Which spatial statistics techniques should be converted to GIS functions?

  • Daniel A. Griffith

Abstract

A fundamental difference between spatial and aspatial data is that observations for geo-referenced data are correlated strictly due to their relative locational positions. This self-correlation activates complications in the statistical analysis of geo-referenced data that lie dormant in the statistical analysis of traditional data comprised of independent observations. Seeking a remedy for this problem has helped motivate the development of an array of procedures labelled spatial statistics. This contribution seeks to identify existing spatial statistical tools that would be routinely of value to the GIS user community — a major analyzer of geo-referenced data — and that currently are theoretically and conceptually ready to be made available through GIS packages (e.g., ARC/INFO, IDRISI). Technical and computational implementation issues are addressed. Current expert opinion, gleaned from two recent international conferences, is summarized. Gateways are identified for spatial statistical tools to be introduced into a GIS. The principal conclusion is that there is evidence from a growing number of initiatives, as well as testimony from spatial scientists and geographic analysis practitioners, supporting the need to convert spatial statistics techniques into GIS functions.

Keywords

Autocorrelation Kriging Glean Gridding 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Anselin L (1989) Spatial regression analysis on the PC: spatial econometrics using GAUSS, unpublished manuscript, Geography Department, UC Santa BarbaraGoogle Scholar
  2. Anselin L, Getis A (1991) Spatial statistical analysis and geographic information systems, paper presented at the 31st European Congress of the Regional Science Association, Lisbon, Portugal, August 27–30Google Scholar
  3. Anselin L, Griffith D (1988) Do Spatial Effects Really Matter in Regression Analysis? Papers Reg Sci Assoc 65: 11–34CrossRefGoogle Scholar
  4. Anselin L, Hudak S (1991) Spatial econometrics in practice: a review of software options, paper presented at the 38th North American meetings of RSAI, New Orleans, November 7–10Google Scholar
  5. Bivand R (1989) User interfaces for geographical software tools: a case study in regression modelling, paper presented at the Sixth European Colloquium on Theoretical and Quantitative Geography, Chantilly, France, September 6–9Google Scholar
  6. Bivand R (1990) Spatial statistics: front-end inference support for GIS, paper presented at the Third Scandinavian Research Conference on Geographical Information Systems, Helsingor, Denmark, November 14–16Google Scholar
  7. Chorley R [chairman] (1987) Handling Geographic Information. Her Majesty’s Stationary Office, Report to the Select Committee on GIS, LondonGoogle Scholar
  8. Griffith D (1988a) Advanced Spatial Statistics. Kluwer, DordrechtCrossRefGoogle Scholar
  9. Griffith D (1988b) Estimating spatial autoregressive model parameters with commercial statistical packages. Geogr Anal 20: 176–186CrossRefGoogle Scholar
  10. Graffith D (1989) Spatial regression analysis on the PC: spatial statistics using MINITAB, Discussion Paper No. 1. Institute of Mathematical Geography, Ann Arbor, MIGoogle Scholar
  11. Griffith D (1990a) A numerical simplification for estimating parameters of spatial autoregressive models. In: Griffith D (ed) Spatial Statistics: Past, Present, and Future. Institute of Mathematical Geography, Ann Arbor, MI, pp 185–195Google Scholar
  12. Griffith D (1990b) Simplifying the normalizing factor in spatial autoregression, invited paper presented at the International Geographical Union Commission on Mathematical Models Workshop on Spatial Modelling, Boston University, November 8, 1990Google Scholar
  13. Griffith D (1990c) Supercomputing and spatial statistics: a reconnaissance. Profess Geogr 42: 481–492CrossRefGoogle Scholar
  14. Griffith D (1991) Spatial regression analysis on the PC: Spatial statistics using SAS, unpublished manuscript, Department of Geography, Syracuse UniversityGoogle Scholar
  15. Griffith D (1992) Simplifying the normalizing factor in spatial autoregressions for irregular lattices, Papers Reg Sci 71: 71–86CrossRefGoogle Scholar
  16. Haining R (1990) The use of added variable plots in regression modelling with spatial data. Profess Geogr 42: 336–344CrossRefGoogle Scholar
  17. IBM (1991) Exploring new worlds with GIS, Directions 5 (#2, Summer/Fall), 12–19Google Scholar
  18. Kehris E (nd) Spatial autocorrelation statistics in ARC/INFO, Research Report No. 16, ESRC North West Regional Research Laboratory, Lancaster UniversityGoogle Scholar
  19. Legendre P (1991) Spatial autocorrelation: trouble or new paradigm? Paper presented at the Statistics Canada Methodology Symposium 91, Spatial Issues in Statistics, Ottawa, Canada, November 12–14Google Scholar
  20. National Research Council (Mapping Science Committee; Commission on Physical Sciences, Mathematics, and Resources) (1990a) Spatial Data Needs: The Future of the National Mapping Program. National Academy Press, Washington, DCGoogle Scholar
  21. National Research Council (Board on Mathematical Sciences) (1990b) Renewing US Mathematics: A Plan for the 1990,s. National Academy Press, Washington, DCCrossRefGoogle Scholar
  22. National Science Foundation (1987) Solicitation: National Center for Geographic Information and Analysis. Directorate for Biological, Behavioral, and Social Sciences. Washington, DCGoogle Scholar
  23. Overton S, Stehman S (1990) Statistical properties of designs for sampling continuous functions in two dimensions using a triangular grid, Technical Report 143. Department of Statistics, Oregon State UniversityGoogle Scholar
  24. Upton G, Fingleton B (1985) Spatial Data Analysis by Example, Vol 1. Wiley, New YorkGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  • Daniel A. Griffith
    • 1
  1. 1.Department of GeographySyracuse UniversitySyracuseUSA

Personalised recommendations