Multivariate analysis of morphological variables has been successfully used to estimate genetic variation within and between local breeds. The objective of this study was to differentiate Hararghe highland goat populations based on their morphometric traits by applying multivariate analysis. Sixteen morphometric traits were collected from 450 goats reared in the three agroecological zones (highland, midland and lowland) of West Hararghe. Multivariate canonical discriminant analysis in combination with cluster and discriminant analysis was applied to identify the combination of variables that differentiate goats of the three agroecological zones. The results indicated that all the morphometric traits were significantly affected by age. The cluster analysis indicated that two main groups of midland goats were included in one group, while group two included highland and lowland goats under one sub-cluster. The canonical discriminant analysis identified two canonical variables (CAN) of which CAN1 and CAN2 accounted for 68.2 and 31.8% of the total variation, respectively. The quadratic discriminant analysis correctly assigned the respective 71.3, 77.3, and 81.3% of lowland, midland, and highland goat populations into their source populations, with an overall accuracy rate of 76.7%. The Mahalanobis distance verified that lowland and highland goats are the closest, while midland and highland goats were the furthest. However, the canonical discriminant analysis indicated a visible overlapping between goat populations of the three agroecological zones, indicating the existence of homogeneity among them. In conclusion, multivariate analysis identified 11 morphometric traits as the most imperative traits to differentiate Hararghe highland goat populations effectively. Genetic potentials of Hararghe highland goat populations can be improved through community-based breeding programs for their sustainable utilization and conservation.