Skip to main content
. Author manuscript; available in PMC: 2021 Jan 1.
Published in final edited form as: Ibis (Lond 1859). 2020 Nov 2;163(1):274–282. doi: 10.1111/ibi.12892

Table 1.

Summary of the significance of terms included in statistical models (see Methods) explaining the number of gulls at the different feeding grounds. s(Time), time of day (15 min) as a smooth term; s(Time : week), time of day on weekdays as a smooth term; s(Time : weekend), time of day on weekend days as a smooth term; df, degrees of freedom; edf, estimated degrees of freedom for the smooth term. Estimates (with standard error) are given for the terms in the final best-fit model except for the categorical variables with more than two levels, which can be found in the Supporting Information Appendix S1.

Model Response Explanatory df χ 2 P β-coefficients ± se/edf
Park Number of gulls (at time of count) Number of people (continuous) 1 13.9 < 0.001*** −0.135 ± 0.004
s(Time) 6.51 46.94 < 0.001*** 5.41
School Number of gulls (at time of count) Number of people (continuous) 1 11.84 < 0.001*** 0.0055 ± 0.001
Food (factor) 1 7.84 0.005** 0.340 ± 0.135
Number of people: Day of the week 1 4.76 0.029* −0.057 ± 0.008
s(Time) 7.08 120 < 0.001*** 5.915
Waste centre Number of gulls (at time of count) Day of the week (factor) 1 241.5 < 0.001*** −0.917 ± 0.189
s(Time) 5.55 31.768 < 0.001*** 4.715
s(Time : weekday) 4.85 15.695 0.023* 4.086
s(Time : weekend) 11.92 1.799 0.509 1.468
Waste centre Percentage on pile (at time of count) Day of the week (factor) 1 14.745 < 0.001*** 0.323 ± 0.109
Waste-related activity level (factor) 3 1601.00 < 0.001***
Time since waste unload (factor) 6 241.58 < 0.001***

P < 0.1. ***P < 0.001. **P < 0.01. *P < 0.05.