How The Public Misinterprets Data Visualizations - Alberto Cairo

ESS DATA VISUALISATION MILAN 2019
How Charts Lie
how the
Getting public
Smarter about
misinterprets data visualizations

Visual Information
and whatAlberto
we canCairo
do about it
New book to be published by W.W. Norton
(October 2019)
To pre-order:
W.W. Norton: https://books.wwnorton.com/books/

detail.aspx?ID=4294998829
IndieBound: https://www.indiebound.org/book/
9781324001560
Barnes&Noble: https://www.barnesandnoble.com/w/how-
charts-lie-alberto-cairo/1130420282?ean=9781324001560
Amazon: https://www.amazon.com/How-Charts-Lie-Getting-
Information/dp/1324001569/ref=sr_1_1?keywords=how
+charts+lie&qid=1551105523&s=gateway&sr=8-1
Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

October 2019
What can I say? I'm a sucker for statistics

explained in funny, engaging, and
How Charts Lie Alberto Cairo has written a wise, witty
and utterly beautiful book. You couldn't
mathematically correct ways, especially hope for a better teacher to improve
when every now and then a line like your graphical literacy.
"charts lie to us because we are prone to
lying to ourselves" is thrown in with good Tim Harford
humor. A must read for anyone who author of The Undercover Economist
works with charts or who wants to stay
informed with an educated, skeptical eye.
Cathy O’Neil
This book will open your eyes to how
author of Weapons of Math Destruction
everyone uses visuals to push agendas. A
master visual designer, Cairo shows you
Getting Smarter about how to read charts and decode design.
After this book, you can’t look at charts
Visual Information with a straight face!
Kaiser Fung
Alberto Cairo author of Numbersense

Visualization has the potential to become a universal language

Visualization has the potential to become a universal language
SOME TOOLS THAT I COVER IN MY
“INTRODUCTION TO INFOGRAPHICS AND DATA VISUALIZATION” COURSES
DATA CLEANING GENERATING STYLING AND OUTPUT

AND EXPLORATION BASE GRAPHICS COMPOSITION
Excel Charts and graphs

Data Wrangler (or Trifacta) iNZight, RAWGraphs, Illustrator
iNZight (for visual exploration) For print
Illustrator: Save as PDF
Adobe
Illustrator
Data maps
QGIS, Tilegrams
(NOTE: I use the R programming language
more every day for these tasks, For the Web
but if you’re a beginner it’s better to have
fun with the tools above first)
Pictorial lllustrations and icons

Illustrator’s pen tool and 3D features For the Web
http://www.thefunctionalart.com/p/instructors-guide.html
To fulfill visualization’s potential to become a universal language…
Maybe we need to spend a bit less time DESIGNING visualizations

and use more time to teach non-experts how to REASON about them
Although we can all do both simultaneously

Common misconceptions
1. “A picture is worth a thousand words”

2. “Visualization is intuitive”
3. “The data should speak for itself ”
4. “Show, don’t tell!”

Common misconceptions
1. “A picture is worth a thousand words”

2. “Visualization is intuitive”
3. “The data should speak for itself ”
4. “Show, don’t tell!”

Visuals are often ambiguous

Visuals are often ambiguous

Challenges to making visualization a universal language
Challenge 1
IGRAPHICACY (“graphical illiteracy)

We believe (wrongly) that charts are illustrations that can be understood at a quick glance.
But charts are visual arguments. They mustn’t bee just seen; they need to be read.

Skills for a modern
educated citizen:
—Literacy
—Articulacy
—Numeracy
—Graphicacy

The symbolical and grammatical level



The semantics level
Liberal pundit: “Obamacare is good for the job market”
The Affordable Care Act

Thousands of (”Obamacare”) was passed
people in the on March 2010
workforce

The semantics level

Thousands of (”Obamacare”) was passed
people in the on March 2010
workforce
Stimulus package passed on February 2009

The semantics level
Skeptical me: But what if Obamacare hadn’t passed?
Thousands of
(”Obamacare”) fails to
?
people in the pass on March 2010
workforce

The semantics level
Skeptical me: But what if Obamacare hadn’t passed?
Thousands of
(”Obamacare”) fails to
?
people in the pass on March 2010
workforce
?

Challenge 2
CONFIRMATION
We constantly see patterns. Once we identify patterns that look meaningful —even if they
are just the product of randomness— our brains try to connect them with causal links and
shape them as stories. Then, we strive to confirm them, never to challenge them

https://twitter.com/TreyYingst?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5Eauthor
Stephen J. Adler, Steve Holland and Jeff Mason
https://www.reuters.com/article/us-usa-trump-100days/exclusive-trump-says-he-thought-being-president-would-be-easier-than-his-old-life-idUSKBN17U0CA

Surface on the
county-level map:
Red: 80%
Blue: 20%
VICTORY BY A LANDSLIDE!!!!!!

Surface on the
county-level map:
Red: 80%
Blue: 20%
VICTORY BY A LANDSLIDE???
Map by Kenneth Field

https://twitter.com/kennethfield/
status/970827334038237184
Surface on the
county-level map:
Red: 80%
Blue: 20%
SHARE OF THE POPULAR VOTE IN THE 2016 PRESIDENTIAL ELECTION

Donald Trump 46.1% 62,984,825 votes
Hillary Clinton 48.2% 65,853,516 votes
Other candidates 5.7%

Surface on the
county-level map:
Red: 80%
Blue: 20%
SHARE OF THE POPULAR VOTE IN THE 2016 PRESIDENTIAL ELECTION

Donald Trump 46.1% 62,984,825 votes
Hillary Clinton 48.2% 65,853,516 votes
Other candidates 5.7%
PERCENTAGE OF ELIGIBLE VOTERS

Didn’t vote 40.0%
Voted for Donald Trump 27.7%
Voted for Hillary Clinton 28.9%
Voted for other candidates 3.4%

These are the numbers that truly matter in a U.S. Presidential Election
ELECTORAL TRUMP CLINTON

VOTES 304 Other: 7 227
270
WHO WON ON EACH STATE STATE SIZE ADJUSTED BY ELECTORAL VOTES
IT CONTRIBUTES TO THE ELECTION

Challenge 3
DISTORTIONS
Sometimes charts are simply badly design —in purpose or unintentionally

1880: 56.8°F 2012: 58.2°F
+1.4°F (+0.8°C)

Chart based on a design by data analyst Sean McElwee
1880: 56.8°F 2012: 58.2°F
+1.4°F (+0.8°C)

Chart based on a design by Richard Reiss
(City University of New York’s Institute for Sustainable Cities)
1880: 56.8°F 2012: 58.2°F
+1.4°F (+0.8°C)

1880: 56.8°F 2012: 58.2°F
+1.4°F (+0.8°C)

Challenge 4
FEAR OF COMPLEXITY
We dislike ambiguity, nuance, and complexity, and prefer simplistic stories with a closure
to proper arguments, which often can’t be clear, linear, or have a definite ending.

The danger of aggregating data too much,
and presenting just averages and other simplistic summaries
Most places are pretty safe, and

have likely remained down here
(these aren’t real data points)

The danger of aggregating data too much,
and presenting just averages and other simplistic summaries
Some places are so

far up that they skew
the national rate
Most places are pretty safe, and

have likely remained down here
(these aren’t real data points)
Careful with amalgamation

paradoxes and outliers
http://journal.frontiersin.org/article/10.3389/fpsyg.2013.00513/full

Challenge 5
UNCERTAINTY
We believe that numbers are accurate, precise, and objective. We must abandon this view.
Data is always biased, incomplete, and uncertain —and that’s fine!
On the other hand, journalists and designers need to show confidence and uncertainty in
their charts when uncertainty is critical.

Articles about uncertainty
https://ec.europa.eu/eurostat/cros/powerfromstatistics/OR/PfS-OutlookReport-Cairo.pdf https://graphics.cs.wisc.edu/Papers/2014/CG14/Preprint.pdf
Collection of papers about visualizing uncertainty:

https://www.dropbox.com/sh/4lu8sqgem5pr68x/AAD_YEfaP7zp15n77VBFA2eGa?dl=0

Actual quote from an
NHC news report:
“Alberto is not very well

organized this morning.”

A photo that terrifies me…

WARNING
This is NOT
how to read
this kind of map

1 2 3 4 5
WARNING
All these
lines are fictional

WARNING
This is closer
to how to read
the map correctly
but still not
entirely right

95%
Of the time, the storm
WARNING
This is NOT
will be inside the cone how to read
this kind of map
5%
may end up going
outside of the cone

67%
AT LAST...
This is
will be inside the cone how to read
this kind of map
33%
may end up going
outside of the cone

WARNING
This is an
imaginary cone

WARNING
This is an
imaginary cone

Challenge 6
DESCRIPTIONS
The way we describe the content of a chart to
ourselves may bias our understanding of that chart

85 Incorrect verbal
descriptions of a chart
80 can greatly bias our
understanding of it
75
70
Smoking
Life expectancy
(in years)
65
cigarettes
can help you
60
Each red dot is a country

live longer!!!
55
Example provided by
50 Heather Krause
0 500 1,000 1,500 2,000 2,500 3,000 https://idatassist.com
Annual cigarette consumption
per person above age 14

85 Incorrect verbal
descriptions of a chart
80 can greatly bias our
understanding of it
75
70
Smoking
cigarettes
Life expectancy
(in years)
65
can help you
60
High income countries

live longer??
Middle income countries
55
Low income countries
Example provided by
50 Heather Krause
0 500 1,000 1,500 2,000 2,500 3,000 https://idatassist.com

HIGH INCOME MIDDLE INCOME LOW INCOME
COUNTRIES COUNTRIES COUNTRIES
85 85
Life expectancy 80 80
Life expectancy
70 70
60 60
50 50
0 1,000 2,000 3,000 0 1,000 2,000 3,000 0 1,000 2,000 3,000

Be careful with amalgamation paradoxes and with the ecological fallacy


100%
People who never
smoke or who stopped
75% smoking years ago
Survival rate
50%
Cigarette
smokers
25%
0%
40 50 60 70 80 90 100
Age

WHAT TO DO?

Let’s spread the word about data reasoning

Let’s test our graphics more often
with non-experts

Let’s not fear redundancy. Show AND tell
Hans Rosling: https://www.gapminder.org/

Scott Klein: https://www.propublica.org/nerds/infographics-in-the-time-of-cholera

We can add “how to read

this chart” explainers to
our visualizations

Let’s spread the word about data reasoning
Let’s test our graphics more often (with non-experts)
How
Let’s not ChartsShow
fear redundancy. LieAND tell
THANK YOU! Getting Smarter about

Visual Information
Alberto Cairo
New book to be published by W.W. Norton
(October 2019)
To pre-order:
W.W. Norton: https://books.wwnorton.com/books/

detail.aspx?ID=4294998829
IndieBound: https://www.indiebound.org/book/
9781324001560
Barnes&Noble: https://www.barnesandnoble.com/w/how-
charts-lie-alberto-cairo/1130420282?ean=9781324001560
Amazon: https://www.amazon.com/How-Charts-Lie-Getting-
Information/dp/1324001569/ref=sr_1_1?keywords=how
+charts+lie&qid=1551105523&s=gateway&sr=8-1

How The Public Misinterprets Data Visualizations - Alberto Cairo

Uploaded by

Copyright:

Available Formats

You might also like

How The Public Misinterprets Data Visualizations - Alberto Cairo

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

How The Public Misinterprets Data Visualizations - Alberto Cairo

Uploaded by

Copyright:

Available Formats

ESS DATA VISUALISATION MILAN 2019

How Charts Lie

misinterprets data visualizations

W.W. Norton: https://books.wwnorton.com/books/

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

What can I say? I'm a sucker for statistics

Alberto Cairo author of Numbersense

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

DATA CLEANING GENERATING STYLING AND OUTPUT

Excel Charts and graphs

Pictorial lllustrations and icons

Maybe we need to spend a bit less time DESIGNING visualizations

Although we can all do both simultaneously

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

1. “A picture is worth a thousand words”

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

1. “A picture is worth a thousand words”

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

IGRAPHICACY (“graphical illiteracy)

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

The Affordable Care Act

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

The Affordable Care Act

Stimulus package passed on February 2009

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Map by Kenneth Field

SHARE OF THE POPULAR VOTE IN THE 2016 PRESIDENTIAL ELECTION

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

SHARE OF THE POPULAR VOTE IN THE 2016 PRESIDENTIAL ELECTION

PERCENTAGE OF ELIGIBLE VOTERS

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

ELECTORAL TRUMP CLINTON

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

1880: 56.8°F 2012: 58.2°F

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

1880: 56.8°F 2012: 58.2°F

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Most places are pretty safe, and

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo

Some places are so

Most places are pretty safe, and

Careful with amalgamation

Alberto Cairo • University of Miami • www.thefunctionalart.com • Twitter: @albertocairo