Abstract

In marginal areas where extensive goat production systems are developed, it is common not to have the technology that allows weighing animals. The aim of this study was to develop equations to predict body weight as a function of morphometric variables. A total of 382 records of “criollo” goats of different ages and sex were analysed, measuring live weight and 15 morphometric variables. Correlation analysis and logarithmic regression were used to analyse the data, selecting only the variables presenting a higher association with body weight to include them in the logarithmic regression model, which was calculated using a “stepwise” procedure. The equations obtained were highly significant (p < 0.05) and explained a high proportion of the variability (R2 > 85%). The predictor variables for live weight were thoracic perimeter and back sternal and/or longitudinal diameters, independently of the sex of the goat. The use of the thoracic perimeter only as the predictor of the live weight of the goat required an equation for each sex.