空间统计分析方法的高校生源分区研究

Zoning for college undergraduate enrollments based on spatial statistical methods

针对高校生源基地建设和招生宣传策略制订的问题,该文以淮海工学院近3年的省内生源数据为基础,运用核密度估计和局部空间自相关分析方法对其生源的空间分布特征进行可视化表达和热点识别,并据此开展生源分区。结果表明,连云港、徐州、宿迁及长江沿岸县市是淮海工学院生源的高密度分布区,淮安南部、扬州和泰州北部以及盐城市全境则为低密度分布区;生源密度的空间分布具有显著的空间自相关性。根据生源密度热点和冷点分布的统计显著性(p<0.05)可将全省划分为3类生源区,且各类生源区均有着各自的生源数量和空间分布特征。基于生源核密度的局部空间自相关分析,可获得连续性较好且具有显著统计学意义的生源热点和冷点分布区,其结果可为生源分区提供坚实的数理统计基础。

Aiming at the problom to zoning for college undergraduate enrollments and the construction of college enrollment base and the formulation of enrollment propaganda strategy,based on the last threeyear provincial enrollment data of Huaihai Institute of Technology(HHIT),the methods of kernel density estimation and local spatial autocorrelation analysis were used to characterize the spatial distribution of enrollments and recognize the hotspots of enrollments,and then HHIT enrollment zoning was carried out.The results showed that the high density areas of HHIT enrollments were mainly distributed in the cities of Lianyungang,Xuzhou,Suqian and counties along the Yangtze river,while the low density areas covered the southern part of Huai’an city,northern part of Yangzhou city and Taizhou city and the whole area of Yancheng city.The local spatial autocorrelation analysis of enrollment density showed that the spatial distribution of enrollment had significant spatial autocorrelation.According to the statistical significance of the distribution of hot and cold spots of enrollment density(p<0.05),the whole province could be divided into three categories of enrollment areas,and each area had its own spatial distribution characteristics of enrollments.The study showed that local spatial autocorrelation analysis of the kernel density of enrollments could provide continuously and statistically significant hot and cold spots,which would settle a solid mathematical statistical basis for enrollment zoning.