Table 1 Summary of statistical models, showing fit indices (marginal R2 and Akaike Information Criterion AIC), the between-group variances for the random effects (τ00 for subject and year_collection), and point estimates, 95% confidence intervals and p-values of predictors and their interactions (if included).

From: Foraging postures are a potential communicative signal in female bonobos

Model

Predictor

Estimate

95% CI

p

p*

All individuals

R2 = 40.2%

AIC = 2,325.9

τ00 subject = 0.61

τ00 year_collection = 0.16

Intercept

0.13

(− 0.77, 1.02)

0.78

Sex (M)

0.63

(− 0.51, 1.76)

0.28

Duration

0.56

(0.68, 0.43)

 < 2.16 × 10–16

Age_class (mature)

1.03

(0.08, 1.98)

0.034

Sex (M) : duration

0.72

(1.01, 0.42)

1.99 × 10–6

5.21 × 10–7

Sex (M) : age_class (mature)

2.31

(3.60, 1.02)

0.0004

0.0016

Mature females

R2 = 20.2%

AIC = 942.9

τ00 subject = 0.20

τ00 year_collection = 0.41

Intercept

1.51

(0.59, 2.43)

0.0013

Swelling (linear)

1.26

(0.89, 1.63)

2.59 × 10–11

2.73 × 10–13

Swelling (quadratic)

0.47

(0.86, 0.08)

0.018

Duration

0.55

(0.72, 0.39)

9.69 × 10–11

  1. The estimates and 95% CIs are in log odds of the position being “up”. For interactions in the “all individuals” model, each p* is the p-value of the whole interaction, while for the “mature females” model, p* is the p-value of the whole linear and quadratic effects. Bold = significance at α-level 0.05.