FISHERY BULLETIN: VOL. 83. NO. 2 



Table 9. — Correlation matrix for selected otolith morphometric, weight, and age data 

 for the calibration subsample of Sebastes pinniger. 



Table 10. — Regression coefficients and associated 

 statistics on the multiple regression models of whole 

 otolith age for Sebastes pinniger. 



cient of determination (R ) suggests that the mod- 

 els of whole otolith age explain 83.4% of the varia- 

 tion in age for females and 81.0% for males. For 

 both males and females, the constant in the re- 

 gression was not significantly different from zero 

 and was not included in the models. The residuals 

 from the models show no distinct trend with the 

 exception of a slight increase at ages >17 yr for 

 males; this included 11.1% of the sample. 



The variables included in the otolith section age 

 models, their coefficients, standard errors, and 

 significance levels are presented in Table 11. As in 

 the whole otolith age models, there are fewer vari- 

 ables included than for S. diploproa; for the male 

 section age model, for example, there is only one 

 variable and the constant included for prediction 

 of age. All variables are highly significant and the 

 coefficients of determination suggest that the 

 otolith section models explain 70.2% of the varia- 

 tion in age for females and 84.6% for males. Mean 

 residuals show a strong trend of increase at ages 

 past 26 yr for male otolith section age models; this 

 represented 23% of the sample. 



A model incorporating both otolith section age 

 and whole otolith age was developed using the 



same criteria for age as in S. diploproa. These 

 models were based upon more independent vari- 

 ables but were not significantly better (as based 

 upon the coefficient of determination) than the 

 otolith section models (Table 12). Based upon the 

 multiple correlation coefficients, the best models 

 for S. pinniger would be the hybrid model for 

 males and the whole otolith model for females. 

 For analyzing the precision of the models, sub- 



Table 11. — Regression coefficients and associated 

 statistics on the multiple regression models of otolith 

 section age for Sebastes pinniger. 



Table 12. — Regression coefficients and associated statistics on 

 the multiple regression models of age in Sebastes pinniger. The 

 ages used for the calibration of these models are based upon 

 either whole otoliths or otolith sections as described in the text. 



112 



