Land use modelling has used rural areas as object for a long time. Modellers start in general with physical landscape factors to determine the spatial distribution of land use, especially agricultural crops. An important factor defining rural areas is the dominance of livelihoods/economies based on agriculture combined with a strong relationship to tradition, high value of family ties, scales of social aggregation, and a framework given by nature. The inter-linkages among economic, social and environmental components determine both the complexity and dynamics of rural development. Both modellers and social scientists seek a better understanding of these complex rural systems in terms of their development paths, the underlying forces and the interactions between physical land use and rural society. In this paper, we use a combination of a land use change model and information about social relationship by referring both to the common spatial unit of villages for a rural region in south west China. We want to show how such a common unit can help to include social science information into a physically oriented land use model. The inclusion of social data results in a substantial improvement of the modelling results. In the model run we could show that land use patterns in the study area were closely related to socio-economic characteristics and governmental rules. In China, collective land and state land are distinguished. This land designation turned out to be the most important factor for the land use types rainfed, irrigated land, and rubber. All land use types were influenced by the ethnicity of village inhabitants because they had different land use traditions.