首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Estimating Small-Area Populations by Age and Sex Using Spatial Interpolation and Statistical Inference Methods
Authors:Qiang Cai  Gerard Rushton  Budhendra Bhaduri  Edward Bright  Phillip Coleman
Institution:Department of Geography The University of Iowa; Department of Geography The University of Iowa; Geographic Information Science and Technology Group Oak Ridge National Laboratory; Geographic Information Science and Technology Group Oak Ridge National Laboratory; Geographic Information Science and Technology Group Oak Ridge National Laboratory
Abstract:The objective of this research is to compute population estimates by age and sex for small areas whose boundaries are different from those for which the population counts were made. In our approach, population surfaces and age‐sex proportion surfaces are separately estimated. Age‐sex population estimates for small areas and their confidence intervals are then computed using a binomial model with the two surfaces as inputs. The approach was implemented for Iowa using a 90 m resolution population grid (LandScan USA) and U.S. Census 2000 population. Three spatial interpolation methods, the areal weighting (AW) method, the ordinary kriging (OK) method, and a modification of the pycnophylactic method, were used on Census Tract populations to estimate the age‐sex proportion surfaces. To verify the model, age‐sex population estimates were computed for paired Block Groups that straddled Census Tracts and therefore were spatially misaligned with them. The pycnophylactic method and the OK method were more accurate than the AW method. The approach is general and can be used to estimate subgroup‐count types of variables from information in existing administrative areas for custom‐defined areas used as the spatial basis of support in other applications.
Keywords:
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号