Table 6.

Summary of 5000 model (binomial GLMM) estimates of the probability that the female (vs. male) incubates in relation to time of a day, incubation period, and start of incubation within the season

Fixed effects	Estimate	95% CI	Number of iterations (P < 0.05)	P
(Intercept)	0.059	(−0.021, 0.14)	318	0.94
cos(time)	0.915	(0.757, 1.079)	4998	0.0004
sin(time)	0.755	(0.61, 0.906)	5000	0
Day of incubation	0.007	(−0.01, 0.024)	101	0.98
Start of incubation	0.006	(−0.006, 0.018)	133	0.97
Start of incubation × Day of incubation	0	(−0.002, 0.003)	22	1
cos(time) × Day of incubation	−0.066	(−0.098, −0.034)	4494	0.10
cos(time) × Start of incubation	0.02	(−0.003, 0.044)	0	1
cos(time) × Day of incubation × Start of incubation	0.003	(−0.002, 0.008)	284	0.94
sin(time) × Day of incubation	0.109	(0.079, 0.139)	5000	0
sin(time) × Start of incubation	0.026	(0.006, 0.046)	0	1
sin(time) × Day of incubation × Start of incubation	−0.012	(−0.017, −0.007)	4928	0.0144
Random effects	Variance	95% CI
Nest (intercept)	0	(0, 0)
cos(time)	4.67	(3.6, 6.0)
sin(time)	2.37	(1.84, 3.04)

Median (range) N = 1722 (1680–1768) incubation data points from 48 nests. Time of a day (in radians) was decomposed into sin and cos; the remaining fixed effects were mean centered. Estimates are means, and their 95% CIs (nonparametric) are 0.025 and 0.975 quintiles, of the fixed effect estimates of 5000 GLMMs. Variances are mean values, and 95% CIs are 0.025 and 0.975 quintiles, of the random effects from 5000 GLMMs. The median (range) partial autocorrelation coefficient of GLMMs was 0.038 (0.001–0.074) in lag 1 and −0.264 (−0.311 to −0.214) in lag 2.