Table 2.

Multiple ridge regression analyses (ridge parameter, k = 0.02) on the number of hens under different UV-filtering shelters throughout the day. Only variables that significantly contributed to the most parsimonious model are presented.

UV-Filtering Shelter	Flock	Predictor 1	Β-Coefficient (Standardised) ‡	t-Value	p-Value	Adjusted R2 and Model’s F-Statistics	Relative Weight of the Predictors in the Model
50%	A	Ambient temperature	−0.59	−7.16	<0.0001	R2-adjusted = 0.52	24.62%
		Relative ambient humidity	−0.36	−4.42	<0.0001	F2.79, 263.03 = 108.58, p < 0.0001	10.52%
		PAR	−0.56	−13.25	<0.0001		64.86%
	B	Ambient temperature	−0.40	−11.24	<0.0001	R2-adjusted = 0.58	34.26%
		UVAB	−0.29	−3.18	<0.01	F2.53, 320.19 = 156.77, p < 0.0001	32.43%
		IR	−0.28	−3.08	<0.01		33.31%
70%	A	Ambient temperature	−0.77	−8.48	<0.0001	R2-adjusted = 0.40	38.52%
		Relative ambient humidity	−0.46	−5.11	<0.0001	F2.79, 263.03 = 71.33, p < 0.0001	15.64%
		UVAB	−0.41	−8.87	<0.0001		45.85%
	B	Ambient temperature	−0.44	−10.63	<0.0001	R2-adjusted = 0.44	45.57%
		UVAB	−0.33	−2.85	<0.01	F2.68, 319.98 = 71.26, p < 0.0001	18.10%
		PAR	0.13	1.66	0.10		17.91%
		IR	−0.22	−2.70	0.01		18.42%
90%	A	Ambient temperature	−0.93	−9.99	<0.0001	R2-adjusted = 0.35	40.75%
		Relative ambient humidity	−0.68	−7.44	<0.0001	F3.08, 262.54 = 51.13, p < 0.0001	18.85%
		PAR	0.20	1.84	0.07		19.14%
		IR	−0.53	−4.92	<0.0001		21.26%
	B	Ambient temperature	−0.54	−12.28	<0.0001	R2-adjusted = 0.38	71.58%
		UVAB	−0.31	−2.51	0.01	F2.68, 319.98 = 56.45, p < 0.0001	9.20%
		PAR	0.24	2.85	<0.01		9.58%
		IR	−0.15	−1.78	0.08		9.63%

^‡ β-coefficients (standardised) of the predictor variables were estimated separately using the ridge regression coefficient in ‘R’ as the original ridge package did not include the ‘β-coefficient’ value in the regression outputs. ¹ UV_AB (ultraviolet radiation A and B wavelengths), PAR (photosynthetically active radiation), IR (infrared radiation).