Elaboration d'une méthode de test d'interaction digestive entre les composants du régime chez les animaux monogastriques

Open Access
Publisher Website
Google Scholar
Abstract
Estimation of the nutritive value of a compound feed is generally based on the additivity of each its ingredients. However, this approach can only be used in the absence of any significant digestive interaction. The described method was used to test statistically the possible existence of such an interaction between two ingredients even when using more than two compound feeds and whatever the form of interaction. The method involved two steps: The first one consisted of testing "diet effect" by an analysis of variance which also included an "animal effect". In the case of repeated measurements, it was also possible to test the "animal-diet" interaction. When there was no diet effect, there was no digestive interaction. If the "diet effect" was significant, it was necessary to use a second step, i.e. a regression method. The accuracy provided with the analysis of variance over the regression allowed to test the possible existence of a digestive interaction. This method could also be used when the measurements were not repeated or when the experimental device was unbalanced. Effects linked to animals or animal-diet interaction could also be easily tested by means of this method.

This publication has 0 references indexed in Scilit: