Behavioural response of a migratory songbird to geographic variation in song and morphology

Table 4 Number of tail flips

	Fixed effects	Estimate	s.e.m	t	CI 2.50%	CI 97.50%
(a) playback	stimulus	−16.65	1.57	−10.58	−19.82	−13.45
	local	21.34	25.35	0.84	−30.85	73.32
	90 km	−37.68	6.18	−6.09	−49.77	−25.20
	180 km	−57.62	6.16	−9.35	−70.01	−45.05
	African	−54.67	6.17	−8.86	−67.06	−42.01
	control	−74.16	6.46	−11.48	−87.36	−60.74
	breeding stage	23.13	10.37	2.23	1.93	44.08
	trial order	−3.10	1.44	−2.16	−5.97	0.44
	date	2.75	1.45	1.89	−0.18	5.69
	time	0.49	0.50	0.97	−0.54	1.51
(b) decoy	stimulus	−18.43	14.13	−1.30	−48.35	11.16
	local	91.14	75.87	0.80	−102.65	224.57
	European	−10.80	43.36	−0.25	−106.25	83.18
	African	−52.38	43.82	−1.20	−146.92	40.28
	control	−48.50	44.02	−1.10	−144.07	46.80
	breeding stage	−2.78	20.12	−0.14	−46.04	39.67
	trial order	59.11	16.39	3.61	24.44	93.31
	date	−15.99	11.76	−1.36	−41.19	8.33
	time	−0.28	1.00	−0.28	−2.40	1.82

Results of general linear mixed model with estimates, standard error, t-value, and credible intervals fitted by maximum likelihood. Estimates for the stimulus locations refer to differences from the intercept estimates, which represent the number of tail flips of the local population. Subjects were included as random intercepts to control for repeated measures. A behavioural response differs significantly from the local population if its credible intervals do not include the mean of the local population. Significant results are shown in bold.

ISSN: 1742-9994