Open Geospatial Data: An Assessment of Global Boundary Datasets. Brigham, C.; Gilbert, S.; and Xu, Q. In Proceedings of the 20th Annual GIS Research UK (GISRUK 2012), of GISRUK Conference Series, pages 35+.
Open Geospatial Data: An Assessment of Global Boundary Datasets [link]Paper  abstract   bibtex   
[Excerpt: Conclusion] Through comparison of GAUL, GADM and UNSALB boundary datasets we found that each dataset has advantages and drawbacks in terms of accuracy and usability, but overall GAUL was the best dataset due to the accuracy and completeness of the dataset. While UNSALB boundaries have the highest rate of accuracy because of validation with national mapping agencies, it is limited in geographic scope. Although GADM has a global scale, many of the boundaries are outdated and it is unclear whether GADM organizers have utilized public feedback to update boundary sets GADM. For local level mapping exercises boundary selection should be selected on a case by case basis, but given GAUL's superior accuracy and completeness, in particular in the developing countries of Africa and Latin America, makes it best suited for mapping subnational data on a global scale. [...]
@inproceedings{brighamOpenGeospatialData2012,
  title = {Open Geospatial Data: An Assessment of Global Boundary Datasets},
  booktitle = {Proceedings of the 20th Annual {{GIS Research UK}} ({{GISRUK}} 2012)},
  author = {Brigham, Charles and Gilbert, Steven and Xu, Qiyang},
  date = {2012},
  pages = {35+},
  url = {http://mfkp.org/INRMM/article/14478120},
  abstract = {[Excerpt: Conclusion]

Through comparison of GAUL, GADM and UNSALB boundary datasets we found that each dataset has advantages and drawbacks in terms of accuracy and usability, but overall GAUL was the best dataset due to the accuracy and completeness of the dataset. While UNSALB boundaries have the highest rate of accuracy because of validation with national mapping agencies, it is limited in geographic scope. Although GADM has a global scale, many of the boundaries are outdated and it is unclear whether GADM organizers have utilized public feedback to update boundary sets GADM. For local level mapping exercises boundary selection should be selected on a case by case basis, but given GAUL's superior accuracy and completeness, in particular in the developing countries of Africa and Latin America, makes it best suited for mapping subnational data on a global scale. [...]},
  keywords = {*imported-from-citeulike-INRMM,~INRMM-MiD:c-14478120,administrative-units,comparison,data-uncertainty,global-scale,open-data,uncertainty},
  series = {{{GISRUK}} Conference Series}
}
Downloads: 0