JOURNAL OF APPLIED ANIMAL RESEARCH, cilt.25, sa.2, ss.97-100, 2004 (SCI-Expanded)
In this research, linear regression models were used for estimation of body weights from various body measurements in Morkaraman sheep. Simple and multiple regression models were fitted with body weight (BW) as dependent variable and body length (BL), heart depth (HD), height at wither (HW), chest width (CW), heart girth (HG) and pump width (PW) as independent variables. The model including heart girth was the best fitted simple regression model (R-p(2) =0.752, MSEp =10.944, Cp =8.777) and the model including heart girth and pump width was the best fitted multiple regression model (R-p(2) =0.784, MSEp =9.678, Cp =1.846). The highest correlation (0.867) was determined between HG and BW and the lowest correlation (0.014) was determined between CW and BL. According to variance inflation factors, multicollinearity among body measurements was not determined.