Transcript Document

Attribute data
• In a vector setting
– Stored in a relational database for each
geographic object
• A relational database is organized in a series of
two-dimensional tables, each of which contains
records for one data object.
• A ‘row’ of data for each point, line, or polygon
Vector setting: attribute data:
Column = property
Row = object
Raster setting: attribute data
• In a raster setting
– Stored as a grid of cells (each cell is given
one number)
rows
Raster attributes: cell values
1
2
3
5
8
4
6
8
3
9
3
5
3
3
1
7
5
4
3
9
2
2
4
5
2
columns
Cell (x,y)
Cell value
Attribute data
Three types of data:
• Nominal scale
• Ordinal level
• Interval and ratio levels
Attribute data types
• Nominal data – descriptors about the
objects – numbers do not quantify an
amount
Examples:
• 1-school, 2-community center, 3-bench mark
• 1-Road, 2-boundary, 3-stream
• 1-Swamp, 2-desert, 3-forest
• 1-Apple tree, 2-peach tree
Ratio data
Ratio data: ratios make sense (3:1) for these data
(which means that ‘0’ means ‘nothing’/‘none’/‘zilch’)
• mm of precipitation
• Resident population
• Elevation (contour lines: 30m, 40m, 50m,
etc.)
• Average household income (of census
tract)
• Age
Interval scale: adding numbers are
meaningful, but ratios aren’t
• Fahrenheit
• Celsius
• Note: 10 degrees cooler means the same
thing at each temperature. But 20
degrees doesn’t mean twice as much as
10 degrees.
• Also, 0 degrees doesn’t mean ‘no
degrees’.
Ordinal data
Numbers identify an order only (no scale)
• Road size:, 1-forest path, 2-dirt road, 3gravel road, 4-country road, 5-major
highway
• 1-never, 2-sometimes, 3-often, 4-always
• 1-sober, 2-tipsy, 3-drunk,
4-smashed/pissed (British), 5-plastered
• 1-small, 2-medium, 3-large