Regression Tree Approach for Assessing the Effects of Non-Genetic Factors on Birth Weight of Hemşin Lamb

Abstract

This study examined effects of birth year, herd type, age of dam, lamb sex, birth type and lamb color onbirth weight in Hemsin lambs using the regression tree method. The study used the data obtained from 9113 Hemsinlambs born between 2007-2010. The average birth weight of Hemsin lambs was found to be 3.55 kg and the standarddeviation was 0.727 kg. Significant differences were found in birth weights of Hemsin lambs according to birth year(P<0.001), birth type (P<0.001) and lamb sec (P<0.01), whereas no statistically significant difference was foundaccording to herd type, age of dam and lamb color (P>0.05). According to the regression tree diagram, the mostimportant factors affecting birth weight of Hemsin lambs were birth year, birth type, sex, herd type and age of dam.The absence of lamb color in the regression tree diagram indicates its insignificant effect on birth weight. The low riskvalue (0.456) and high R-Squared value (0.862) of the regression tree model shows that the model is sufficient toexplain birth weight. Variables of birth year, birth type, sex, herd type and age of dam explained 86.2% of the variationin birth weight.