SCOPeS Visual Properties

An introduction of visual properties in module 2 of IT 7113 Data Visualization (http://idi.kennesaw.edu/it7113/) at Kennesaw State University.
About Jack Zheng

Faculty of IT at Kennesaw.edu
visual variable color visual property data visualization scopes Color Visual Data http Visual property http www
An introduction of visual properties in module 2 of IT 7113 Data Visualization (http://idi.kennesaw.edu/it7113/) at Kennesaw State University.
About Jack Zheng

Faculty of IT at Kennesaw.edu
visual variable color visual property data visualization scopes Color Visual Data http Visual property http www
Visual Encoding with 
SCOPeS Visual Variables/Properties
Jack Zheng
Fall 2023
IT 7113 Data Visualization
http://idi.kennesaw.edu/it7113/
©
https://www.edocr.com/v/631d1wpb/jgzheng/SCOPeS-Visual-Properties
https://www.edocr.com/user/jgzheng/collection/datavisualizationlecturenotes
Visual Properties: SCOPeS
 Visual variable or property is the “decoration” applied to visual 
elements to represent data values
 A visual property is used to encode different values of a particular 
dimension of data
 There are six basic visual properties - easier to be remembered as 
“SCOPeS” (Dr. Jack’s term)
Size
Color
Orientation
Position
Texture
Shape
Bertin’s Original Version
https://infovis-wiki.net/wiki/Visual_Variables
Semiology of Graphics: Diagrams, Networks, Maps 
1st Edition, by Jacques Bertin, 
https://www.amazon.com/dp/1589482611
Data Characteristics
 Data encoding or mapping is determined both by the kind of visual 
property and the type of data. The encoding process is the fit of the 
two.
 The value of a data item is mapped to the value of a particular 
visual property based on the types of data
 Some properties can be more effectively represent values of certain 
data types than others
Continuous 
quantitative data
Numerical values. Example: sales amount, age, height, 
etc. usually they can be aggregated (sum, average, etc.)
Ordinal data
Discrete data, but with an order; often qualitative (for 
example month in a calendar year) but can be 
quantitative (ranked or ranged, like age groups/intervals).
Nominal data
The data is a collection of non-numerical and non-
ordered data (discrete): categorical. Example: 
departments in the college
Expanded readings on data characteristics https://dwbi.org/pages/18
Visual Property: Size
 Size is a physical measures of the visual element like length, width, 
height, area, angle, quantity of items, etc. It is commonly used for 
continuous data values.
 Examples
 Scaling issue
 For various reasons, it is common that the size property does not directly 
and truly represent the underlying value. In these cases, it must be very 
careful to design the size property, because unreasonable distortions will 
impact human perception.
Person symbol height represents 
number of years
Area size represents 
percentages
Visual Property: Color
 Color is the most common visual property that can be used for both 
dimensions (categorical) and measures (continuous or discrete)
 Color has three major properties (HSV)
 hue (color spectrum)
 saturation
 brightness (value)
 Extended reading:
 http://learn.leighcotnoir.com/artspeak/elements-color/hue-value-
saturation/
 Color and color schemes: https://blogs.ifgi.de/digital-
cartography/symbols/color-and-color-schemes/
 https://en.wikipedia.org/wiki/HSL_and_HSV
Color coding based on data characteristics

Two basic ways of using colors: categorical and continuous
 Categorical data (dimensions or measures) – using hues

Represent and differentiate individual data item -
http://en.wikipedia.org/wiki/Pie_chart (colors in pie chart 
commonly represent countries, which is nominal or 
categorical) – however, color is ineffective if there are more 
than 7 items or categories.
 Group a set of similar items or data points (dimensions or 
measures)
 Highlight (dimensions or measures): alert, distinguish, etc.
 Continuous data (measures) – using values or saturations

Sequential system: using one color’s value range
 Diverging system: two color gradient (or step) system, often 
used in heat maps where two colors represent opposite 
directions. (Example: 
https://www.theinformationlab.co.uk/2014/12/03/geospatial
-heat-maps-tableau-via-alteryx/)
 Gradient or stepped: 
 gradient is a smooth transition, often used for continuous
measure
 Stepped usually categorize measures in ranges and uses 
more discrete color values, https://covid.cdc.gov/covid-
data-tracker/#global-counts-rates; 
https://gasprices.aaa.com/
 More references

https://spectrum.adobe.com/page/color-for-data-
visualization/
Color Systems, Cases and Best Practices
 Traffic light system in performance management
 Uses green, yellow, and red to represent different levels of performances (good, 
warning, bad); three colors corresponds to target values; it is a hue-based system 
used ranged measures.
 https://citoolkit.com/articles/traffic-light-assessment/
 Temperature system in maps (heatmap) – a bipolar hot/cold system: red for hotter 
temperature and blue for colder
 Heat maps (https://en.wikipedia.org/wiki/Heat_map) use color to represent target 
values, either continuous or in groups/ranges (ordinal)
 Originally used for data related to temperature and weather 
https://weatherboy.com/north-america-chills/
 Also for data that has a temperature metaphor, such as click map 
(ttps://www.ometrics.com/heat-maps/), population density 
(https://andyarthur.org/thematic-map-county-population-density.html), lighting, 
salary, etc.
 More best practices and examples of using colors in data visualization:
 https://cambridge-intelligence.com/choosing-colors-for-your-data-visualization/
 https://www.dataquest.io/blog/what-to-consider-when-choosing-colors-for-data-
visualization/
 https://colorbrewer2.org (a good color choice tool)
Other considerations when selecting color
 Color symbolism
 Match the color of visual element to the 
real-world counterpart (things or 
phenomenon) it represents.
 CI color: use the color consistent with 
the company image to represent the 
company or organization.
 Reasonably consider aesthetics that 
contributes to human’s emotional 
perception
 temperature - warm vs. cool
 harmony - sharp vs. soft
 Use beautiful colors: 
https://blog.datawrapper.de/beautifulc
olors/
 Color blindness
 Prepare for a second alternative, usually 
using textual instead of color.
http://www.dataatworkbook.com/data-work-
03-beyond-visual-perception/
Mismatch of natural color of the 
vegetable with the line color that 
represents the vegetable.
Color represents gas prices. Is yellow 
cheaper or more expensive? 
https://www.gasbuddy.com/GasPriceMap
Which color system is better in the following 
charts? Any other better designs?
0
1
2
3
4
Emily
Rachel
Joy
Roger
GPA
0
1
2
3
4
Emily
Rachel
Joy
Roger
GPA
0
1
2
3
4
Emily
Rachel
Joy
Roger
GPA
Hue based color coding for each person – not 
necessary, as each person is represented by 
position. Color may be used to represent a 
second dimension, like school or class.
Value based color coding for each 
person – not recommended same 
reason as the one on the left.
Unified color 
(color has no 
specific meaning)
0
1
2
3
4
Emily
Rachel
Joy
Roger
GPA
Can we use value-based color 
coding to represent GPA values? 
I.e. darker color for higher GPA? 
Generally not needed, as GPA is 
represented by size of the bars.
Visual Property: Orientation
 Orientation is about direction. It can be seen as variations of a 
particular shape or pattern pointing to different directions. It can 
also be seen as a variation of position (pointing to different positions)
 A common example is the use of arrows or hands pointing to 
different directions.
 Examples
 Gauge chart
http://voyager8.blogspot.com/2014/01/the-historical-
relationship-between.html
Visual Property: Position
 Position refers to the location where a data item is 
placed. Data values can be visualized as absolute 
positions in the visualization, or as the relative distance 
between elements. 
 Position is commonly used in
 Dot plot – a variation of column chart
 Strip plot (one-dimension scatter plot)
 The placement of data items against a pre-established 
scheme (such as a Cartesian coordinate system)
 Spatial distances (especially used with maps, geo maps, 
or any spatial locations like building, stadium, campus, 
etc.)
 Examples
 http://www.gartner.com/technology/research/methodo
logies/research_mq.jsp (top left chart)
 http://en.wikipedia.org/wiki/Cluster_analysis (bottom left 
chart) 
Visual Property: Texture
 Texture is used much like colors but seldom used for continuous 
data.

It is important when color sensitivity is an issue. Implementations 
include fill patterns, border patterns, shadow, etc.
 Examples
Visual Property: Shape

Type of shapes
 Shapes can be formed using simple shapes: square, triangle, etc.
 More complex shapes also can be formed by combination of simple shapes: icon, marker, etc.
 Shapes are usually used to represent different type of things, or nominal/discrete data 
(e.g., category). Shapes are rarely used for continuous data.
 Shapes are usually used for dimensions but sometimes used for measures as well.
 Examples
 http://www.masters.com/en_US/scores/ - in this example, shapes are used for measures (ordinal) 
actually
 http://v8doc.sas.com/sashtml/gref/z15-ex.htm
- shapes are used for species
Motion - on Top of SCOPeS
 Motion is the movement or transformation of the basic visual properties, which can represent 
richer meanings and grab greater attention.
 Dynamic change of positions, shape, color, orientation, texture, size
 Note: simply animating an object in the visualization can just be decorative, but not data driven
 Examples:

Simple position change: move one position to the other
 Growing or shrinking size, and/or accompanies by shape shifting
 Moving direction (of position change): used like orientation 

Flickering/blinking/flashing/spinning pattern: can be used like color or texture for categorical data

Speed (of movement and changing) can be added as additional properties in motion driven 
visualizations

Typical usages and examples

Show trends along a period of time -
https://www.linkedin.com/feed/update/activity:6464536253118373888/
 Operational process simulation: e.g., disk optimization
 http://metrocosm.com/global-immigration-map/
 http://www.storybench.org/role-motion-visualizations/
 https://www.gapminder.org/tools/#$chart-type=bubbles&url=v1
 https://www.digitalattackmap.com
We do not cover motion in depth in IT 7113, but it can be a very good research topic for class project.
Composition of Multiple Properties
 Combinations of these properties can be used to represent multi-
dimensional data in the same visualization.
 Example: here is one product visualized as an object with small (size) 
square (shape), red color (color), and double-line border (texture).
 Size: representing sales amount
 Shape: representing product type
 Color: representing profit/loss
 Line textual: representing sales channel
 More complex visual elements (such as icons and symbols) can be 
built based on the basic elements and properties discussed above.
 See more examples of the visual property use at 
http://innovis.cpsc.ucalgary.ca/innovis/uploads/Courses/Informatio
nVisualizationDetails/09Bertin.pdf
Visual Properties Choice
 Choose appropriate visual properties to code data based on data 
types
Refer to https://infovis-
wiki.net/wiki/Visual_Variables
The shaded box indicates 
most inappropriate.
Best choice on top 
and then less 
appropriate toward 
the bottom. For 
example, position 
property can be used 
to code all types of 
data really good; while 
area is somewhat good 
for quantitative data, 
but it is not good for 
ordinal and nominal 
data.
For example, do not use shape for ordinal 
type of data, like month, or class level, etc.
Types of Data
Also see https://www.qlik.com/blog/visual-encoding
• Choose one visual property to represent one dimension. E.g., use colors for 
products.
• Use different visual properties, instead of different values of the same visual 
property, for different dimensions (or dimension attributes). E.g. if colors are used for 
products, then do not use color for countries again. Choose position or shape for 
countries.
Visual Properties used for 
Dimensions and Measures
*
Occasionally, only in specific cases
**
Sometimes, depending on more factors
***
Very often and appropriate
Measure 
(continuous)
Measure 
(ordinal)
Dimension
(categorical: ordinal or 
nominal)
Shape
X
*
***
Color (Hue)
*
*
***
Color (Value)
**
**
*
Orientation
**
**
**
Position
***
**
**
Texture
X
*
**
Size
***
**
*
Exercise: which visual coding method is better?
VS.
VS.
 Representing 
product 
categories
 Representing 
population
Readings and Resources
 Visual variable/property basics
 https://infovis-wiki.net/wiki/Visual_Variables
 http://innovis.cpsc.ucalgary.ca/innovis/uploads/Courses/InformationVisualiza
tionDetails/09Bertin.pdf
 Choosing visual encoding
 https://www.qlik.com/blog/visual-encoding
 Choose Appropriate Visual Encodings 
https://www.oreilly.com/library/view/designing-data-
visualizations/9781449314774/ch04.html
 For maps https://www.e-education.psu.edu/geog486/node/594
 More articles and books
 http://wiki.gis.com/wiki/index.php/Visual_variable
 Considering Visual Variables as a Basis for Information Visualisation, 
https://prism.ucalgary.ca/handle/1880/45758
 Roth, R. E. 2017. Visual Variables. The International Encyclopedia of 
Geography. 
https://onlinelibrary.wiley.com/doi/abs/10.1002/9781118786352.wbieg0761
 Semiology of Graphics: Diagrams, Networks, Maps 1st Edition, by Jacques 
Bertin, https://www.amazon.com/dp/1589482611