社会调查的空间抽样方法研究

The Spatial Sampling Method of Social Survey

社会调查的传统抽样方法是在抽样框内进行,无法查看抽样结果的空间分布并评价抽样结果的合理性。针对此问题,本文提出了一种社会调查的空间抽样方法,首先将抽样数据空间可视化,结合抽样框确定抽样的限定条件,然后利用GIS技术将抽样结果空间可视化,最后利用最邻近指数法评判抽样结果的空间分布均衡性,从而优化样本选取方案。本文在武汉市中心城区中选择洪山区和江岸区为研究区域,使用武汉市2011年的人口普查数据进行实验,得到抽样框要求下的最佳抽样方案,将抽样结果最邻近指数的置信度设为95%,不断增加样区所需抽取样本数量进行实验,找出每个研究区域下最合适的样本数量,从而对抽样框的设计提出了改进建议。

The sampling of traditional social survey uses traditional sampling methods in the sampling frame without effective use of geospatial information and GIS spatial analysis technology.The spatial distribution of the sampling results and the evaluation of the sampling results＇ rationality cannot be seen immediately.In consideration of this,this paper proposes a spatial sampling method of social survey.First,it visualizes sampling data space,and determine the qualification of sampling according to the sampling frame.Then it uses GIS technology to visualize sampling space.Finally,it uses the nearest neighbor index to judge the spatial equilibrium of the sampling results＇ distribution,so as to optimize the sampling points selection scheme of the sampling of social survey.This paper chooses Hongshan district and Jiang＇an district as the research areas,uses the 2011 population data in Wuhan city for experiments and gets the best sampling result under the sampling frame.With the sampling results＇ most neighboring index confidence level set at95%,the number of sampling points needed to be chosen on research areas increases to find the most appropriate sampling size under each research area.Meanwhile,suggestions for improvement in designing the sampling frame are offered.