The Nature of Geographic Data. The Paper Map. A long and rich history Has a scale or representative fraction The ratio of distance on the map to distance on the ground Is a major source of data for GIS Obtained by digitizing or scanning the map and registering it to the Earth’s surface
Discrete: definitive; with concrete, observable, boundaries
Continuous: no easily discernable boundaries, “fuzziness” depends on scale
Raster representation. Each color represents a different value of a nominal-scale field denoting land cover class.
First law of geography: “everything is related to everything else, but near things are more related than distant things” – Waldo Tobler
Many new geographers would say “I don’t understand spatial autocorrelation” Actually, they don’t understand the mechanics, they do understand the concept.
Unobserved similarities between places?
Diffusion? (disease spread)
Common activity in neighboring areas? (crime)
Common policy across neighboring areas? (zoning)
Every location is equally likely to be chosen
Sample points are spaced at regular intervals
Requires knowledge about distinct, spatially defined sub-populations (spatial subsets such as ecological zones)
More sample points are chosen in areas where higher variability is expected
As always, error propagates and grows through subsequent analyses
Correlation does not mean causation
Sampling method may introduce bias
Models and measurements must be appropriate for your dataset
With GIS data, model must be geo-aware
r is the correlation value between two or more sets of values
Ranging from -1 to +1, r identifies the degree of positive or negative correlation
Squaring r produces a percentage to which two sets of data share the same values
r can be plotted as a “best-fit” or trend line
Gravity model applies concepts in physics to the social sciences
The “masses” and distance between two urban places influences the migratory bond between two places
Population (people, employment) and distance decay effect the degree to which two places are “bonded”
(work with me here, people)
. . . well, a snowflake.
. . . but that the area it bounds is finite (indeed, it is
contained in the white square).