04 data viz concepts
Download as PPTX, PDF1 like690 views
This document provides an overview of concepts and techniques in data visualization. It discusses the work of Gregory Bateson in information and mind, Jacques Bertin's seven visual variables to represent data, and color use guidelines from Brewer. It also includes examples of TGV transportation networks displayed using visual variables, a dashboard example, theories on pre-attentive visual variables, and techniques for displaying quantity through location from historical figures like Playfair, Minard and Neurath.
04 data viz concepts
- 1. Data Visualization Concepts Prepared by: Paul Kahn – Experience Design Director February, 2013 Media Lab, Aalto University Helsinki, Finland
- 2. Gregory Bateson (1904-1980) British anthropologist, social scientist, linguist, visual anthropologist, semiotician and cyberneticist whose work intersected that of many other fields Major books: Steps To An Ecology of the Mind, 1972 Mind and Nature: A Necessary Unity, 1979
- 3. Information and Mind All information is communicated as differences The mind operates with hierarchies and networks to create gestalten. Hierarchies are nested containers Networks are links connecting discrete nodes Information architecture is the re/shaping of information/differences into hierarchies and networks we search for and visualize the patterns that connect The pattern that connects is the pathways for accessing differences
- 4. Jacques Bertin (1918-2010) Visual Variables for Quantitative Information “Matrix theory of graphics,” Information Design Journal, Vol. 10, No. 1. (2002) Semiology of graphics: Diagrams, Networks, Maps (Univ of Wisconsin, 1983; ESRI, 2010) originally published as Sémiologie graphique (1967)
- 5. Seven Visual Variables To Represent Data 5
- 6. 6 Variables of the Image (1-3) • X/Y Position • Size: Z value of quantity (area) superimposed on position • Value: Z value of content (fill) superimposed on position
- 7. 7 Variables of the Image (Beniot Martin)
- 8. 8 Differential Variables (4-5 ) • Grain/Pattern: Variation of value within glyph • Color: hue of glyph content
- 9. 9 Differential Variables (6-7 ) • Orientation: relative position in relation to XY grid • Shape: abstract shapes distinguished by outline: dots, squares, triangles, diamonds, metaphors
- 10. 10 Les variables visuelles (Beniot Martin)
- 11. 11 TGV Network Network map 2011
- 12. 12 TGV Network • X/Y Position • Size: Z value of quantity (area) • Value: Z value of content (fill) • Grain/Pattern • Color • Orientation • Shape
- 13. 13 TGV Network • X/Y Position • Size: Z value of quantity (area) • Value: Z value of content (fill) • Grain/Pattern • Color • Orientation • Shape
- 14. 14 TGV Network TGV Change of service speed to Marseille BEFORE
- 15. 15 TGV Network TGV Change of service speed to Marseille AFTER
- 16. 16 Color Use Guidelines for Data Representation Brewer, C. A. 1999. Color Use Guidelines for Data Representation, Proceedings of the Section on Statistical Graphics, American Statistical Association
- 17. 17 Online resources Brewer, C. A. 1999. Color Use Guidelines for Data Representation, Proceedings of the Section on Statistical Graphics, American Statistical Association http://www.personal.psu.edu/cab38/ColorSch/ASApaper .html No more excuses: a list of references to learn how to use color http://diuf.unifr.ch/people/bertinie/visuale/2009/0 5/infovis_color_theory_in_few_li.html
- 20. 20 CogSci Theory (Dan Berlin) Pre-attentive Visual Variables (1-4) From Designing Interfaces by Jenifer Tidwell
- 21. 21 Pre-attentive Visual Variables (5-8)
- 22. 22 Don’t make me think Immediate Visual Scan Repeated Visual Scan An interaction is intuitive when the user makes the least effort to grasp the difference.
- 23. 23 Steps of Visual Cognition Preattentive Perception Cognition Processing Perception • All based on changes in contrast: hue, brightness, and color palette • We detect differences, physiologically and psychologically Pre-attentive Processing • Processed in under 250 milliseconds (Healey, Booth, and Enns, 1995) • Parallel (bottom-up) processing Cognition • Serial (top-down) processing
- 24. 24 Elementary Perceptual Tasks We are good at some tasks, but not others • Good at: position, length, direction • Bad at: area (of a circle), volume, saturation This is why you will see line or bar graphs to convey data • You will never (well, shouldn’t) see a graph that uses color saturation to convey data (i.e. using different shades of orange)
- 25. 25 Preattentive Processing Second step of visual perception “The perception of a pattern can often • Sits between perception and cognition be the basis of a new insight.” • Processed in under 250 milliseconds - Colin Ware, Information Visualization • Understanding without training or cognition • Serial vs. parallel processing • Forms objects in the mind’s eye Preattentive variables • Proximity, similarity, connectedness, continuity, symmetry, closure, relative size, figure and ground, intensity, curvature, line length, color, orientation, brightness, and direction of movement. • Overlapping variables • Many theories as to how we deal with these – Feature Integration Theory, for one (2 variables at most) Variable hierarchy
- 26. Example: Periodic Table of Elements Dmitri Mendeleev’s original table (1869)
- 29. Periodic Table as a metaphor
- 30. Displaying Quantity in Location William Playfair (1759-1823): space as a metaphor for quantity
- 31. 31 Charles Joseph Minard (1781-1870) Thickness of line (also known as a Sankey Diagram)
- 32. Otto Neurath (1882-1945), Gerd Arntz (1900-1988) — Isotype: Repeated unit as an expression for quantity
- 33. Otto Neurath, Modern Man in the Making (1939)
- 35. Maps & Diagrams | September 2011 | 35
- 36. US Population density (2000), Read Agnew & Don Moyers, UNDERSTANDING USA