GIS-Based Statistical Analysis of Population Spatial Distribution Patterns-Taking Gansu Province as an Example

The existence of spatial dependence breaks the basic assumption that the samples are independent of each other in most of the classical methods of statistical analysis. Supported by GIS, we used the method of spatial autocorrelation analysis to make a preliminary analysis of the population spatial differences and distribution characteristics in various districts and counties in Gansu Province in 2006. The results showed that the population distribution in Gansu have positive spatial correlationship, and presented a significant spatial aggregating feature. But from local spatial autocorrelation analysis, we can see that the relevance of regional distribution of population in majority of Gansu is not significant, indicating that these regions within a certain range had not form regional growth poles with strong attraction. This conclusion can serve better for the formulation of population policies and sustainable development strategy of the economy and society.