This study was conducted to investigate the relationships among morphological traits, body weight, carcass
traits, and carcass primal cuts†of goats and to fit the best regression equation for goat body weight, carcass traits
and carcass primal cuts prediction. Body weight, carcass traits such as slaughter weight, empty weight, hot
carcass, dressing percentage and carcass primal cuts: thin cut, loin, leg, rib, and fore leg+ shoulder+ neck were the
dependent variables. Morphological traits: body length carcass primal cuts (thin cut, loin, leg and leg+shoulder+neck)
(BL), wither height (WH), heart girth (HG),†paunch girth (PG),†rump height (RH), hip width (HW), pin bone
width (PBW), neck girth (NG), scrotum circumference (SC) and Scrotum length (SL) were an important parameters
used for prediction of the equation. Crossbred Boer goats exhibited higher (p<0.05) body weight, carcass
traits, and carcass primal cuts than pure Central Highland (CH) goat except for dressing percentage. Body weight
and carcass traits for both genotypes had a moderate to higher significant correlation with all morphological
traits (r= 0.42 to 0.94) except with SL and SC. For both genotypes two principal components were extracted and
which in a multiple regression analysis explained 80.17% of the total variance for CH with Central Highland goat
and 82.25% of the total variance for crossbred Boer goat. HG, NG, BL, and PG were the most important (R2 = 0.71)
morphological traits used to predict body weight, carcass traits, and carcass primal cuts of Central Highland goat,
whereas HG, BL, HW, and SL were the most important (R2 = 0.84) predictor for crossbred Boer goat. Therefore,
this study suggests that morphological traits could be a suitable criteria for early selection of bucks for their body
weight and carcass traits without slaughtering. The prediction of body weight and carcass traits based on principal
component factor scores is more reliable than the use of individual morphological traits, because the uses of factor
scores in multiple regression models get rid of the problem of the interdependency of explanatory variables, thereby
improve the accuracy of interpretation of the regression results.
Keywords
Genotype, morphological trait, principal component analysis, selection