Research Journal of Agricultural Sciences Vol : 9 - Issue : 2 ; 01-05 Sheela Kharkwal and *Hamid Nazir Division of Dairy Economics, Statistics and Management, ICAR-National Dairy Research Institute (Deemed University), Karnal - 132 001, Haryana, India *Department of Animal Husbandry Extension, College of Veterinary and Animal Sciences, GBPUA&T, Pantnagar - 263 145, Uttarakhand, India Abstract The present study was carried out in Kumaon hills of Uttarakhand state to identify typical dairy production systems and its relationship with farm and socio-economic characteristics of farm households. The study area comprised of five randomly selected villages from three Tehsils in each of two hill districts of Kumaon division, viz. Nainital and Almora. A total of 300 households constituted the ultimate sampling units for the study. Farm household typologies were constructed by using two multivariate statistical techniques, viz. Principal Component Analysis (PCA) and Cluster Analysis (CA). PCA was used to transform linearly an original set of 20 variables, representing farm and socioeconomic characteristics, into a smaller set of uncorrelated variables (factors) that represents most of the information in the original set. The factors retained from the PCA were used for cluster analysis. Four homogenous groups (clusters) were obtained. Cluster I (28.42%) was defined as households with low land holding, high stock of buffalo and low degree of technology adoption. Cluster II (40.75%) was defined as households with high land holding, high degree of technology adoption, and high buffalo stock with low farm family labour involvement. Cluster III (22.60%) was defined as households with low intensity of Market Participation in Dairying and Cluster IV (8.22%) was defined as households with high intensity of market participation, high stock of crossbred animals and high degree of technology adoption.