Table 1 Random effects standard deviations from the Bayesian logistic regression model for the web data; ordered from highest to lowest standard deviation; larger SDs indicate that accuracy levels varied more for that random effect variable.

From: Novel vocalizations are understood across cultures

Random effect term

SD

95% Bayesian CI

Meaning

1.12

[0.84, 1.50]

Vocalization

0.56

[0.46, 0.68]

Listener

0.33

[0.30, 0.35]

Language

0.26

[0.17, 0.37]

Language family

0.24

[0.02, 0.56]

Creator of vocalization

0.13

[0.01, 0.40]