Lecture 3

The Basics, Part II

0d - point phenomena located in 2d or 3d space (eg data collected at weather monitoring stations)

1d - linear phenomena (eg the path an AUV takes while taking measurements)

2d - areal phenomena (eg data collected on the surface of a lake)

2.5d - volumetric phenomena - each x, y position has a single z value associated with it (eg the maximum depth at any point in the lake)

3d - volumetric phenomena - each x, y, z position has a value associated with it (eg the ph values collected at various points and depths in the lake)

•     nominal - grouping (categorization) but no ordering of those categories (eg census data on religion)
  

• ordinal - categorization plus ordering of those categories (eg low, medium, high)
  
  
• numerical
•     interval - ordering of the data plus explicit numerical differentiation between those categories (eg temperature values) with an arbitrary zero point (eg SAT scores, F, C - eg getting 2x an SAT score doesnt mean you are twice as smart, the temperature isnt twice as hot from 10 degrees C to 20 degrees C)
•     ratio - ordering of the data plus explicit numerical differentiation between those categories (eg temperature values) with an nonarbitrary zero point (eg temperatures in degree kelvin)
  

• orientation - direction/orientation of the marks/symbol
  
• shape - different shapes are used
  
• arrangement - different arrangement of marks making up the symbol
  
• hue - different colors are used but careful choices need to be made here so there is no sense of 'less' to 'more' in the different hues

• spacing (texture) - smaller spacing between marks suggest higher value
• size - larger symbol or larger marks making up the symbol suggests a higher value
  
  
• perspective height - higher elevation suggests a higher value (cant be used for 3D phenomena because all 3 dimensions are already in use)
  
  
• color (hue) - what is the dominant wavelength (red, green, blue, etc but careful choices need to be made here so there is a sense of 'less' to 'more' in the progression on hues) 
  
• color (lightness) - how light or dark the color is (light green, dark green)
• color (saturation) - how far is the intensity is from grey (bright red, muted red)

• commonly used to portray data collected for units such as counties or states
• regions are shaded / colored based on the phenomena - in the example below lightness is used
  
• good for when values change abruptly at unit boundaries but hides variation within units, and the boundaries may be artificial in relation to the phenomena.
• the units may also be different sizes, so the raw data may need to be standardized to better show the underlying phenomena - eg California has much more land mass and more people than Illinois which has more land mass and people than Rhode Island. The visualization above showing the raw number of brew pubs may give people the wrong impression; it might be better to (also) view the number of brew pubs per million people in each state.
  

• good when data collected was from a smooth continuous phenomenon
• regions are shaded / colored based on the phenomena - in the example below lightness is used
• interpolating set of isolines between sample points of known values
• the data should be standardized
• we will be talking about algorithms for this in 2D and 3D next week

•     scale symbols in proportion to the magnitude of the data
•     symbol might be a true point (located at a data collection point) or a conceptual point (at the center of a unit)
•     normally used to show raw data

• one dot is set equal to a certain amount of the phenomenon
• dots should be placed where the phenomena occurs (much higher level of accuracy than other maps)

Coming Next Time

VTK