Table 1

Deviance tables obtained after fitting binary responses to playbacks to a logistic regression model.

VariableCoefficientDeviance*dfSignificance
Bunching
Variables included in final model
Family 33.520P = 0.029
Age of matriarch−0.5148.061P = 0.005
Association index98.02.111P = 0.147
Age of matriarch × association index−4.316.501P = 0.011
Variables excluded from final model§
Mean age of other females in group| −0.2011.341P = 0.248
Number of females in group| 0.0330.031P = 0.867
Number of calves in group 0.0150.001P = 0.946
Age of youngest calf in group0.0321.691P = 0.194
Presence or absence of adult males−0.8511.921P = 0.166
Smelling
Variables included in final model**
Age of matriarch0.01880.641P = 0.423
Association index97.27.741P = 0.005
Age of matriarch × association index−2.548.261P = 0.004
Variables excluded from final model§
Family 7.020P = 0.997
Mean age of other females in group| 0.02100.181P = 0.669
Number of females in group| −0.03090.291P = 0.590
Number of calves in group −0.0930.791P = 0.374
Age of youngest calf in group0.01481.111P = 0.291
Presence or absence of adult males−0.2460.371P = 0.543
  • * Deviances reported are those obtained when significant variables are dropped and when nonsignificant variables are added to the final model.

  • The final model explains a deviance of 56.8 out of a total deviance of 190.3.

  • Twenty-one families were fitted to the model as a factor, and hence coefficients for individual families are not reported.

  • § All interactions between these variables and the age of the matriarch were not significant.

  • | The number and age of adult females present (“adult” defined by having reached the age of first conception: 11 years and over) were noted for every playback.

  • Calves were individuals of up to 3 years in age.

  • ** The final model explains a deviance of 10.7 out of a total deviance of 211.9.