Morphometric-Based Body Weight Prediction in Red Sokoto and West African Dwarf Goats Using Regression and Classification and Regression Tree

Résumé fourni par la source

Linear body measurements and live body weight were obtained from 633 adult goats comprising 394 West African Dwarf (WAD; 282 females and 112 males) and 239 Red Sokoto (RS; 68 females and 171 males) sampled from selected goat markets in Osun, Ondo and Oyo States, South-West Nigeria. This study compared sex- and breed-specific correlation patterns among morphometric traits and compared the predictive traits usings multiple regression and classification and regression tree (CART) models for body weight (BW) estimation. Body weight and twenty-three morphometric traits were measured, and data were analysed using Pearson’s correlation, backward stepwise multiple regression, and CART algorithms. Across both breeds and sexes, body weight showed strong and consistent associations with trunk and chest dimensions, particularly heart girth, chest depth, and body depth, highlighting the central role of thoracic development in determining live weight. Predictive models based on selected morphometric traits accounted for most of the observed variation in body weight (P < 0.001), although the relative contribution of individual traits differed between sexes and breeds, reflecting sexual dimorphism and breed-specific body conformation. Tree-based analysis confirmed the regression findings, identifying HG as the primary discriminator of BW in both breeds, followed by CD and BD. The regression equations and CART models generated in this study provide reliable, field-applicable tools for objective BW estimation and informed selection decisions in WAD and RS goats where weighing scales are unavailable.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.