Table 3 

Multiple linear regression results. 

Mean F 
SD F 
CV F 
Mean S 
SD S 
CV S 



R^{2} 
0.56 
0.47 
0.40 
0.58 
0.70 
0.71 
F(4,12) 
3.87 
2.69 
2.00 
4.19 
7.16 
7.37 
p 
0.032 
0.082 
0.157 
0.024 
0.0035 
0.0031 
pFDR 
0.048 
0.098 
0.157 
0.048 
0.0105 
0.0186 


Age 
N. S. 

b(Hz/yr) 
14.7 
4.0 
50.4 
32.6 
0.01 (no units) 

t 
3.3 
2.63 
3.7639 
t = 4.97 
4.04 

p 
0.006 
0.02 
0.0027 
0.0003 
0.0016 



Dom 
N.S. 
N. S. 
N. S. 
N. S. 
N.S. 

t 
2.93 

p 
0.012 



Sex 
N.S. 
N. S. 
N. S. 
N. S. 
N. S. 
N. S. 


Treat 
N.S. 
N. S. 
N. S. 
N. S. 
N. S. 
N. S. 


Age*Dom 
Not done 

F(1,13) 
3.24 
0.01 
2.4 
0.09 
5.07 

p 
N.S. 
N.S 
N. S. 
N. S. 
0.04 



The table shows the statistical results of the linear regression analysis performed for the mean fundamental (Mean F), the standard deviation of the fundamental (SD F) the coefficient of variation of the fundamental (CV F), the mean spectral density (Mean S), the standard deviation of the spectral density (SD S) and the CV of the spectral density (CV S) as separate dependent variables and age Dominance (Dom), Sex, Treatment (Treat) as predictor variables. The pvalues are also shown corrected for multiple testing using the false discovery rate adjustment. Age is correlated with both the mean fundamental frequency and its standard deviation; they both decrease as the animal ages. The mean, SD, and CV of the mean spectral value (S) also decrease with age. The Age*Dom shows the results for the interaction between age and dominance performed with the posthoc ANCOVA. The cells shown in bold italic are the significant results that are illustrated in the figures of the paper. The linear regression between Mean F and Age is shown in figure 2D and the regression and interaction between CV of S and Age and Dominance is shown in figure 3B. N.S. = Not significant. 

Mathevon et al. BMC Ecology 2010 10:9 doi:10.1186/14726785109 