The Use of the Method of Partial Regression in the Analysis of Comparative Feeding Trial Data, Part I

Abstract

Statistical analysis by the method of partial regression has been employed to determine the relationship between gains and initial weight and feed consumption of swine of four different age groups. Highly significant relationships between gains and feed consumption are exhibited by all groups, the increase in gain due to each additional unit of feed consumed decreasing with advancing age of pig. In the case of weanling pigs and young pigs fed to market weight, the gain made is also dependent upon the initial weight of the pig. The gain-feed ratio commonly employed does not efficiently correct the observed final weights for variations in initial weight and feed intake. A satisfactory correction may, however, the made by means of the regression coefficients, markedly reducing the variance of the results without introducing errors inherent in the gain-feed ratio. This procedure is also applicable to comparative feeding trials, in which the animals are divided into two or more differently treated lots.