Chapter 14. Advanced Spatial Analysis

1. What is data mining?
2. What is the centroid or mean center?
3. What does looking at the change in centroid in time help you understand?
4. What is the MAT and what can it be used for?
5. Why is the MAT a normative spatial analysis method?
6. Explain one measure of dispersion of spatial data.
7. How can histograms be used to summarize spatial datasets?
8. You want to make a scatterplot showing the relationship between two interval variables in the raster model.  How does it work?
9. Making a scatterplot of two interpolated datasets is one way of combining data from different sets of spatial objects.  Methods of spatial interpolation make it easy to invent geographic data, a potentially dangerous practice.  Why might the scatterplot be suspect?
10. The Moran index and a semivariogram are spatial summarization methods.  What does each summarize?
11. What is fragmentation?
12. What are some measures of fragmentation?
13. Ian McHarg was a pioneer.  Why?
14. What is the difference between the MAT and an analogous problem on a linear network?
15. What are location-allocation problems?
16. What are routing problems?
17. What is the TSP?
18. Software packages such as ArcLogistics use procedures called heuristics. What are they?
19. What is optimum path routing through continuous space in a raster GIS system?
20. What are friction values?
21. Optimum routing requires the use of move sets including the rook's case and the queen's case.  Explain the difference between these two different move sets.
22. What is a cost layer?
23. What are inferential statistics?
24. Are least-cost path problems in a raster resolution dependent?
25. Why are geographic datasets not usually independent samples that are appropriate for inferential statistics?
26. We cannot have it both ways- if we believe in spatial interpolation, we cannot at the same time believe in independence of geographic samples, despite the fact that this is a basic assumption of statistical tests.  Explain.
27. What are three ways of dealing with spatial dependence and spatial heterogeneity in spatial data sets?