Application of principal component factor analysis to identify key morphological traits in Garut ewes

Abstract

This research utilized principal component factor analysis (PCA) to pinpoint significant morphological characteristics affecting the body structure of Garut ewes, a native Indonesian breed crucial for smallholder agriculture. The study analyzed data from 85 mature ewes, focusing on body weight (BW) and six linear measurements: body length (BL), chest depth (CD), chest girth (CG), rump width (RW), rump height (RH), and withers height (WH). Descriptive statistics indicated moderate variability in BW (mean = 47.84 kg, CV = 3.75%) and greater variation in skeletal traits (RH and WH CV > 20%). Correlation analysis revealed notable relationships between BW and BL (*r* = 0.294) and RW (*r* = 0.296), with RW and WH showing the strongest correlation (*r* = 0.429). PCA reduced dimensionality, with PC1 (35.7% variance) heavily loading on RW (0.776), WH (0.665), and RH (0.622), highlighting their significance in structural size. Regression models showed that PCA-derived components (PC1 and PC2) were more effective than individual traits, accounting for 52% of BW variability compared to just 9% for BL alone. These results emphasize the value of multivariate analysis in breeding programs, suggesting that composite indices (e.g., PC1 as a "size" factor) improve prediction accuracy over single-trait methods. Focusing on RW, WH, and RH in selection could boost productivity, while PC-based models provide practical BW estimation tools for resource-constrained farms. This study highlights the significance of integrated morphological analysis for the sustainable management of Garut sheep.