DETERMINATION OF PREDICTION EQUATIONS TO ESTIMATE BODY CONDITION SCORE FROM BODY SIZE AND TESTICULAR TRAITS OF YANKASA RAMS

The study was aimed to develop prediction models using stepwise multiple linear regression analysis for estimating the body condition score (BCS) from the body weight (BW), testicular length (TL), testicular diameter (TD) and scrotal circumference (SC) of indigenous Yankasa rams. Data were obtained from 120 randomly selected rams with approximately two and half years of age, from different extensively managed herds in Nasarawa State, Nigeria. Although pairwise phenotypic correlations indicated strong association (P<0.01) among the measured variables, there was collinearity problem between BW and SC as revealed by the variance inflation factors (VIF) and tolerance valves (T). The VIT was higher than 10 (VIF = 19.45 and 16.65 for BW and SC, respectively). The Twas smaller than 0.1 (T = 0.05 and 0.06 in BW and SC, respectively). BW was retained among the collinear variables, and was singly accounted for 83.7% of the variation in BCS. However, a slight improvement was obtained from the prediction of BCS from BW and TL [coefficient of determination (R2), adjusted R2 and root mean squares error (RMSE) were 85.3%, 85.1% and 0.305, respectively]. The prediction of the BCS of Yankasa rams from BW and testicular measurements could therefore be a potential tool for sustainable production and improvement of small ruminants in Nigeria.