Table 1 Results of statistical analyses.

From: Habitat preferences, estimated abundance and behavior of tree hyrax (Dendrohyrax sp.) in fragmented montane forests of Taita Hills, Kenya

(a) Estimated regression parameters, standard errors (SE), z-values and P-values for the ZINB GLMM with model: calls—size*time + zq99 + cc10 with all the forests of Taita Hills

 

Estimate

SE

z-value

P-value

Intercept

4.488580

0.18664

24.049

< 2E − 16

Size

− 0.407968

0.25540

− 1.597

0.00180

zq99

0.317512

0.152841

2.077

0.16564

cc10

0.334376

0.242165

1.381

0.46496

(b) AIC and ∆AIC with step1 function with the model calls—size*time + zq99 + cc10 with all the forests of Taita Hills

 

Df

AIC

∆AIC

 

None

 

2805.7

0

 

Size*time

11

2891.3

85.6

 

cc10

2

2902.2

96.5

 

zq99

2

2924

118.3

 

(c) Estimated regression parameters, standard errors (SE), z-values and P-values for the NB GLMM with model: calls—size*time + zq99 + cc10 from Mbololo, Ngangao and Vuria Forests

 

Estimate

SE

z-value

P-value

Intercept

4.498331

0.181644

24.765

< 2E − 16

zq99

0.339377

0.147054

2.308

0.03776

Size

− 0.375196

0.241463

− 1.554

0.11019

cc10

− 0.274987

0.228903

1.201

0.16735

(d) AIC and ∆AIC with step1 function with the model calls—size*time + zq99 + cc10 from NB GLMM analysis

 

Df

AIC

∆AIC

 

None

 

2102.1

0

 

Size*time

11

2109.9

7.3

 

cc10

2

2121.0

18.9

 

zq99

2

2147.4

45.4

 

(e) Estimated regression parameters, standard errors (SE), z-values and P-values for the Bernoulli GLMM model of singing probability with forest size as the only variable

 

Estimate

SE

z-value

P-value

Intercept

− 2.3104

0.2481

− 9.313

< 2E − 16

Size

0.8513

0.2210

3.853

0.000117

Back to article page