Breed characterization in the livestock is important for the breeding program in the future. This research aimed to characterize Thin-tail sheep in the highland and lowland areas of Jambi Province based on 7 body measurements and body indices using multivariate analysis. Data were collected from 160 sheep consisted of 80 rams and 80 ewes from each area (320 sheep in total with 1-2 years old). The research showed that the principal component analysis (PCA) in this study was explained 65.84%-72.30% by body measurements and 78.23%-84.99% by body indices of the total phenotypic variance of animals. The body measurement of cannon circumference (CC) and body indices of area index (AI), relative cannon index (RCI), dactyl thorax index (DTI), conformation index (CI), and index of body weight (IBW) were selected as the discriminating variable for Thin-tail sheep in different areas. However, this study’s canonical correlation (rc) values were 0.44 for body measurements and 0.47 for body indices. Therefore, about 67.5% of Thin-tail sheep from the lowland area and 57.5% of Thin-tail sheep from the highland area can be characterized with body measurements. Hence, about 61.2% of Thin-tail sheep from the lowland area and 65.6% of Thin-tail sheep from the highland area can be characterized with body indices. Temperature, humidity, rainfall, and length of radiation in the highland areas are lower than in the lowland areas. The cluster analysis in four Thin-tail sheep populations revealed two clusters, i.e., cluster 1 consisted of Kerinci and Sungai Penuh and cluster 2 consisted of Muaro Jambi and Batanghari. It was concluded that about 60% of Thin-tail sheep could be characterized through their body indices.