Using Principal Component Factor Scores in Multiple Linear Regression Models to Predict Body Weight of Indigenous Chickens from Morphometric Traits in Bench Maji Zone, Southwestern Ethiopia

Body weight; Correlation; Morphometric traits; Multivariate analysis; Principal component factor analysis

Authors

January 1, 2022

Downloads

Background: There is a rich genetic resource base of indigenous chickens in Ethiopia. However, the productivity of indigenous chicken in the country and their generic base has not been explored sufficiently through characterization using multivariate analysis.
Objective: The objectives of this study were to understand complex interrelations among morphometric traits of indigenous chickens and predict body weight using principal component (PC) factor analysis.
Materials and Methods: A total of 660 (180 males and 480 females) randomly selected chickens of age six months and above were used for the study. Data were collected on body weight and morphometric traits.
Results: In factor solution of the PCA with varimax rotation of the transformation matrix, two principal components (PCs) were extracted (PC1 and PC2) explaining 75.76% of the total variation in the original variables. PC1 had the largest share (62.43%) of the total variance and had its loadings on comb length, wattle length, wingspan, comb height, shank length, and keel length while the PC2 shared only 13.33% of the total variance with positive loadings on body length, back length, and neck length. Prediction model based on PC factor scores accounted for 48% of the variation in the body weight and was more valid than the inter-dependent based models (which accounted for 49% of the variation in the body weight) as it removed multi-collinearity which was present as inter-dependent traits were used in the model.
Conclusion: According to the findings of this study, body weight can be estimated more accurately from PC factor scores than inter-dependent original morphometric traits (i.e. comb height, comb length, wattle length, neck length, back length, body length, wingspan, shank length, and keel length) and the results obtained could be used by chicken producers and researchers for selection, management purposes and estimating market values of the chickens, since weight is the pivotal point on which animal production thrives.