TY - JOUR
T1 - Distance in Spatial Analysis
T2 - Measurement, Bias, and Alternatives
AU - Mu, Wangshu
AU - Tong, Daoqin
N1 - Funding Information:
We sincerely thank the editor Prof. Racheal Franklin, the editor of the special issue Prof. Alan Murray and Prof. Keith Clarke, and the anonymous reviewers for their insightful comments and suggestions that significantly strengthened this manuscript during the reviewing process. The research is supported by the National Science Foundation under Grant No. 1461390. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.
Publisher Copyright:
© 2020 The Ohio State University
PY - 2020/10/1
Y1 - 2020/10/1
N2 - Distance is an important and basic concept in geography. Many theories, methods, and applications involve distance explicitly or implicitly. While measuring the distance between two locations is a straightforward task, many geographical processes involve areal units, where the distance measurement can be complicated. This research investigates distance measurement between a location (point) and an area (polygon). We find that traditional polygon-to-point distance measurements, which involve abstracting a polygon into a central or representative point, could be problematic and may lead to biased estimates in regression analysis. To solve this issue, we propose a new polygon-to-point distance metric along with two algorithms to compute the new distance metric. Simulation analysis shows the effectiveness of the new distance metric in providing unbiased estimates in linear regression.
AB - Distance is an important and basic concept in geography. Many theories, methods, and applications involve distance explicitly or implicitly. While measuring the distance between two locations is a straightforward task, many geographical processes involve areal units, where the distance measurement can be complicated. This research investigates distance measurement between a location (point) and an area (polygon). We find that traditional polygon-to-point distance measurements, which involve abstracting a polygon into a central or representative point, could be problematic and may lead to biased estimates in regression analysis. To solve this issue, we propose a new polygon-to-point distance metric along with two algorithms to compute the new distance metric. Simulation analysis shows the effectiveness of the new distance metric in providing unbiased estimates in linear regression.
UR - http://www.scopus.com/inward/record.url?scp=85089183506&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85089183506&partnerID=8YFLogxK
U2 - 10.1111/gean.12254
DO - 10.1111/gean.12254
M3 - Article
AN - SCOPUS:85089183506
SN - 0016-7363
VL - 52
SP - 511
EP - 536
JO - Geographical Analysis
JF - Geographical Analysis
IS - 4
ER -