Attribute Data
Attribute data describes the characteristics of spatial features:
- Each row (record) represents a spatial object and each column (field) describes a characteristic of that object
- Two types of attribute data: textual or numerical
Attribute data is managed by relational database. It is a collection of tables (relations) that are defined for each class of objects describing the same subject
- Allows to organize, update, and interrogate data of different subjects
- Each table can be managed separately from other tables, but can be connected when needed (i.e. for queries)
There are two types of attribute tables in ArcGIS:
- Feature attribute tables: has access to the spatial data and uses a FID to link the spatial data with the attribute data
- Non-spatial attribute table: stores additional information and has one field in common with the feature attribute table
The connection between tables is made though a key (i.e. an ID), which is a common field whose values can uniquely identify a record in a table. They cannot be linked through the FID.
Types of relationship between tabular data
- One-to-One relationship
- One-to-Many relationship
- Many-to-One relationship
- Many-to-Many relationship
There are 4 measurement levels associated with attribute data:
- Ratio - has an absolute 0
- Interval - distance between numbers is meaningful
- Ordinal - can be ordered
- Nominal - only named variables
| Level | Summary | Possible Operations | Statistics | Examples |
|---|---|---|---|---|
| Nominal | Categories only. Data cannot be arranged into an ordering scheme | =, | Counting : mode, amplitude | Name, State, gender, color |
| Ordinal | Categories are ordered, but differences can’t be found or are meaningless | Median, quartile | Type of car (compact, mid- size, full-size) | |
| Interval | Differences are meaningful, but there is no natural zero starting point and ratios are meaningless | Mean, variance, correlation, etc. | Temperature | |
| Ratio | There is a natural zerostarting point and ratios are meaningful | Mean, variance, correlation, etc. | Computing distance |
Attribute analysis is done by spatial queries which search for and select a subset of features and table records. All query expressions in ArcGIS use SQL to formulate these search specifications. The purpose of this is to:
- Explore the data
- Focus on a data subset of interest
Basic structure in SQL (i.e. syntax) is: SELECT WHERE
- Logical operators: =, >,
- Arithmetic operators: +, -, *, /, ^
- Boolean operators: AND, OR, NOT, XOR
- Mathematical functions (trigonometric functions (e.g. sin, tan), logarithms, etc.)
EXAMPLE: Select* (=all columns) From ContaminatedSites Where “Toxicity” > 8