Table 1 Factors impacting the rate of whole-group encounters in ursine colobus monkeys, as well as the propensity of adult males and adult females to participate aggressively in these whole-group encounters.

From: When population growth intensifies intergroup competition, female colobus monkeys free-ride less

Model and fixed effects

b Estimate

SE

χ2

p

− 95% CI

+ 95% CI

Whole-group encounter rate

(Intercept)

− 2.81

0.15

–

–

–

–

Rainfall

− 0.03

0.06

0.28

0.598

− 0.14

0.08

Population size

0.33

0.07

23.01

< 0.001

0.19

0.48

Food availability in home range

− 0.07

0.07

0.89

0.346

− 0.22

0.07

Number adult females

− 0.38

0.11

16.60

< 0.001

− 0.62

− 0.18

Number adult males

− 0.10

0.07

2.09

0.148

− 0.24

0.03

Male aggression in whole-group encounters

(Intercept)

1.33

0.47

–

–

–

–

Rainfall

0.05

0.16

0.12

0.729

− 0.25

0.37

Population size

− 0.42

0.22

3.85

0.049

− 0.88

− 0.00

Food availability in home range

0.67

0.25

–

–

–

–

Number adult females

0.43

0.32

–

–

–

–

Single male vs multi-male

0.10

0.48

0.05

0.830

− 0.84

1.10

Food availability * Number adult females

0.47

0.22

4.75

0.029

0.05

0.90

Female aggression in whole-group encounters

(Intercept)

− 1.03

0.18

–

–

–

–

Rainfall

− 0.44

0.15

9.44

0.002

− 0.72

− 0.16

Population size

1.08

0.19

38.37

 < 0.001

0.72

1.47

Food availability in home range

− 0.32

0.17

–

–

–

–

Number adult females

0.11

0.22

–

–

–

–

Single male vs multi-male

0.07

0.30

0.05

0.831

− 0.57

0.66

Food availability * Number adult females

0.58

0.25

6.06

0.013

0.12

1.09

  1. The whole-group encounter rate model explained half of the variation in IGE rates (R2GLMM(C) = 0.56) and performed significantly better than the null model, which contained the intercept, offset, and random effects (likelihood ratio test: N = 146, χ2 = 53.35, p < 0.001). The models examining male participation (likelihood ratio test: N = 146, χ2 = 15.27, p = 0.018; R2GLMM(C) = 0.55) and female participation in whole-group encounters (likelihood ratio test: N = 146, χ2 = 72.02, p < 0.001; R2GLMM(C) = 0.50), both performed significantly better than the null models (i.e., models with intercept, number of encounters as a weighted term, and random effects only). Significant predictors are presented in bold and trends are italicized.
Back to article page