Akademik Veri Yönetim Sistemi
PAKISTAN JOURNAL OF STATISTICS, cilt.28, sa.1, ss.159-165, 2012 (SCI-Expanded)
The aim of this study is to predict carcass weight from some morphological traits (total length, fork length, standard length, head length, body height and weight) by using jointly factor and multiple regression analyses. A total of 91 Capoeta capoeta capoeta fish was used to estimate carcass weight. The suitability of factor analysis was determined with Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy (0.87) and Bartlett's test of sphericity (P < 0.01). VARIMAX rotation was used to facilitate interpretation of factor loadings (L-ik). Other morphological traits (total length, fork length, standard length, head length, body height, and weight) except for carcass weight were exposed to factor analysis. As a result of factor analysis, three latent variables were obtained from six morphological traits and considered as independent variables in multiple regression analysis. Additionally, carcass weight was used as a dependent variable in multiple regression analysis. The developed model was determined as CW = 0.484 FS1 - 0.324 FS2 + 0.755 FS3. The obtained results shown that, the three selected factors had significant effects (P < 0.01) and explained 95.3% of variation in carcass weight. With using factor scores in the multiple regression analysis, carcass weight was predicted successfully by using these morphological traits.