Multivariate analysis of characteristics associated with performance of dairy goats fed integral mango meal

• The high values of factor analysis indicated increases in milk production and composition as it increased the number of daily meals.
• The factor analysis showed a linear increase in the rumination and chewing activities.
• The latent variable composed of metabolic characteristics and milk performance did not detect the effect of the replacement of corn with integral mango meal.
• The total replacement of corn by integral mango meal may be advantageous during periods of feed scarcity, since the metabolic profile was not altered by diet.

Factor analysis was used for reduction of dimensionality and to obtain latent variables, constructed from linear combinations of feeding behavior characteristics, physiological and performance, to evaluate the effects of diets containing integral mango meal (IMM) replacing corn (0, 330, 660, and 1000 g/kg) for dairy goats. Eight lactating Saanen crossbred goats (48.7 ± 1.99 kg of BW) were introduced in the experiment 48 days postpartum and maintained up to 124 days of lactation. The latent variable composed of behavioral characteristics and milk performance showed a cubic effect for the tested diets (P < 0.05), with a maximum point for milk performance with 78 g/kg of replacing corn with integral mango meal, and a minimum point with 763 g/kg replacing corn with integral mango meal. Similarly, there was a linear increase (P < 0.01) in rumination and chewing activities with replacement levels. The latent variable composed of metabolic characteristics and milk performance did not (P > 0.05) detect the effect of the replacement of corn with integral mango meal. From the factor analysis, it was possible to reduce the dimensionality of the set of data. Considering that the feeding behavior characterized by the amount of meals and activities of rumination interfered the performance of dairy goats, may be recommended replacing 78 g/kg of corn by integral mango meal.