Table 2 Model averaging results for the best candidate models from QAICc-based (Quasi Akaike Information Criterion Corrected for small sample size) model selection (delta < 2) for large, small and all (pooled data) mammal species in Iran. Relative importance values represent the comparison of the contribution of each covariate in the original models and the simulated models. β is the estimated beta coefficient. Variable codes explained in Table 1.

From: Hunting and persecution drive mammal declines in Iran

Model

Parameter

β ± unconditional SE

Confidence interval

Exp (β)

P-value

Relative importance

Original

Simulation

All mammals (n = 156)

 

Intercept

− 0.93 ± 0.14

(− 1.21, − 0.65)

0.39

0.00

–

–

AET

0.18 ± 0.09

(0.00, 0.35)

1.19

0.05

0.75

0.75

DB

0.14 ± 0.12

(− 0.10, 0.39)

1.15

0.24

0.25

0.25

HB

− 0.08 ± 0.18

(− 0.42, 0.27)

0.92

0.66

0.09

0.09

HII

0.14 ± 0.10

(− 0.06, 0.34)

1.15

0.17

0.39

0.39

HV

0.64 ± 0.09

(0.46, 0.82)

1.89

0.00

1.00

1.00

LS

− 0.42 ± 0.14

(− 0.70, − 0.14)

0.65

0.00

1.00

1.00

Large mammals (n = 39)

 

Intercept

0.13 ± 0.18

(− 0.23, 0.975)

1.13

0.49

-

-

ABM

0.33 ± 0.12

(0.09, 0.58)

1.39

0.01

1.00

1.00

AET

0.25 ± 0.11

(0.02, 0.47)

1.28

0.03

0.75

0.75

HB

− 0.43 ± 0.32

(− 1.08, 0.23)

0.65

0.20

0.28

0.28

HV

0.41 ± 0.17

(0.06, 0.76)

1.50

0.02

0.87

0.87

LS

− 0.39 ± 0.19

(− 0.77, − 0.01)

0.67

0.04

0.73

0.73

Small mammals (n = 117)

 

Intercept

− 1.41 ± 0.19

(− 1.80, − 1.02)

0.24

0.00

–

–

ABM

0.09 ± 0.17

(− 1.79, − 1.02)

1.09

0.59

0.05

0.05

AET

− 0.28 ± 0.20

(− 0.67, 0.12)

0.75

0.17

0.51

0.51

DB

− 0.25 ± 0.21

(− 0.68, 0.17)

0.77

0.24

0.23

0.23

GL

0.29 ± 0.17

(− 0.06, 0.63)

1.33

0.10

0.45

0.45

HII

0.38 ± 0.13

(0.12, 0.64)

1.46

0.00

1.00

1.00

HV

− 0.18 ± 0.22

(− 0.61, 0.25)

0.83

0.41

0.17

0.17

LS

− 0.26 ± 0.19

(− 0.64, 0.13)

0.77

0.19

0.15

0.15

Back to article page