ASA WORKSHOP ON DATA VISUALIZATION TECHNIQUE S

S T O R Y T E L L I N G W I T H D ATA

VEENA MENDIRATTA
january 31, 2014 veena.mendiratta@gmail.com
Outline

§  Model and Characterization

§  Simple Examples
§  Elements of a Data Story
§  Summary
 Story Definition and Model

•  Story
–  ordered sequence of steps (<- order is key)
each step contains text, images, visualizations, video, etc
–  defined path (or paths) through the steps
•  Journalism Model
–  journalists collect information through research, interviews,
etc to assemble the key facts
–  tie together the key facts (raw material) to produce a story
•  Data Analyst Model
–  use visualization for exploration and analysis
–  use visualization for presentation (storytelling) using the
results from the analysis
–  tools used for analysis may not work for presentation

Robert Kosara, Jock Mackinlay, Storytelling: The Next Step for Visualization,
IEEE Computer (Special Issue on Cutting-Edge Research in Visualization), vol. 46, no. 5, pp. 44–50, 2013.
 Characterizing the Area
−Narrative Visualization
Edward Segel and Jeffrey Heer, Narrative Visualization: Telling Stories with Data, IEEE TRANSACTIONS ON
VISUALIZATION AND COMPUTER GRAPHICS, VOL. 16, NO. 6, NOVEMBER/DECEMBER 2010

•  Narrative framework Genres

magazine style
–  visual structure (genre): support the story annotated chart
partitioned poster
–  interactivity: engage the story flow chart
comic strip
–  messaging: tell the story slide show
film, video, animation
•  Approaches
–  author driven: strong ordering, heavy messaging, limited interactivity
–  reader driven: weak ordering, light messaging, free interactivity

•  Common schemas
–  martini glass structure: prioritizes author-driven approach

–  drill-down story: prioritizes reader-driven approach

–  interactive slide-show: promotes dialog between above approaches

 Narrative Framework

Edward Segel and Jeffrey Heer, Narrative Visualization: Telling Stories with Data, IEEE TRANSACTIONS ON
VISUALIZATION AND COMPUTER GRAPHICS, VOL. 16, NO. 6, NOVEMBER/DECEMBER 2010
https://multimedia.journalism.berkeley.edu/media/upload/presentations/2012/01/13/InteractiveJournalism_Berkeley.pdf
Design Space
… genres + interactivity + messaging
Author Driven Reader Driven
̶  strong order ̶  weak order
 ̶ heavy messaging  ̶ light messaging
 ̶ minimal interactivity  ̶ free interactivity
è  clear story è  query
è  fast delivery è  explore

è  author’s message è  reader driven

martini glass interactive

slide-show drill-down
story
§  Model and Characterization
§  Simple Examples
§  Elements of a Data Story
§  Summary
Simple story example

http://www.storytellingwithdata.com/2012/10/my-penchant-for-horizontal-bar-graphs.html
… this tells a better story
Chart Junk …
… when is it helpful for a story?

better recall in the long-term – elements, trend, message

Scott Bateman, Regan L. Mandryk, Carl Gutwin, Aaron Genest, David McDine, Christopher Brooks,
Useful Junk? The Effects of Visual Embellishment on Comprehension and Memorability of Charts. ACM
Conference on Human Factors in Computing Systems (CHI), 2010.
Pictograph (1930)

“Home and Factory Weaving in England, 1820-1880″ Otto and Marie Neurath Isotype Collection, University of Reading
as seen in The Functional Art by Alberto Cairo.
§  Model and Characterization
§  Simple Examples
§  Elements of a Data Story
§  Summary
Storytellying elements

story depth

simple
charts

tells a story
Storytellying elements

story depth visuali-

zation
info scent

simple info
charts graphics

tells a story
narrative
Storytellying elements

story depth visuali- visual

zation data
info scent

stories
audience

simple info
charts graphics

tells a story
narrative
author
Building Narrative
Napoleon’s March 1812-1813 flowmap

“ … a means of telling the story of the soldiers’ plight …”

 Driving Safety in Fits and Starts
Building Narrative
http://www.nytimes.com/interactive/2012/09/17/science/driving-safety-in-fits-and-starts.html

Connected Scatterplot
 Why is Her Paycheck Smaller
Adding Depth
http://www.nytimes.com/interactive/2009/03/01/business/20090301_WageGap.html?_r=0
 Housing’s Rise and Fall in 20 Cities
Adding Depth
http://www.nytimes.com/interactive/2011/05/31/business/economy/case-shiller-index.html?ref=economy
 Jobless Rate for People Like You
Adding Depth
http://www.nytimes.com/interactive/2009/11/06/business/economy/unemployment-lines.html
 How Different Groups Spend Their Day

Adding Interaction
http://www.nytimes.com/interactive/2009/07/31/business/20080801-metrics-graphic.html
§  Model and Characterization
§  Simple Examples
§  Elements of a Data Story
§  Summary
Discovery vs Presentation
Discovery Presentation
explore how much narrative to provide?
build views … depends
test hypothesis is there a presenter?
lots of data or is the graphic provided?

Presentation
̶  è focus!
 ̶ common mistake – lack of focus, too much data
 ̶ present only what is relevant for the story
 ̶ minimum amount of information needed to make
the story work
Summary
•  Understand the context – audience, data, takeaway/outcome STRUCTURE
•  Choose right display type – text, scatterplots, line charts, bar charts, ...
•  Eliminate clutter – use Gestalt principles to cut if no information value
•  Draw attention where you want to – preattentive attributes of color, size, …
•  Linear is better for storytelling
•  Guide readers through the story –where to start, how to get back, reset
•  Limit complexity initially, reveal as needed
•  Cool and readability maybe at odds – recognize tradeoffs, tailor to audience
•  Text good for storytelling MESSAGING
•  State the point you want to make – don’t leave the reader wondering
•  Start with an interesting view
•  Put numbers and facts in context
•  Connect relevant text and graphics, e.g., see Figure 1
•  Add summary/conclusions/”so what?”
•  Labels and significant digits suggest what deserves attention
•  Show how the interactivity works, make it intuitive INTERACTIVITY
•  Limit interactivity to key elements – too much can distract from story