Skip to main content

Organizing Data

Introduction:

  • Real world is messy
  • To get data into a GIS, we need to measure "The real world"
  • Measurement is not perfect
    • Makes assumptions
    • Represents a particular view of the world
  • GIS data becomes the medium by which users see the "REAL WORLD"

 

 

Levels of measurement:

  • Categorical
    • Numbers are used as "names" for mutually exclusive sets of objects, events, processes, persons
    • Different values of the variable do NOT indicate different magnitudes of any property
    • Equality and inequality
    • no < or >, no + or -
  • Ordinal
    • Values of the variable only indicate the rank of the measured objects
    • Higher values => more of the property, but not how much more
    • = and !=, < and >
    • no + or -
  • Interval
    • Equal differences in the values of the variable indicate equal difference in the property
    • Ratios of variable values do NOT indicate equal ratios in the property
    • = and !=, < and >, + and -
    • No X or /

 

  • Ratio
    • Equal intervals in variable values indicate equal intervals in the property, AND
    • Equal ratios in variable values indicate equal ratios in the property
    • = and !=, < and >, + and -, X and /

 

Accuracy: correctness of measure: difference between prediced, measured, or observed value and true value; freedom from error.

 

Precision: Closeness of measurements to each other

 

Reliability: reproducibility of results

 

Validity: Extent to which measurements reflect what we intend them to. Measurements can be reliable without being valid; but not valid without being reliable.

 

 

Scale:

1:1 is largest

1:24 is smaller