+1(978)310-4246 credencewriters@gmail.com
  

Introduction

Initial Post: This website (

5 Ways Writers Use Misleading Graphs To Manipulate You [INFOGRAPHIC]

) explains the top five reasons charts can be misleading. These were also seen in the

Ch 2 of

How Charts Lie

as well. For this class discussion – after reading through the chapters and viewing this website to summarize, find a chart that was displayed through traditional and/or social media or a company organization that created a chart that may be considered misleading. Explain specific reasons why it is misleading i.e. zero baseline, not scaled properly etc. and summarize the consequences that this may have on the viewer and the larger society as well.

Response Posts: After reviewing your colleagues’ posts, follow up with a comment, constructive criticism, or additional view to at least

TWO

colleague posts.

How Charts Lie
Getting Smarter about
Visual Information
Alberto Cairo
Praise for
How Charts Lie
“Funny and engaging. A must-read for anyone who wants to stay informed.”
—CATHY O’NEAL, author of Weapons of Math Destruction
“This book offers a succinct, elegant, accessible look at the ways data can be represented or
misrepresented and is a perfect primer for anyone who cares about the difference. I loved
this book!”
—CHARLES WHEELAN, author of Naked Statistics
“I wish we lived in a world where you didn’t need to read Alberto Cairo’s How Charts Lie, a
robust guide to self-defense against graphs and figures designed to mislead. But here we are,
and yes, you do.”
—JORDAN ELLENBERG, author of How Not to Be Wrong
“Alberto Cairo has written a wise, witty, and utterly beautiful book. You couldn’t hope for a
better teacher to improve your graphical literacy.”
—TIM HARFORD, author of
The Undercover Economist and presenter of More or Less
“Alberto Cairo shares great examples of data visualization and storytelling for anyone who
wants to dig into their data.”
—DONA WONG, author of The Wall Street Journal
Guide to Information Graphics
“A picture may be worth a thousand words, but only if you know how to read it. In this book,
Alberto Cairo teaches us how to get smarter about visual information by reading charts with
attention and care. I found a lot to steal here, and you will, too.” —AUSTIN KLEON, author
of Steal Like an Artist
“This book will open your eyes to how everyone uses visuals to push agendas. A master visual
designer, Alberto Cairo shows you how to read charts and decode design. After this book, you
can’t look at charts with a straight face!”
—KAISER FUNG, author of
Numbers Rule Your World
This document contains the
Introduction of Alberto Cairo’s
How Charts Lie
To purchase the book visit
Indibound.com
Amazon.com
Amazon.co.uk
Barnes&Noble
How
Charts
Lie
Getting Smarter about
Visual Information
Alberto Cairo
HowChartsLie_txt_final.indd 3
7/15/19 11:12 AM
Introduction
O
n April 27, 2017, President Donald J. Trump sat with Reuters journalists Stephen J. Adler, Jeff Mason, and Steve Holland to discuss
his accomplishments in his first 100 days in office. While talking
about China and its president, Xi Jinping, Trump paused and handed the
three visitors copies of a 2016 electoral map:1
Share of vote in the
2016 presidential election
More Democratic More Republican
(Source: Cook Report)
HowChartsLie_txt_final.indd 1
7/15/19 11:12 AM
2
How Charts Lie
The president then said, “Here, you can take that, that’s the final map of
the numbers. It’s pretty good, right? The red is obviously us.”
When I read the interview, I thought that it was understandable President Trump was so fond of that map. He won the 2016 election despite most
forecasts, which gave him between 1% and 33% chances of succeeding; a
Republican establishment that distrusted him; a bare-​bones campaign that
was often in disarray; and numerous controversial remarks about women,
minorities, the U.S. intelligence services, and even veterans. Many pundits
and politicians predicted Trump’s demise. They were proved wrong. He
seized the presidency against all odds.
However, being victorious isn’t an excuse to promote misleading visuals.
When presented alone and devoid of context, this map can be misleading.
The map appeared in many other places during 2017. According to The
Hill,2 White House staffers had a large, framed copy of it hanging in the
West Wing. The map was also regularly touted by conservative media organizations, such as Fox News, Breitbart, and InfoWars, among others. Right-​
wing social media personality Jack Posobiec put it on the cover of his book,
Citizens for Trump, which looks similar to this:
CITIZENS
TRUMP
FOR
THE INSIDE STORY OF THE PEOPLE’S MOVEMENT
TO TAKE BACK AMERICA
JACK POSOBIEC
HowChartsLie_txt_final.indd 2
7/15/19 11:12 AM
Introduction
3
I’ve spent the last two decades making charts and teaching others how
to design them. I’m convinced that anyone—​including you, reader—​can
learn how to read and even create good graphics, so I’m usually happy to
offer my free and constructive advice to whoever wants to take it. When I
saw Posobiec’s book on social media, I suggested that he needed to change
either the title or the map, as the map doesn’t show what the book title says.
The map is misleading because it’s being used to represent the citizens
who voted for each candidate, but it doesn’t. Rather, it represents territory. I
suggested that Posobiec either change the graphic on the cover of his book to
better support the title and subtitle, or change the title to Counties for Trump,
as that is what the map truly shows. He ignored my advice.
Try to estimate the proportion of each color, red (Republican) and grey
(Democratic). Roughly, 80% of the map’s surface is red and 20% is grey. The
map suggests a triumph by a landslide, but Trump’s victory wasn’t a landslide at all. The popular vote—​Posobiec’s “citizens”—​was split nearly in half:
Share of the popular vote in the 2016 presidential election
Donald Trump
Hillary Clinton
Other candidates
46.1% 62,984,825 votes
48.2% 65,853,516 votes
5.7%
We could be even pickier and point out that turnout in the election was
around 60%;3 more than 40% of eligible voters didn’t show up at the polls.
If we do a chart of all eligible voters, we’ll see that the citizens who voted for
each of the major candidates were a bit less than a third of the total:
Percentage of eligible voters
Didn’t vote
Voted for Donald Trump
Voted for Hillary Clinton
Voted for other candidates
40%
28%
29%
3%
And what if we count all citizens? There are 325 million people in the
United States. Of those, around 300 million are citizens, according to the
HowChartsLie_txt_final.indd 3
7/15/19 11:12 AM
4
How Charts Lie
Kaiser Foundation. It turns out that “Citizens for Trump” or “Citizens for
Clinton” are just a bit more than one-​fifth of all citizens.
Critics of President Trump were quick to excoriate him for handing out
the county-​level map to visitors. Why count the square miles and ignore the
fact that many counties that went for Trump (2,626)4 are large in size but
sparsely populated, while many of the counties where Clinton won (487) are
small, urban, and densely populated?
That reality is revealed in the following map of the continental U.S.,
designed by cartographer Kenneth Field. Each dot here represents a voter—​
grey is Democratic and red is Republican—​and is positioned approximately—​
but not exactly—​where that person voted. Vast swaths of the U.S. are empty:
As someone who strives to keep a balanced media diet, I follow people
and publications from all ideological stripes. What I’ve seen in recent years
makes me worry that the increasing ideological polarization in the U.S. is
also leading to a divide on chart preferences. Some conservatives I read love
the county-​level map President Trump handed out to reporters. They constantly post it on their websites and social media accounts.
HowChartsLie_txt_final.indd 4
7/15/19 11:12 AM
Introduction
5
Liberals and progressives, on the other hand, prefer a bubble map proposed by Time magazine and other publications.5 In it, bubbles are sized in
proportion to the votes received by the winning candidate in each county:
Bubble size is proportional
to the number of votes received
just by the candidate who
won in each county.
More votes for Donald Trump
More votes for Hillary Clinton
Both conservatives and liberals laugh at the other side’s stupidity. “How can
you tweet that map? Don’t you see that it distorts the results of the election?”
This is no laughing matter. Both sides in this debate are throwing different charts at each other because we all often use information to reinforce
our beliefs: conservatives love to convince themselves of a crushing victory
in the 2016 election; liberals console themselves by emphasizing Hillary
Clinton’s larger share of the popular vote.
Liberals are correct when they claim that the colored county map isn’t an
adequate representation of the number of votes each candidate received, but
the bubble map favored by liberals is also faulty. By showing only the votes
for the winning candidate in each county, this chart ignores those received
by the losing candidate. Plenty of people voted for Hillary Clinton in conservative regions. Many voted for Donald Trump in very progressive ones.
Kenneth Field’s map or the pair of maps below may be a better choice if
what we care about is the popular vote. There are many more visible red bubbles (votes for Trump) than grey bubbles (votes for Clinton), but the fewer
grey ones are often much bigger. When these maps are put side by side, it’s
easier to see why the election was decided by a relatively small number of
HowChartsLie_txt_final.indd 5
7/15/19 11:12 AM
6
How Charts Lie
votes in a handful of states; if you add up the area of all red bubbles and the
area of all grey bubbles, they are roughly the same:
Votes for Donald Trump
Votes for Hillary Clinton
Bubble size is proportional to the number of votes per county
Having said this, both conservatives and liberals are missing the point.
What makes you win a presidential election in the United States is neither
the territory you control, nor the number of people you persuade to vote for
you nationally. It’s the Electoral College and its 538 electors. To win, you
need the support of at least 270 of electors.
Each state has a number of these folks equal to its congressional representation: two senators plus a number of representatives in the House
that varies according to the state’s population. If you are a small state with
the fixed number of senators (two per state) plus one representative in the
House, you are allotted three electors.
Small states often have more electors based on their populations than
what pure arithmetic would give them: the minimum is three electors per
state, no matter how small the population of that state is.
Here’s how you receive the support of a state’s electors: with the exception of Nebraska and Maine, the candidate who wins even a razor-​thin
advantage in a state’s popular vote over his or her opponents is supposed to
receive the support of all that state’s electors.
In other words, once you’ve secured at least one more vote than any of
your opponents, the rest of the votes you receive in that state are useless. You
don’t even need a majority, just a plurality: if you get 45% of the popular vote
HowChartsLie_txt_final.indd 6
7/15/19 11:13 AM
Introduction
7
in one state, but your two opponents get 40% and 15%, you’ll receive all the
electoral votes from that state.
Trump got the support of 304 electors. Clinton, despite winning the
national popular vote by a margin of three million and getting tons of support in populous states like California, received only 227. Seven electors
went rogue, voting for people who weren’t even candidates.
Therefore, if I ever got elected president—​which is an impossibility,
since I wasn’t born in the U.S.—​and I wanted to celebrate my victory by
printing out some charts, framing them, and hanging them on the walls of
my White House, it would be with the ones below. They are focused on the
figures that really matter—​neither the number of counties, nor the popular
vote, but the number of electoral votes received by each candidate:
Electoral votes
Trump
Clinton
304
227
270
Who won in each state
Other: 7
State size adjusted by electoral votes
it contributes to the election
Maps are among the many kinds of charts you’ll learn about in this book.
Sadly, they are among the most misused. In July of 2017, I read that a popular
U.S. singer called Kid Rock was planning to run for the Senate in the 2018
election.6 He’d later claim that it was all a joke,7 but it sounded like a serious
bid at the time.
I didn’t know much about Kid Rock, so I wandered through his social
HowChartsLie_txt_final.indd 7
7/15/19 11:13 AM
8
How Charts Lie
media accounts and saw some of the merchandise he was selling in his
online store, KidRock.com. I love graphs and maps, so one T-​shirt with an
intriguing map of the results of the 2016 election was irresistible. Its legend
indicated that, according to Mr. Rock, the results of the election matched
the boundaries of two separate countries:
United
States
ofofAmerica
United
States
America
Dumbfuckistan
Dumbfuckistan
As you might now expect, this map isn’t an accurate representation
of the borders between the United States (read: Republican America) and
Dumbfuckistan (read: Democratic America). An electoral precinct-​or
county-​level map may be much more accurate.
Now, as an aside, I want to point out that I lived in North Carolina
between 2005 and 2008. Originally from Spain, I knew little about the Tar
Heel State before I arrived, other than that it was often red on the presidential
election maps I’d always seen in Spanish newspapers. I was expecting to settle in a conservative place. Fine with me. I’m ideologically moderate. But my
expectations were misguided. To my surprise, when I arrived, I didn’t land
in the United States of America—​if we follow Kid Rock’s nomenclature—​I
landed in deep Dumbfuckistan! The Chapel Hill–​Carrboro area, in Orange
County (North Carolina), where I lived, is quite progressive and liberal,
more so than most of the rest of the state.
The city where I am now, Kendall (Florida), part of the greater Miami
area, is also quite proud of its Dumbfuckistani heritage. The following maps
HowChartsLie_txt_final.indd 8
7/15/19 11:13 AM
Introduction
9
reveal what I’d say are the true borders between the two countries Mr.
Rock’s T-​shirt depicts:
United States of America
Dumbfuckistan
Places where I’ve lived
President Donald Trump gave his first State of the Union address on Jan-
uary 30, 2018. Pundits on the right sang praises to his great performance
as he read from a teleprompter, and those on the left criticized him. Trump
devoted some time to talking about crime and got the attention of economist
and Nobel Prize winner Paul Krugman, a columnist for the New York Times.
On several occasions during the presidential campaign in 2016, and
also during his first year in office, Trump mentioned a supposedly sharp
increase of violent crime in the United States, particularly murders.
Trump blamed undocumented immigrants for this, an assertion that has
been debunked many times over and that Krugman called a “dog whistle”
in his column.8
However, Krugman didn’t stop there. He added that Trump wasn’t
“exaggerating a problem, or placing the blame on the wrong people. He was
inventing a problem that doesn’t exist” as “there is no crime wave—​there
have been a few recent bobbles, but many of our big cities have seen both a
surge in the foreign-​born population and a dramatic, indeed almost unbelievable, decline in violent crime.”
Here’s a chart that Krugman provided as evidence:
HowChartsLie_txt_final.indd 9
7/15/19 11:13 AM
10
How Charts Lie
U.S. murder rate (yearly murders per 100,000 people)
12
8
4
0
1960
1966
1972
1978
1984
1990
1996
2002
2008
2014
(Source: Bureau of Crime Statistics)
It seems that what Krugman said is true: the United States has witnessed a noticeable drop in murders since the peaks in the 1970s, 1980s, and
early 1990s. The trend is similar for violent crime in general.
However, isn’t it odd that an article published at the beginning of 2018
includes only years up to 2014? While detailed crime statistics are hard to
obtain, and it would be impossible to get a good estimate up to the day when
Krugman’s column was published, the FBI already had solid stats for 2016
and a preliminary estimate for 2017.9 This is what the chart looks like if
we add those years. The murder rate increased in 2015, 2016, and 2017. It
doesn’t look like a “bobble” at all:
U.S. murder rate (yearly murders per 100,000 people)
12
8
2017*
4
0
1960
1966
1972
1978
1984
1990
1996
2002
2008
2014
*Preliminary 2017 estimate (obtained on January 31, 2018)
I doubt that someone with Krugman’s record would conceal relevant
data intentionally. Based on my own experience as a chart designer and
journalist who’s made plenty of silly mistakes, I’ve learned to never attribute
to malice what could be more easily explained by absentmindedness, rashness, or sloppiness.
It’s true, as Krugman wrote, that the murder rate today is way lower
HowChartsLie_txt_final.indd 10
7/15/19 11:13 AM
Introduction
11
than it was thirty years ago. If you zoom out and take a look at the entire
chart, the overall long-​term trend is one of decline. Tough-​on-​crime politicians and pundits often ignore this, quite conveniently, and focus instead
on the last few years.
However, the uptick since 2014 is relevant and shouldn’t be concealed.
How relevant is it, though? That depends on where you live.
This national murder rate chart, as simple and easy to read as it looks,
hides as much as it reveals. This is a common feature of charts, since they are
usually simplifications of very complex phenomena. Murders aren’t increasing everywhere in the United States. Most places in the U.S. are pretty safe.
Instead, murder in the U.S. is a localized challenge: some neighborhoods in mid-​sized and big cities have become so violent that they distort
the national rate.10 If we could plot those neighborhoods on the chart, they
would be way above its upper gridline, perhaps even beyond the top edge of
the page! If we took them off the chart, the national-​murder-​rate line might
stay flat or even go down in recent years.
Doing this wouldn’t be appropriate, of course: those cold numbers represent people being killed. However, we can and should demand that, when
discussing data like this, politicians and pundits mention both overall rates
and the extreme values—also called “outliers”—that may be distorting those rates.
Here’s an analogy to convey the statistics and help you grasp the role
of outliers: Imagine you’re in a bar enjoying a beer. Eight other people are
drinking and chatting. None of you has killed anyone in your life. Then, a
tenth person comes in, a hitman for the Mob who’s dispatched 50 rivals in his
career. Suddenly, the average kill count per drinker in the bar jumps to 5! But
of course that doesn’t automatically make you an assassin.
Charts may lie, then, because they display either the wrong information or
too little information. However, a chart can show the right type and amount
of information and lie anyway due to poor design or labeling.
HowChartsLie_txt_final.indd 11
7/15/19 11:13 AM
12
How Charts Lie
In July 2012, Fox News announced that President Barack Obama was
planning to let President George W. Bush’s cuts to the top federal tax rate
expire by the beginning of 2013. The very wealthy would see their tax bills
increase. By how much? Please estimate the height of the second bar in comparison with the first one, which represents the top tax rate under President
Bush. It’s a massive tax increase!
If Bush
tax cuts
expire
Top tax rate:
35%
Now
?
Jan. 2013
The chart that Fox displayed for a few seconds contained the figures,
but they were quite tiny and hard to read. Notice that the tax increase was
roughly five percentage points, but the bars were grossly misshapen to exaggerate it:
If Bush
tax cuts
expire
39.6%
40%
38%
Top tax rate:
36%
35%
Now
HowChartsLie_txt_final.indd 12
Jan. 2013
34%
7/15/19 11:13 AM
Introduction
13
I like taxes as little as anyone else, but I dislike arguments defended
with dubious charts even more, regardless of their creators’ political leanings. Whoever designed this chart broke an elementary principle of chart
design: if your numbers are represented by the length or height of objects—​
bars, in this case—​the length or height should be proportional to those numbers. Therefore, it’s advisable to put the baseline of the chart at zero:
If Bush
tax cuts
expire
39.6%
35%
Top tax rate:
Now
Jan. 2013
0%
Starting a bar chart at a baseline different from zero is the most conspicuous trick in the book to distort your perception of numbers. But fudging
with scales is just one of the many strategies used by swindlers and liars
from all ideological denominations. There are many others that are far less
easy to spot, as we’ll soon see.
Even if a chart is correctly designed, it may still deceive us because we
don’t know how to read it correctly—​we can’t grasp its symbols and grammar, so to speak—​or we misinterpret its meaning, or both. Contrary to what
many people believe, most good charts aren’t simple, pretty illustrations
that can be understood easily and intuitively.
On September 10, 2015, the Pew Research Center published a survey
HowChartsLie_txt_final.indd 13
7/15/19 11:13 AM
14
How Charts Lie
testing U.S. citizens’ knowledge of basic science.11 One of the questions
asked participants to decode the following chart. Try to read it and don’t
worry if you get it wrong:
Relationship between sugar consumption
per person and average number of decayed teeth
Each dot is
a country
10
8
Average number 6
of decayed teeth
per person in 4
different countries
2
0
0
(Source: Pew Research Center)
20
40
60
80 100 120 140
Average sugar consumption
(grams per person per day)
In case you’ve never seen a chart like this, it’s called a scatter plot. Each
dot is a country; we don’t need to know which one. The position of each dot
on the horizontal axis corresponds to the daily sugar consumption per person. In other words, the farther to the right a dot is, the more sugar people
consume in that country, on average.
The position of a dot on the vertical axis corresponds to the number of
decayed teeth per person. Therefore, the higher up a dot is, the more bad
teeth people in that country have, on average.
You’ve probably detected a pattern: in general, and with some exceptions, the farther to the right a dot is, the higher up it tends to be as well.
This is called a positive correlation between two metrics: sugar intake is positively correlated with worrisome dental health at the country level. (This
chart on its own does not prove that more sugar leads to more decayed teeth,
but we’ll get to that soon.) Correlations can also be negative; for instance,
HowChartsLie_txt_final.indd 14
7/15/19 11:13 AM
Introduction
15
the more educated countries are, the smaller the percentage of poor people
they usually have.
The scatter plot is a kind of chart that is almost as old as those we all
learn to read in elementary school, such as the bar graph, the line graph,
and the pie chart. Still, roughly 4 out of 10 people in the survey (37%)
couldn’t interpret it correctly. This may have to do in part with how the
questions in the survey were posed or some other factors, but it still suggests to me that a large portion of the population struggles to read charts
that are commonplace in science and that are also becoming common in
the news media.
And it’s not just scatter plots. It also happens with charts that, at least at
first glance, look easy to read. A group of researchers from Columbia University showed the following pictorial chart to more than 100 people:12
Fruit servings per week
Victor
Other men in
Victor’s age group
Recommendation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
(Source: Adriana Arcia, Columbia University School of Nursing)
The chart reveals that “Victor,” an imaginary fellow, is consuming more
fruit servings per week than other men of his age, but fewer than the recommended 14 servings per week.
What the chart is intended to say is: “Victor is currently eating 12
servings of any kind of fruit every week. He’s eating more than the average man in his age group, but 12 servings aren’t enough. He should be
eating 14.”
Many participants read the chart too literally. They thought that Victor
needed to eat the exact same amount and kinds of fruits portrayed in the
HowChartsLie_txt_final.indd 15
7/15/19 11:13 AM
16
How Charts Lie
chart 14 times every week! A participant even complained, “But a whole
pineapple?” The results were similar if the icon used to represent “fruit serving” was a single apple. In that case, one participant complained about the
“monotony” of eating the same fruit every day.
Charts are still seductive and persuasive, whether or not many people
are able to read them correctly. In 2014, a team of researchers from New
York University conducted several experiments to measure how persuasive
charts are in comparison with textual information.13 They wanted to see
whether three charts—​about the corporate income tax, incarceration rates,
and the reasons children play video games—​modified people’s opinions. For
instance, in the case of video games, the goal was to show participants that,
contrary to some messages in the media, children don’t play video games
because they enjoy violence, but because they want to relax, let their imaginations fly, or socialize with friends.
Many participants’ minds changed because of the charts, particularly
if they didn’t have strong preexisting opinions about the charts’ topics. The
authors conjectured that this happened “partially due to the increased sense
of objectivity” that “evidence supported by numbers carries.”
Studies like this have limitations, as their authors themselves acknowledged. For instance, it’s hard to tell what exactly participants found persuasive: Was it the visual representation of the numbers or the numbers
themselves? As the saying goes, more research is needed, but the tentative
evidence we have suggests that many of us are cajoled by the mere presence
of numbers and charts in the media we consume, no matter whether we can
interpret them well.
The persuasiveness of charts has consequences. Very often, charts lie to
us because we are prone to lying to ourselves. We humans employ numbers
and charts to reinforce our opinions and prejudices, a psychological propensity called the confirmation bias.14
HowChartsLie_txt_final.indd 16
7/15/19 11:13 AM
Introduction
17
Republican congressman Steve King, a strong proponent of strict limits
to immigration, posted on Twitter in February 2018:
Illegal immigrants are doing what Americans are reluctant to do.
We import young men from cultures with 16.74 times the violent
death rate as the U.S. Congress has to KNOW more Americans
will die as a result.15
King also added a table. The United States isn’t shown, but it’s in the
85th position, with a violent death rate of around 6 per 100,000 people:
Violent death rate per 100,000 people
Rank
1
2
3
4
5
6
7
8
9
10
Country
Rate
93
El Salvador
Guatemala
71
Venezuela
47
Trinidad-Tobago 43
Belize
43
Lesotho
42
Colombia
37
Honduras
36
Swaziland
36
Haiti
35
Rank
11
12
13
14
15
16
17
18
19
20
Country
Panama
D.R. Congo
Brazil
South Africa
Mexico
Jamaica
Guyana
Rwanda
Nigeria
Uganda
Rate
34
31
31
29
27
27
26
24
21
20
King was fooled by his own data and chart and, as a result, he likely
also fooled some of his constituents and followers. These countries are very
violent, yes, but you cannot infer from the chart alone that the people moving
from them to the United States have violent inclinations. The opposite may
be true! It may well be that immigrants and refugees from dangerous countries are the meek and the peaceful, fleeing from societies where they can’t
work hard and thrive because they’re being harassed by criminals.
To give you an anecdotal analogy, an enormous number of Spanish men
my age love soccer, bullfighting, Flamenco dance, and the reggaeton song
“Despacito.” I’m a Spaniard, but I don’t like any of those, and neither do any
of my closest Spanish friends, who prefer to engage in much dorkier rec-
HowChartsLie_txt_final.indd 17
7/15/19 11:13 AM
18
How Charts Lie
reations, such as strategic board games and reading comic books, popular-​
science books, and science fiction. We must always be wary of inferring
features of individuals based on statistical patterns of populations. Scientists
call this the ecological fallacy.16 You’ll soon learn more about it.
Charts may lie in multiple ways: by displaying the wrong data, by includ-
ing an inappropriate amount of data, by being badly designed—or, even if
they are professionally done, they end up lying to us because we read too
much into them or see in them what we want to believe. At the same time,
charts—​good and bad—​are everywhere, and they can be very persuasive.
This combination of factors may lead to a perfect storm of misinformation and disinformation. We all need to turn into attentive and informed
chart readers. We must become more graphicate.
Geographer William G. V. Balchin coined the term “graphicacy” in the
1950s. In a 1972 address to the annual conference of the Geographical Association, he explained its meaning. If literacy, said Balchin, is the ability to
read and write, articulacy is the ability to speak well, and numeracy the
ability to manipulate numerical evidence, then graphicacy is the ability to
interpret visuals.17
The term “graphicacy” has appeared in numerous publications since
then. Two decades ago, cartographer Mark Monmonier, author of the classic book How to Lie with Maps, wrote that any educated adult should possess a good level of not just literacy and articulacy but also numeracy and
graphicacy.18
This is even truer now. Public debates in modern societies are driven by
statistics, and by charts, which are the visual depiction of those statistics. To
participate in those discussions as informed citizens, we must know how to
decode—​and use—​them. By becoming a better chart reader, you may also
become a better chart designer. Making charts isn’t magic. You can create
them with programs installed on common personal computers or available
HowChartsLie_txt_final.indd 18
7/15/19 11:13 AM
Introduction
19
on the web, such as Sheets (Google), Excel (Microsoft), Numbers (Apple),
open-​source alternatives such as LibreOffice, and many others.19
By now you’ve seen that charts can indeed lie. I hope to prove to you,
however, that by the end of this book you’ll be able to not only spot the lies
but also recognize the truths in good charts. Charts, if designed and interpreted properly, can indeed make us smarter and inform conversations. I
invite you to open your eyes to their wondrous truths.
HowChartsLie_txt_final.indd 19
7/15/19 11:13 AM
Anusha Ramachandruni posted Jul 7, 2022 1:23 AM
Although the following example shows how the same data can be
displayed in many ways, you shouldn’t assume that they are always
interchangeable. With a bar chart, you can select any numbers you like
for the numeric value axis.
The sums across groups will typically be equal to the sum across the
data as a whole for numeric values that represent metric totals or data
point counts. In scenarios like this, pie charts and bar charts are both
acceptable visualization options.
In this situation, a bar chart works better than a pie chart. It is too
simple for a reader to think that the sum of slices represents some kind
of total because the circular shape suggests that slices are components
of a whole.
A pie chart is designed for showing how each portion contributes to
that whole and can only be used if the sum of the different parts adds
up to a meaningful whole. A bar chart, on the other hand, can be
utilized for a wider variety of data types in addition to dissecting an
entire thing into its component parts.
Overall, compared to a pie chart, a bar chart is a far more informationdense depiction. In actuality, a bar chart should be your first option. It’s
best to play it safe with a bar chart if you’re unsure whether a pie chart
will be a decent choice of visualization.
Sri Priya Nalla posted Jul 6, 2022 11:14 PM
Subscribe
I found the graph shown below on the internet which used colors to indicate
difference between the data. The data we are trying to convey to the user can be
easier to understand by using color, but it can cause substantial confusion for the
user if we use inappropriate colors. It is important to choose the colors wisely
which helps to communicate the story you want to tell the user.
The map chart shown below is misleading as it used too many colors and some
of them being too contrast makes it difficult for the viewer to quickly understand
what we are trying to convey. It would be easy to understand if we use different
shades of the same color family in which lighter shade indicating a smaller
number of cases and darker shade indicating a greater number of cases. If we use
dark color to indicate a smaller number of cases, it might mislead the customer.it
would also be easier to understand if the map has the state name indicated on
each state.
The inaccuracy of the chart may appear to be a minor issue, but it causes
improper understanding of the data by user which may lead to flawed insights
and poor business decisions. Misrepresenting Data Visualization is particularly
alarming in terms of manipulating the public conscience.

Purchase answer to see full
attachment

  
error: Content is protected !!