Estimation of body weight from some body measures of Saanen kids with regression analysis based on factor analysis scores

Authors : Melis Çelik Güney, Özge Uyguner, Murat Durmuş, Nazan Koluman
Pages : 798-807
Doi:10.37908/mkutbd.1750767
View : 55 | Download : 171
Publication Date : 2025-12-31
Article Type : Research Paper
Abstract :The present study was conducted to estimate the body weight of Saanen kids by using some body measurements. The least squares method and regression analysis were employed, based on factor analysis scores. For this purpose, the body weight of a total of 90 kids (90 days old) and body measurements such as height at withers, rump height, rump width, chest width, chest depth, chest girth, and body length were determined. As a result of the analysis, the highest correlation was between height at wither and rump height, and this correlation was found to be statistically significant. In the analysis conducted with the least squares method, it was determined that the variance inflation factor values for the height at wither and rump height variables were above 10. Because there was a multicollinearity problem in these variables, regression analysis based on factor analysis scores was performed. Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity were used to determine the suitability of the data set for factor analysis. The eigenvalues of the two factors were determined to be greater than 1 based on the results of varimax rotation. The first factor extracted has an explained variance of 70.37% and the second factor of 17.36%. The body weight estimation model obtained with factor scores was found as BW=2.794FS1+1.072FS2. The factor scores obtained through factor analysis have eliminated the multicollinearity problem, as confirmed by the variance inflation factor values falling below 10. This allowed for a more consistent and reliable estimation of body weight in Saanen kids.
Keywords : keçi, faktör analizi, Çoklu doğrusallık, Regresyon

ORIGINAL ARTICLE URL