Full Citation
Title: Spatio-temporal Small Area Analysis for Improved Population Estimation Based on Advanced Dasymetric Refinement
Citation Type: Miscellaneous
Publication Year: 2016
ISBN:
ISSN:
DOI:
NSFID:
PMCID:
PMID:
Abstract: Demographic datasets are aggregated over areas to protect privacy. To study micro-scale demographic processes, those datasets have to be collected over temporally consistent small areas. However, the availability of such data is limited. That is, the demographic data is either aggregated over large geographical areas (i.e., counties), or collected over small population-derived census units that are temporally inconsistent (i.e., census tracts). Areal interpolation methods transfer the variable of interest from source zones to target zones. The methods can be used in temporal demographic applications to create temporally consistent population estimates over small areas by transferring population values from the areal units of one census year (i.e., source zones) to the units of another census year (i.e., target zones). In this research, spatial refinement is incorporated into areal interpolation methods to enhance their population interpolation accuracy. Moreover, one method called Enhanced Expectation Maximization (EEM) is introduced. Areal interpolation methods -- with and without spatial refinement are used to estimate total population values from census tracts in 1990 to census tract boundaries in 2010 in Mecklenburg County, North Carolina. Based on validation results, EEM is the most accurate method to create temporally consistent population estimates for the 1990-2010 period in the study area.
Url: http://www.cartogis.org/docs/proceedings/2016/Zoraghein_Leyk_Buttenfield_and_Ruther.pdf
User Submitted?: No
Authors: Zoraghein, Hamidreza; Leyk, Stefan; Buttenfield, Barbara; Ruther, Matt
Publisher: Cartography and Geographic Information Society
Data Collections: IPUMS NHGIS
Topics: Methodology and Data Collection, Other
Countries: