.16 and a standard deviation of 1.08. The number of observations was 160 and the number of cities 14. Effect size is the z-transformed Pearson product-moment correlation coefficient. Statistically significant terms are shown in bold font.