A Brief History of Data

Visualization
 Mary Eleanor Spear
1952, 1969

• Common-sense
advice
• Invented box plot
• Worked for various
US government
agencies
Jacques Bertin 1967
• Principle of
expressiveness:
• Say everything you want
to say — no more, no less
• Don’t mislead
• Principle of effectiveness:
• Use the best method
available for showing
your data
• Cartographer
 Jacques Bertin
Seven Visual Variables
• Position
• Size
• Shape
• Color
• Brightness
• Orientation
• Texture
 Edward Tufte 1983

• Disciplined design
principles
• Minimalist approach
• Professor emeritus at
Yale University
 Jock Mackinlay
1986

• Automatically encode data with

software
• Enable people to focus on ideas,
concepts
• Added eighth variable to Bertin’s
list: motion
• VP of Research and Design,
Tableau Software
When a Chart hits our Eyes
1. Visuals aren’t read in a predictable,
linear way
• Create charts spatially, from the visual
outward
2. We see first what stands out
• Whatever stands out should support idea
3. We see only a few visuals at once
• Plot as few visual elements as possible
4. We seek meaning and make connection
• Relate visual elements in a
meaningful way
5. We rely on conventions and
metaphors
• Embrace deeply ingraine49d
conventions
What is data visualization
and why is it important?
Is data visualization a part of
data science?
 What are the best data
visualization software of 2019?
What is data discovery and
visualization?
What are data visualization
tools?
Is Excel a data visualization tool?
How do you create good data
visualization?
What kind of visual communication
do you want to create?
1. Is my information
conceptual or data- driven?
• Conceptual information is
qualitative
• Data-driven information is
quantitative
2. Are my visuals meant to be
declarative or exploratory?
• A declarative purpose is to make a
statement
• An exploratory purpose is to look
for new ideas
Four Types of Data Visualizations
Declarative
Idea
illustration

Conceptual Data-Driven

Idea
generation

Exploratory
Four Types of Data Visualizations
Declarative

Idea Everyday
illustration dataviz

Conceptual Data-Driven

Idea Visual
generation discovery

Exploratory
 Data Visualization

provide clear understanding of patterns in data

detect hidden structures in data

condense information
What makes a good chart?
What can you learn from this map?

http://www.popvssoda.com/countystats/total-county.html
 Some basic principles (adapted
from Tufte 2009)

1 • The chart should tell a story

2 • The chart should have graphical

integrity

3 • The chart should minimize graphical

complexity

Tufte’s fundamental principle: Above all else show the data

Principle 1: The chart should tell a
story
Graphics should be clear on their own

The depictions should enable meaningful comparison

The chart should yield insight beyond the text

“If the statistics are boring, then you’ve got the

wrong numbers.” (Tufte 2009)
 Principle 2: The chart should have
graphical integrity
• Basically, it shouldn’t “lie” (mislead the reader)
• Tufte’s “Lie Factor”:
• LF = 𝑠𝑖𝑧𝑒 𝑜𝑓 𝑒𝑓𝑓𝑒𝑐𝑡𝑠ℎ𝑜𝑤𝑛 𝑖𝑛 𝑔𝑟𝑎𝑝ℎ𝑖𝑐
𝑠𝑖𝑧𝑒 𝑜𝑓 𝑒𝑓𝑓𝑒𝑐𝑡𝑖𝑛 𝑑𝑎𝑡𝑎

Should be ~ 1

> 1 = exaggerated < 1 = understated

effect effect
 Principle 3: The chart should
minimize graphical complexity
Generally, the simpler the better…

Key concepts
Sometimes Chart junk
a table is Less Data-ink should not
better be there
 When a table is better than a
chart
For a few data points, a table can do just as well…
Salespers Total
on Sales
Total Sales by Salesperson
Peacock $225,763.68
$
2 Leverling $201,196.27
5
0 Davolio $182,500.09
,
0 Fuller $162,503.78
0
0 Callahan $123,032.67
.
0 King $116,962.99
0
Dodsworth $75,048.04
$
2 Suyama $72,527.63
0
0
Buchanan $68,792.25
,
0
0 The table carries more information in less space
0
. and is more precise.
0
 The Ultimate Table: The Box
Score
• Large amount of
information in a
very small
space
• So why does
this work?
• Depends on the
reader’s
knowledge of the
data
 Data Ink
• The amount of “ink” devoted to data in a chart
• Tufte’s Data-Ink ratio:
𝑑𝑎𝑡𝑎−𝑖𝑛𝑘
Data
𝑖𝑛𝑘 Ink Ratio
𝑢𝑠𝑒𝑑 𝑖𝑛 𝑔𝑟𝑎𝑝ℎ𝑖𝑐
𝑎𝑙 𝑡𝑜

Should be
~1
< 1 = more non-data = 1 implies all ink
related ink in devoted to data
graphic
Tufte’s principle: Erase ink whenever possible
Being conscious of
dataLower
inkdata-ink Hypothetical City Crime
ratio 425

(worse) 375

Thefts per 100000 citizens

325

275

225

175
Hypothetical City Crime 125
425 75
375 25
Thefts per 100000 citizens

325 2003 2004 2005 2006 2007

2008 2009 2010
275

225

175 Hypothetical City

125 Crime 37
40
0 37
75 0 35 0
32 33 0
25 0 0
2003 2004 2005 2006 27
2007 2008 2009 2010 0
20
0

Higher data-ink
ratio 2003 2004
2009
2005
2010
2006 2007 2008

(better)
What makes a good
chart?
Sum of
Extended Price

2011 Total Sales

1
6
0
0
0
Sometimes
0
1
4
it’s really a
matter of
0
0
0
0
1
2
0
preference.
These
0 Order Date
0
0

both
1 Sum of
0 Extended Price
0
0
2011 Total Sales
0
1
0
6
8
0
minimize
data
Whyink.isn’t a
0
0
0
0

table
1
6
4
0
0
0
0
4
1
better
here?
0
2
0
0 Order Date
0
 3-D
Charts

Evaluate this from a data-ink

perspective. How does it affect
the clarity of the chart?
One of the golden rules
of data visualization
Never useis…..
3D! Data • 3D skews
numbers,
Integrity/
making them
Lie Factor
difficult to
• Adding
interpret3D or to
graphs
compare
Graphic introduces
al unnecessary
Comple chart elements
xity like side and
Source: Knaflic (2015). Storytelling withfloor panels
Data: A Data Visualization Guide for
Chartjunk: Data Ink “gone
wild”
Unnecessary visual clutter
that doesn’t provide
additional insight

Distraction from the story

the chart is supposed to
convey

When the data-ink ratio is low,

chartjunk is likely to be high
 Example: Moiré effects
(Tufte 2009) Creates
Total Sales by Salesperson
$ illusion of
2
5 movement
0
,
0
Stands out, in
0
0
a bad way
.
0
0
Hypothetical City
$
2 4 Crime
0 2
0 5

Thefts per 100000 citizens

, 3
0 7
0 5
0
. 3
0 2
0 5

2
$ 7
1 5
5
2 200 200 200 200 200 200 201
0
2 4 5 6 7 8 9 0
,
5
 Example: The
Grid Hypothetical City Crime
425

375
Why are
Thefts per 100000 citizens

325

275 these
225

175 examples
125

75
of
25
2003 2004 2005 2006
chartjunk?
Hypothetical City
2007 2008 2009 2010
425 Crime
375

Thefts per 100000 citizens

325

275
What could 225

you do to 175

125

remedy it? 75

25
20 200 200 200 200 200 200 201
03 4 5 6 7 8 9 0
Data Ink Working
For Us
Evaluate
this chart
in terms of
Data Ink.

Imagine
this as a
bar chart.
As a
table!!
Review: Data principles
(adapted from Tufte 2009)
1 • The chart should
tell a story

2 • The chart should have

graphical integrity

3 • The chart should minimize

graphical complexity

Tufte’s fundamental
principle: Above all
else show the data
Infographics
• Information graphics
• Visualization of
information, data or
knowledge intended to
present information
quickly and clearly
• We will have an ICA
to create
infographics using
Piktochart.
 Summa
• Use ry
data visualization principles to assess a
visualization
• Tell a story
• Graphical integrity (lie factor)
• Minimize graphical complexity (data ink, chartjunk)
• Explain how a visualization can be improved based
on those principles
• Types of visualization