Table 3. Summary statistics of posterior samples for regression coefficients of the 5 best models.

Model		Intercept	Tibetan	Female	Herdpeople	Ratio meadows	Ratio Forest
1	Mean	1.18	0.77	0.28	0.12	−8.72	−10.50
	CI 95%	(−6.36, 8.32)	(0.17, 1.52)	(0.11, 0.45)	(−0.12, 0.36)	(−18.95, 1.65)	(−18.51, −2.47)
	p	0.380	0.005	<0.0001	0.159	0.952	0.004
2	Mean	1.51	0.76	0.28	0.12	−9.20	−10.52
	CI 95%	(−5.99, 9.57)	(0.18, 1.47)	(0.11, 0.45)	(−0.11, 0.36)	(−20.73, 1.06)	(−18.11, −2.92)
	p	0.342	0.003	<0.0001	0.155	0.965	0.007
3	Mean	1.30	0.76	0.28	0.12	−10.28	−10.03
	CI 95%	(−6.75, 9.2)	(0.12, 1.52)	(0.1, 0.46)	(−0.12, 0.35)	(−21.03, 0.82)	(−20.83, 0.08)
	p	0.378	0.013	0.004	0.144	0.966	0.026
4	Mean	0.68	0.78	0.28	0.11	−8.72	−9.40
	CI 95%	(−6.7, 8.08)	(0.1, 1.57)	(0.11, 0.44)	(−0.14, 0.35)	(−18.42, 1.11)	(−18.41, −0.79)
	p	0.422	0.011	<0.0001	0.180	0.953	0.020
5	Mean	0.96	0.78	0.28	0.12	−9.79	−9.17
	CI 95%	(−6.38, 9.02)	(0.1, 1.49)	(0.11, 0.44)	(−0.12, 0.37)	(−20.99, 0.23)	(−19.17, 0.45)
	p	0.413	0.012	0.001	0.143	0.971	0.029

p, the probability that the coefficient be ≤0 (Tibetan, female, herd people, meadows) or ≥0 (forest); CI 95%, 95 percent credibility interval.