Hypothesis Annotation for readings
EK17
Schwabish_1-2.pdf
B efore we start creating our charts and graphs, we need to cover some basic theory of how the brain perceives visual stimuli. This will guide you as you decide what chart type
is most appropriate to visualize your data. When we consider how to visualize our data, we must ask ourselves how accurately the
reader can perceive the data values. Are some graphs better equipped to guide the reader to the specific difference between, say, 2 percent and 2.3 percent? If so, how should we think about those differences as we create our visualizations?
There’s a thread of research in the data visualization field that explores this very ques- tion. Based on original research over the past thirty years or so, the image on the next page shows a spectrum of graphs—or more generally, types of data encodings like dots, lines, and bars—arrayed by how easily readers can estimate their value. The encodings that readers can most accurately estimate are arranged at the top, and those that enable more general estimates are at the bottom.
The rankings are unsurprising. It is easier to compare the data in line charts, bar charts, and area charts that have the same axis or baseline. Graphs on which the data are positioned on unaligned axes—think of a pair of bars that are offset from one another on different axes—are slightly harder for us to accurately discern the values.
Farther down the vertical axis are encodings based on angle, area, volume, and color. You intuitively know this: it’s much easier to discern the exact data values and differences between values when reading a bar chart than when reading a map where countries are shaded with different colors.
VISUAL PROCESSING AND PERCEPTUAL RANKINGS �1
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
Enable accurate
May enable general
Length
Area
Angle Parts of a whole
Color hue
Perceptual ranking diagram. What kind of data visualization you choose to create will depend on your goals and your audience’s needs, experiences, and expertise. This image is
based on Alberto Cairo (2016) from research by Cleveland and McGill (1984), Heer, Bostock, and Ogievetsky (2010), and others.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
VISUAL PROCESSING AND PERCEPTUAL RANKINGS � 15
Standard graphs, like bar and line charts, are so common because they are perceptually more accurate, familiar to people, and easy to create. Nonstandard graphs—those that use circles or curves, for instance—may not allow the reader to most accurately perceive the exact data values.
But perceptual accuracy is not always the goal. And sometimes it’s not a goal at all. Spurring readers to engage with a graph is sometimes just as important. Sometimes, it’s
more important. And nonstandard chart types may do just that. In some cases, nonstandard graphs may help show underlying patterns and trends in better ways that standard graphs. In other cases, the fact that these nonstandard graphs are different may make them more engaging, which we may sometimes need to first attract attention to the visualization.
This graphic from information designer Federica Fragapane shows the fifty most vio- lent cities in the world in 2017. The vertical axis measures the population of each city and the horizontal axis captures the homicide rate per 100,000 people. The number of lines in each icon represents the number of homicides, and additional colors, shapes, and markers capture metrics like country of origin (the symbol in the middle of each), region (vertical dashed line), and change since 2016 (blue for decreases, red for increases). It could be a bar
Graphic from Frederica Fragapane for La Lettura—Correier della Serra that shows the fifty most violent cities in the world. See the next page for a closer look at the legend.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
A zoom-in of the graphic from Frederica Fragapane. Notice all of the details and data ele- ments included in each icon. It could be a bar chart or line chart, but would you then be
inclined to zoom in and read it closely?
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
VISUAL PROCESSING AND PERCEPTUAL RANKINGS � 17
chart or a line chart or some other chart type. But if it were, would you be inclined to zoom in, read it closely, and examine it?
Data visualization is a mix of science and art. Sometimes we want to be closer to the sci- ence side of the spectrum—in other words, use visualizations that allow readers to more accurately perceive the absolute values of data and make comparisons. Other times we may want to be closer to the art side of the spectrum and create visuals that engage and excite the reader, even if they do not permit the most accurate comparisons.
Sometimes you must make your visuals interesting and engaging, even at the cost of absolute perceptual accuracy. Readers may not be as interested in the topic as we hope or may not have enough expertise to immediately grasp the content. As content creators, how- ever, our job is to encourage people to read and use the graph, even if we “violate” perceptual rules that we know will hamper someone’s ability to make the most accurate conclusions. Thinking about different audience types is not just about considering among decision mak- ers, scholars, policymakers, and the general public—it also means thinking about different levels of interest or engagement with the visual itself. As historian Cecelia Watson writes in her book about the history and use of the semicolon, “What if we thought less about rules and more about communication, and considered it our obligation to one another to try to figure out what is really being communicated?”
We should not operate from the assumption that readers will pay attention to everything in our visual, even if we use a common, familiar chart type. Let’s be honest: People see bar charts and line charts and pie charts all the time, and those charts are often boring. Boring graphs are forgettable. Different shapes and uncommon forms that move beyond the bor- ders of our typical data visualization experience can draw readers in. Reading a graph is not like the spontaneous comprehension of seeing a photograph. Instead, reading a graph has more of the complex cognitive processes as reading a paragraph.
This isn’t to say we should not concern ourselves with visual perception or allowing our readers to make the most accurate comparisons, but the goal of engagement can be worth a lot in its own right. Elijah Meeks, a data visualization engineer, wrote that, “Charts, like any other communication, need to be compelling to be convincing, and if your bar chart, as opti- mal as it may be, has been reduced to background noise by the constant hum of bar charts crossing a stakeholder’s screen, then it’s your responsibility to make it more compelling, even if it’s not any more precise or accurate than a more simple form.”
Introducing a new or different graph type can also introduce a hurdle to your reader. These can be big hurdles, like a completely new graph type or an exceptionally unusual
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
18 � PRINCIPLES OF DATA VISUALIZATION
representation of the data. Or they can be small hurdles, graphs that rank lower on the perceptual-accuracy scale or graphs that people may have only seen a few times before. To overcome these hurdles, you may need to explain how to read the graph. But that might be worth it because sometimes different charts attract reader’s attention and pique their curiosity.
When should you use a nonstandard graph? Likely not for many scholarly purposes, because they do not enable the most accurate perceptions of the data. For scholarly writing, accuracy is paramount. We want our reader to clearly and efficiently compare the values we’re presenting. But in other cases—headline-style or standalone graphics, blog posts, shorter briefs or reports, or graphs for social media—creating something different may draw people in and hold their attention just long enough to convey your argument, data, or content.
This graphic from an interactive visualization from the Organisation of Economic Co-Operation and Development (OECD) enables users to explore the different metrics
and definitions of what it means to have a “better life.” A more standard chart type, like a bar chart, might enable easier comparisons, but would it be as much fun?
Source: Organisation for Economic Co-Operation and Development
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
VISUAL PROCESSING AND PERCEPTUAL RANKINGS � 19
This visualization from artist and journalist Jaime Serra Palou is a lovely example of this kind of nonstandard and creative data visualization. He plots his coffee consump- tion every day over the course of a year by using the stains from his coffee cups. You can immediately see those parts of the year when he needed an extra burst of caffeine. Yes, a line chart might convey the same data, but would you pause to spend an extra moment reading it?
Sometimes you can do both—a nonstandard, attention-grabbing graphic accompanied by a more familiar graph next to it. What you present and how you present it depends on your audience. The Serra piece might work as the lead graphic on a book or report about coffee consumption, but more detailed charts inside might take the form of standard charts and tables. Some academic research has shown that creating novel graphs, such as
Artist and journalist Jaime Serra Palou plotted his coffee consumption every day for a year by using stains from his coffee cup.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
20 � PRINCIPLES OF DATA VISUALIZATION
those that enable the user to personalize the content (by inputting their own information) or are simply more aesthetically appealing, encourages readers to actively process the content.
ANSCOMBE’S QUARTET
The value of visualizing data is best illustrated by Anscombe’s Quartet, published in 1973 by statistician Francis Anscombe. The Quartet demonstrates the power of graphs and how they, together with statistical calculations, can better communicate our data.
Examine the table below, which shows four pairs of data, an X and a Y. We can make some basic observations about these data. We can see that the first three
series of X’s are all the same; the values of X’s in the last series are all 8 except for the one 19; and the X’s are all whole numbers while the Y’s are not. We might even notice that the 12.7 value in the third column of Y is larger than the rest. In my experience, most people don’t comment about the relationship between the different series, which, at the end of the day, is what we want to understand. It turns out that each of the four pairs yield the same standard information: the same average values of the X series and the Y series; the same variance for each; the same correlation between X and Y; and the same estimated regression equation.
Known as Anscombe’s Quartet, this example demonstrates how difficult it is for us to pull out basic patterns and summary statistics.
Data set 1 1 2 2 3 3 4 4
Variable x y x y x y x y
Obs. No. 1 : 10 8.0 10 9.1 10 7.5 8 6.6
2 : 8 7.0 8 8.1 8 6.8 8 5.8
3 : 13 7.6 13 8.7 13 12.7 8 7.7
4 : 9 8.8 9 8.8 9 7.1 8 8.8
5 : 11 8.3 11 9.3 11 7.8 8 8.5
6 : 14 10.0 14 8.1 14 8.8 8 7.0
7 : 6 7.2 6 6.1 6 6.1 8 5.3
8 : 4 4.3 4 3.1 4 5.4 19 12.5
9 : 12 10.8 12 9.1 12 8.2 8 5.6
10 : 7 4.8 7 7.3 7 6.4 8 7.9
11 : 5 5.7 5 4.7 5 5.7 8 6.9
Mean 9.0 7.5 9.0 7.5 9.0 7.5 9.0 7.5
Variance 11.0 4.1 11.0 4.1 11.0 4.1 11.0 4.1
Correlation
Regression line y = 3 + 0.5x y = 3 + 0.5x y = 3 + 0.5x y = 3 + 0.5x
0.816 0.816 0.816 0.817
Source: Francis Anscombe
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
VISUAL PROCESSING AND PERCEPTUAL RANKINGS � 21
When we see the same data presented in four graphs, however, we can immediately see these relationships, for example, the positive correlation in all four pairs, the curvature in the second pair that you couldn’t see in the table, and the outliers 12.7 and 19.0.
We are much more likely to remember these four small graphs than we are the origi- nal table. In his bestselling book, Brain Rules, molecular biologist John Medina writes, “The more visual the input becomes, the more likely it is to be recognized and recalled.” The more we can make our data and content visual, the more we can expect our readers to remember it and, hopefully, use it.
0
5
10
15
0 5 10 15 20 0
5
10
15
0 5 10 15 20
0
5
10
15
0 5 10 15 20 0
5
10
15
0 5 10 15 20
The data visualization representation of Anscombe’s Quartet. Notice how much easier it is to see the positive relationship between the two variables, the curvature in the
pattern in the top-right graph, and the outliers in the bottom two graphs. Source: Francis Anscombe (1973).
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
22 � PRINCIPLES OF DATA VISUALIZATION
GESTALT PRINCIPLES OF VISUAL PERCEPTION
How do we perceive information? And how, as chart creators, can we use these perceptual rules to more eff ectively communicate our data? “Gestalt theory” is one such way we can think about how our readers will look at our graphs. Gestalt theory was developed in the early part of the twentieth century by German psychologists and refers to how we tend to organize visual elements into groups. Further developments in the fi eld were interrupted by the rise of the Nazi regime in Germany and then by World War II, and aft er the war it was criticized for not having rigorous methodological methods. But the ideas persist in many disciplines, including information theory, vision science, and cognitive neuroscience.
Th ese six organizational principles from Gestalt theory are especially useful for creating graphs and visuals that tap into our reader’s visual processing network.
PROXIMITY
We perceive objects that are close to one another as belonging to a group. Th ere are lots of graphical elements that we can group together: labels with points, bars with each other, or, like this graph, clusters of points in a scatterplot in which we can see two groups or clusters, one in the top-right and the other closer to the bottom-left .
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
VISUAL PROCESSING AND PERCEPTUAL RANKINGS � 23
SIMILARITY
Our brains group objects that share the same color, shape, or direction. Adding color to the above scatterplot reinforces the two groups.
ENCLOSURE
Bounded objects are perceived as a group. Here, in addition to using color, we can enclose the two groups with circles or other shapes.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
2 4 � PRINCIPLES OF DATA VISUALIZATION
CLOSURE
Our brains tend to ignore gaps and complete structures with open areas. In its basic form, we don’t have a problem viewing a simple graph that has a horizontal axis and a vertical axis as a single object because the two lines are enough for us to define the closed space. In a line chart with missing data, for example, we tend to mentally close the gap in the most direct way pos- sible, even if there might be something different going on in that missing area. For example, in the line graph on the left, we mentally close the gap between the two segments with a straight line even though the missing data might yield a pattern that moves up and then down.
A B C D E
CONTINUITY
Here, objects that are aligned together or continue one another are perceived as a group. Hence, our eyes seek a smooth path when following a sequence of shapes. You don’t need the horizontal axis line in this bar chart, for example, because the bars are aligned along a consistent path between the labels and the bottoms of the bars.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
VISUAL PROCESSING AND PERCEPTUAL RANKINGS � 25
CONNECTION
According to this principle, we perceive connected objects as members of the same group. Take this series of dots: At first, we perceive it as a single series, a mass of blue dots. Adding color makes it clear there are two different series. Connecting the dots makes it clear how the two initially track each other but then diverge.
PREATTENTIVE PROCESSING
The concept of “preattentive processing” is a subset of Gestalt theory, and it is the visual pro- cess I consider most when creating my data visualizations. As we just saw, because our eyes can detect a limited set of visual characteristics, we combine various features of an object and unconsciously perceive them as a single image. In other words, preattentive attributes draw our attention to a specific part of an image or, in our case, a graph.
For example, try to find the four largest numbers in this table.
Table 1. Our sales grew to $600 million this year Q1 Q2 Q3 Q4
Bob 26 35 72 84 Ellie 22 15 61 35 Gerrie 19 20 71 55 Jack 22 95 13 64 Jon 83 62 46 48 Karen 30 65 98 82 Ken 38 28 45 71 Lauren 98 81 41 63 Steve 16 50 23 41 Valerie 46 24 30 57
Total $400 $475 $500 $600
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
26 � PRINCIPLES OF DATA VISUALIZATION
Preattentive attributes here direct our attention to the large numbers immediately.
Table 1. Our sales grew to $600 million this year Q1 Q2 Q3 Q4
Bob 26 35 72 84 Ellie 22 15 61 35 Gerrie 19 20 71 55 Jack 22 95 13 64 Jon 83 62 46 48 Karen 30 65 98 82 Ken 38 28 45 71 Lauren 98 81 41 63 Steve 16 50 23 41 Valerie 46 24 30 57
Total $400 $475 $500 $600
Table 1. Our sales grew to $600 million this year Q1 Q2 Q3 Q4
Bob 26 35 72 84 Ellie 22 15 61 35 Gerrie 19 20 71 55 Jack 22 95 13 64 Jon 83 62 46 48 Karen 30 65 98 82 Ken 38 28 45 71 Lauren 98 81 41 63 Steve 16 50 23 41 Valerie 46 24 30 57
Total $400 $475 $500 $600
It’s easier to fi nd the numbers in these two tables than the fi rst because the numbers are encoded using preattentive attributes : color and weight. Each distinction helps us eff ortlessly identify the key number.
Hard to do, right? Now try it with these versions that use color (on the left ) and intensity (on the right) to highlight those four numbers.
Shape Enclosure Line Width Saturation Color
Size Markings Orientation Position 3D
Length Curvature Density Closure Sharpness
Examples of preattentive attributes that we can use in our visualizations to direct our reader’s attention.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
VISUAL PROCESSING AND PERCEPTUAL RANKINGS � 27
Notice how your eye gravitates toward the four tomatoes in the top-right part of the image on the left. The image on the right is balanced, so your eye doesn’t
immediately focus on any particular area. Photos by NordWood Themes (left) and Tim Gouw (right) on Unsplash.
Preattentive attributes are effects that seem to pop out from their surroundings. There are many we can use to tap into our reader’s visual processing network to draw their attention: shape, line width, color, position, length, and more.
Preattentive processing works in photographs too. Consider these images of fruits and vegetables. In the photo on the left, the eye is drawn to the upper-right corner. The group of tomatoes is slightly larger than the rest and positioned away from the group. In the photo- graph on the right, however, the eye is not drawn to any specific position. This photograph is more evenly balanced, so no one object stands apart from the rest.
We can apply these attributes to data visualization. A line chart uses the position of the points to indicate the data, while a bar chart uses length. You can use preatten- tive attributes to draw your audience’s attention to aspects of your graphs, guiding their focus.
For example, on the next page, I can enclose the ‘Forecast’ area of the line chart on the left with the gray box—notice how it immediately draws your eye to the right side of the graph. Similarly, I can use the color attribute to highlight a few points in the scatterplot on the right (and keep the other dots gray).
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
28 � PRINCIPLES OF DATA VISUALIZATION
WRAPPING UP
With these basic rules of perception, we are now better equipped to recognize and interpret the visual features we can use to encode and highlight our data. Before we start adding more graphs to our data visualization toolbox, let’s lay out some basic guidelines of more eff ec- tive data visualizations—things you should keep in mind no matter what kind of graph you are creating.
Applying simple preattentive attributes to these graphs directs your eye to the “Forecast” area of the graph on the left and to the two highlighted countries in the graph on the right.
Actual Forecast 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
2010 2012 2014 2016 2018 2020 2022 2024
US Real GDP growth is projected to decline and stabilize around 1.7%
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
W henever I create a data visualization, whether it’s static, interactive, or part of a report or blog post or even a tweet, I follow five primary guidelines.
1. Show the data 2. Reduce the clutter 3. Integrate the graphics and text 4. Avoid the spaghetti chart 5. Start with gray
Showing the data and reducing the clutter means reducing extraneous gridlines, markers, and shades that obscure the actual data. Active titles, better labels, and helpful annotations will integrate your chart with the text around it. When charts are dense with many data series, you can use color strategically to highlight series of interest or break one dense chart into multiple smaller versions.
Taken together, these five guidelines remind me of the needs of my audience and how my visuals can tell them a story.
GUIDELINE 1: SHOW THE DATA
Your reader can only grasp your point, argument, or story if they see the data. This doesn’t mean that all the data must be shown, but it does mean that you should highlight the values
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS �2
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
30 � PRINCIPLES OF DATA VISUALIZATION
that are important to your argument. As chart creators, our challenge is deciding how much data to show and the best way to show it.
Consider this dot density map of the United States (see page 244 for more on this kind of map). It uses data from the 2010 U.S. decennial census and places a dot for each of the country’s 308 million residents in their census blocks (a census block roughly corresponds to a city block). Notice how there is nothing in the image except for the data. There are no state borders, roads, city markers, or labels for lakes and rivers. We still recognize it as the United States because people tend to live along borders and coasts, which helps give shape to the country.
This doesn’t mean we must show all of the data all the time. Sometimes charts show too much data, making it hard to see which data points matter most. On the next page are two line charts that both show the average number of years of schooling for fifty countries around the world. In the graph on the left, each country is assigned its own color. This makes
The Gestalt principle of similarity helps us see the clusters of people around the country. Source: Image Copyright, 2013, Weldon Cooper Center for Public Service, Rector and Visitors of the
University of Virginia (Dustin A. Cable, creator).
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 31
it busy and confusing, impossible to pick out a trend for any single country. In the graph on the right, just six countries of interest are highlighted while the remaining are set in gray, blending them into a neutral background. This gives the reader a clear view of the countries we want to highlight. It’s not about showing the least amount of data, it’s about showing the data that matter most.
GUIDELINE 2: REDUCE THE CLUTTER
The use of unnecessary visual elements distracts your reader from the central data and clut- ters the page. There are lots of different types of chart clutter we might want to avoid. There are basic elements like heavy tick marks and gridlines, which we should remove in almost every case. Some graphs use data markers like squares, circles, and triangles to distinguish between series, but when the markers overlap they jumble the patterns. Some use textured or filled gradients when simple, solid shades of color work just as well. Some use unnec- essary dimensions that distort the data. And others contain too much text and too many labels, cluttering the space and crowding out the data.
Take this three-dimensional column chart of average schooling for the United States and Germany for a few select years.
Highlighting just a few countries in the chart on the right makes it easier to read.
0
2
4
6
8
10
12
14
16
1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Source: Our World in Data
Average years of schooling has increased around the world
(Number of years)
0
2
4
6
8
10
12
14
16
Average years of schooling has increased around the world
(Number of years)
Source: Our World in Data
China
United States
Nepal
Mexico
Germany
Spain
1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
32 � PRINCIPLES OF DATA VISUALIZATION
If you think that this looks so outlandish that no one would ever style a chart this way, you’d be wrong. I’ve copied the exact style from another chart, even down to the gradient styl- ing. Th e three-dimensional bars and shimmering stripes, mismatched data and axis labels, the abundance of decimals that suggest a level of data precision that’s not actually there— all these combine to create a graph that is diffi cult to read and, quite honestly, uncomfortable to look at. Also notice how the three-dimensional view distorts the data. Th e fi rst bar never touches the gridline even though it should match it exactly. Th is distortion occurs because the unnecessary third dimension requires adding perspective to the graph. Simplifying the graph by discarding these extraneous, distracting elements and showing the data makes your argument clear and comprehensible.
While much of our understanding of perception and how our eyes and brains work is rooted in scientifi c research, our decisions of which graph to use, where we place labels and annotation, which colors and fonts to use, and how we lay out our visualizations is mostly subjective. Th ere are cases where certain graphs are wrong, but many other cases call for
You’ve seen these kinds of 3D charts before—they are distracting, hard to read, and distort the data.
0.00%
500.00%
1000.00%
1500.00%
1997 2007
2017
10,000
13,700 14,10012,700 12,900 13,400
Germany United States
Average years of schooling has grown faster in Germany than in the United States
(Number of years)
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 33
nothing more than your best judgment. As you create more visualizations and read more graphs, you’ll develop your own eye and aesthetic—and your own balance of art and science.
GUIDELINE 3: INTEGRATE THE GRAPHICS AND TEXT
Although our primary focus on creating a visualization is the graphic elements—bars, points, or lines—the text we include in and around our graphs is just as important. Far too often, we treat the text and annotations as an afterthought, but these elements can be used to explain how to read the content in the graph as well as how to read the graph itself. Amanda Cox, the Data Editor at the New York Times, once said that “The annotation layer is the most important thing we do . . . otherwise it’s a case of ‘here it is, you go figure it out.’ ”
Adding the right annotations to a graph can be vitally important to your reader’s com- prehension. There are three ways we can integrate our graphs and our visuals: removing legends, creating active titles, and adding detail.
Average years of schooling has grown faster in Germany than in the United States (Number of years)
10.0
13.7 14.1
12.7 12.9 13.4
1997 2007 2017
Germany United States
A basic bar chart eliminates the clutter and the distortion caused by the 3D effect, so the graph is easier to read and understand.
Data Source: Our World in Data.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
34 � PRINCIPLES OF DATA VISUALIZATION
1. REMOVE LEGENDS WHEN POSSIBLE AND LABEL DATA DIRECTLY
Let’s start with the easiest type of annotation: Removing legends and directly labeling your data. Many software tool defaults create a data legend and place it around the chart, discon- nected from the data. This forces more work upon your reader to connect each line or bar to its label. A better approach is to directly label your data series.
Take the line chart of average schooling for fifty countries from earlier. Rather than the default approach of putting a legend somewhere around the chart, as in the graph on the left, in the version on the right, I directly label the lines at the right end of the graph.
Help your reader more easily find the labels for the data values by placing the labels directly on the chart.
0
2
4
6
8
10
12
14
16
1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Average years of schooling has increased around the world
(Number of years)
Source: Our World in Data
Germany United States Spain Mexico China Nepal
0
2
4
6
8
10
12
14
16
Average years of schooling has increased around the world
(Number of years)
China
United States
Nepal
Mexico
Germany
Spain
1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Source: Our World in Data
In graphs that have fewer lines, we might also be able to place the labels directly on the graph. In these cases, I try to align the labels instead of placing them in random positions. Notice how in the graph on the left of the next page your eye needs to jump around to find each label. And because we might start reading the graph with the title, the proximity of the label for the United States could give that series greater emphasis. In the version on the right, the labels are aligned along a single vertical line, making it easier to read the entire visual.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 35
We can take a similar approach to labeling this bar chart of average schooling in Germany and the United States. With just two countries, instead of a legend disconnected from the data, what if we added the labels inside the bars or used color in the title of the graph itself to link the title to the content of the graph?
By integrating the text and the data, we’re doing a better job of considering the reader’s needs. Do they need to see every line equally, or will including all the lines clutter the graph? Is it important to label every point in the scatterplot, or will highlighting just a few points suffice? How can we integrate labels and chart elements to help the reader understand the content quickly and easily?
Align the labels and match the colors with the data as in the graph on the right rather than placing them in random positions.
1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Average years of schooling has increased around the world
(Number of years)
Source: Our World in Data
China
United States
Nepal
Mexico
Germany
Spain
0
2
4
6
8
10
12
14
16
1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
Average years of schooling has increased around the world (Number of years)
Source: Our World in Data
China
United States
Nepal
Mexico
Germany
Spain
0
2
4
6
8
10
12
14
16
These are just two examples of how to integrate labels into the graph. Data Source: Our World in Data.
Average years of schooling in Germany and the United States (Number of years)
10.0
13.7 14.1 12.7 12.9 13.4
1997 2007 2017
13.4
1997 2007 2017
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
36 � PRINCIPLES OF DATA VISUALIZATION
2. WRITE THE TITLE LIKE A NEWSPAPER HEADLINE
Most titles are neutral descriptions of the data, as in “Figure 1. Labor Force Participation Rate, Men and Women, 1950–2016.” But better titles capture the takeaway of the chart, telling the reader what conclusions can be drawn from the data. I call these “active titles” or “headline titles.” In my book on presentations, I follow the advice of author Carmine Gallo and urge presenters to use “Twitter-like headlines” in their slides. These are concise, active phrases that make it easy to understand what the slide—or chart—is aiming toward.
Notice the inconsistency between the order of the lines and the legend in the chart on the left. The redesigned version removes unnecessary clutter and directly labels the lines.
Source: Social Security Advisory Board, February, 2012.
DI and SSI allowance rates have generally moved in tandem over the past 25 years (Percent)
80%
70%
60%
50%
40%
30%
20%
10%
0%
1986 1988
1990 1992
1994 1996
1998 2000
2002 2004
2006 2008
2010
Removing the legend isn’t always possible. A bar chart with several categories or a map with different colors requires a legend, because directly labeling the chart will add too much clutter to the visual. In these cases, at least keep the order of the legend consistent with the order of the data. Notice the inconsistency between the order of the lines and the legend in this line chart from the Social Security Advisory Board. Not only do we need to jump back and forth between the lines and labels, there is an extra task of figuring out the order of the two. A redesigned version removes much of those unnecessary data markers and extra gridlines, and integrates the legend onto the chart by adding labels directly next to the lines.
We won’t be able to remove the legend on every single graph we create, but we should strive to link the data and the labels as best we can, and that starts with labeling the data series on our charts.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 37
Too often, we attach a title to the chart that describes the data instead of the point or argu- ment we want to make.
While “Labor Force Participation, Men and Women, 1950–2016” is certainly a correct and accurate description of the data in this graph from the Pew Research Center, it does not describe what the reader should learn about the labor participation rate among men and women between 1950 and 2016. The more active title that Pew uses instead—“Labor force participation rate has risen for women, fallen for men”—tells the reader exactly what they should take away from the graph.
Do people even read titles? A 2015 study from researchers at Harvard University showed that they do: “Titles and text attract people’s attention, are dwelled upon during encoding,
The active title in this chart from the Pew Research Center tells you exactly what you are supposed to learn from it.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
38 � PRINCIPLES OF DATA VISUALIZATION
and . . . contribute to recognition and recall.” If it’s indeed the case that people read titles (and text more generally), then we should treat a chart’s title as carefully as the chart itself.
But can so-called “active” titles make us seem biased or partisan? If we use active titles only to faithfully represent the results and showcase the message of the graph, then no. I’ve worked with many people who have debated my inclination for active titles by arguing that such titles will make their work appear biased. In most of those cases, I can look at the text around their chart and see a single argument for what’s being shown in the graph and how to interpret the data. Their argument is right next to the graphic, but, like the legend we saw earlier, it’s disconnected from it.
Active titles don’t make us biased, but descriptive titles do waste an opportunity to make a clear, compelling case. Of course, short, active titles aren’t always possible—you may be making more than one point or your sole goal is to simply describe the data. Generally speaking, however, integrating your graphs as part of your argument creates a more cohesive approach to making your argument and telling your story.
In the case above, Pew doesn’t leave it to the reader to search for a point in the graph, but neither are they biasing the results by adding commentary in the title. They are simply foregrounding the takeaway of the visual.
If you are having trouble coming up with a concise, active title, that may be a sign that your chart doesn’t have a concise takeaway or—and maybe this is more common—you haven’t thought through what you want the graph to communicate.
3. ADD EXPLAINERS
Once the chart is made and the title is settled, ask yourself, Would this chart benefit from more text?
Sometimes data sets have peaks or valleys, outliers or variations that bear explanation. Adding detail in graphs can push your argument, highlight points of interest, or (in cases of nonstandard graphs) even explain how to read it.
Take this line chart of the popularity of the name “Neil” in the United States created by, yes, Neil Richards, a data visualization consultant in the United Kingdom. Anyone could make the simple line chart on the left—it’s only one data series—but with only a quick glance the reader might immediately ask some obvious questions: Why did the decline stop in the late-1960s? Why did the line spike upwards a few years later in the 1970s? And what halted the decline in the early 2000s?
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 39
Now look at this second version of the chart with short explainers. The late-1960s spike might be attributed to Neil Armstrong landing on the moon, followed by the popularity of musicians like Neil Young, Neil Sedaka, and Neil Diamond in the 1970s. The flattening of the trend in the mid-2000s could be attributed to “modern Neils” like Neil DeGrasse Tyson. These annotations are not complicated and don’t require complex programming or design techniques—they are often just interesting points in the data thoughtfully added with short sections of text.
Annotation allows readers—especially those who may have less experience with data visualization —to grasp and understand the content quickly. The bubble chart from the Los Angeles Times on the next page is a great illustration of how to do so. The change in the violent crime rate is plotted along the horizontal axis and the change in the property crime rate is plotted along the vertical axis for about thirty-five cities in California. The average LA Times reader is probably not a bubble-plot expert, so the authors have added annotations to help readers navigate the format of the graph and its content.
Notice the use of color and annotation to help the reader understand this graph. The top-right quadrant is shaded red with the word “Worse” in large, red letters. The bottom- left quadrant is shaded blue with the word “Better” in large, blue letters. Immediately, you
Short explainers in this graph on the right from Neil Richards explain some of the basic features of the data.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
40 � PRINCIPLES OF DATA VISUALIZATION
understand that the cities in the top-right have worsened and the cities in the bottom-left have improved. Short, bold headlines (“Reduced violent and property crime rate” in the bottom-left quadrant) explain the substance of the changes. Then, a short sentence high- lights a city or two and what has transpired over the past year. This graph does an expert job of explaining how to read it and how to understand the content in it.
This graph from the Los Angeles Times is one of my favorite examples of how to use anno- tation to explain how to read a graph and its content.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 4 1
In a 2016 interview, John Burn-Murdoch, an interactive data journalist at the Financial Times said, “The annotation layer is where the ‘journalism’ really comes into ‘visual journal- ism.’ Making a graphic is the equivalent to interviewing your source. But it’s then your job to actually pick out . . . the bits the reader should know about.” Not everyone is a journalist, but everyone can find ways to help our readers clearly see what’s important and what we want them to learn.
GUIDELINE 4. AVOID THE SPAGHETTI CHART
It’s obvious when a certain graph contains too much information—line charts that look like spaghetti, maps with dozens of colors and icons, or bar after bar after bar littering a chart. Sometimes we face the challenge of including lots of data in a single graph but we don’t need to try to pack everything into a single graph.
Two examples of the small multiples approach. The graph on the left, from Zeit Online, shows the average temperature in Germany over the past 140 years. The graph on the
right, from the Centers of Disease Control and Prevention, shows how facial hair can affect the fit of respirators. The Gestalt principle of connection helps us track the changes from
one image to the next in both graphics.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
42 � PRINCIPLES OF DATA VISUALIZATION
One way to address the packed, single chart is to break it into smaller parts. Known as grid charts or panel charts (also called facets, trellis charts, or, most commonly, small mul- tiples) these are smaller charts that use the same scale, axes, and scope but spread the data across multiple visuals. In other words, instead of putting all of the data on one graph, create multiple, smaller versions with variations on the basic data.
The small multiples approach isn’t a new or revolutionary approach to communicating data. In 1878, photographer Eadweard Muybridge (see page 44) was tasked with deter- mining whether a horse becomes fully airborne when it gallops. Muybridge developed a technique to take a sequence of fast-action photographs (what we now call stop-motion) of a horse at gallop. His photos proved that horses do indeed leave the ground entirely. The sequence of images, which also gives a sense of motion and animation, is an early example of small multiples.
The small multiples approach has at least three advantages. First, once the reader understands how to read one chart, they know how to read all the charts. Second, you can display lots of information without confusing your reader. Third, small mul- tiples let readers make comparisons across multiple variables. This example from the Guardian shows voting results in 2016 for the Brexit resolution in the United Kingdom across six different demographic variables. The horizontal axes stay unchanged and it’s easy to see the direction of the relationships for each demographic measure.
But there are pitfalls to the small multiples approach that will muddy the visual if not avoided. First, the charts should be arranged in a logical order. Don’t make your reader navigate around the page—use an intuitive arrangement based on something like geography or alphabetical order.
Second, the graphs should share the same layout, size, font, and color. Remember, we are breaking up one chart into many, so it should look like one chart replicated multiple times. The vertical and horizontal axes may change, but you wouldn’t want to have one chart where blue dots represent “no” and another where they represent “yes.” Third, small multiples should be relatively easy to read. You are not necessarily asking your reader to zoom in and uncover all the specific details in all of the graphs—the purpose is to give them a view of the overall patterns. The graphs are intended to be small, so including annotations and labels or repeating long axis labels and data markers features can over- whelm the reader.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
Small multiple scatterplots from the Guardian shows the relationship between voting choice and six demographic variables. Notice how the Gestalt principle of similarity lets us
easily see the two clusters of circles within each scatterplot.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
44 � PRINCIPLES OF DATA VISUALIZATION
GUIDELINE 5. START WITH GRAY
I end this section with a practical technique that I think can be an easy step to creating clear, comprehensible visualizations: Start with gray. Whenever you make a graph, start with all- gray data elements. By doing so, you force yourself to be purposeful and strategic in your use of color, labels, and other elements.
Consider a simpler version of the average schooling chart from earlier, this time with only ten countries as shown on the next page. With color and labels (top-left), I could put this graph in my report or handout, and with a little work (and a more active title), my reader could figure out which labels correspond to which lines. But if I make all the lines gray (top-right), the reader can’t accomplish that same task because it’s impossible to figure out which country is represented by which line.
Photographer Eadweard Muybridge used the small-multiples approach back in 1878 to determine whether a horse becomes fully airborne when it gallops.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 45
Now I can be purposeful about what I want to do with this graph. I could add color and even vary the thickness of the lines to better highlight only, say, the two countries I want to emphasize. (Leaving all the labels in the version on the bottom-left is less useful than label- ing the lines directly as in the version on the right.) Starting with gray forces us to deliber- ately choose what elements to put into the foreground.
0
2
4
6
8
10
12
14
16
1997 1999 2001 2003 2005 2007 2009 2011 2013 2015 2017
United States Germany China India Indonesia Brazil Pakistan Philippines Russia Japan
Source: Our World in Data
Average years of schooling has increased around the world (Number of years)
Starting your graphs with all-gray data elements forces you to make purposeful, strategic decisions about where you want to direct
your reader’s attention.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
DATA TYPES The bedrock of any data visualiza on is the data. Without data and a good understand- ing of what our data is, how it was collected, and what it tells us, we are just pain ng pictures. This book is not the venue for a thorough review of data types and sta s cal methods, but a short primer can help us organize our data types just as we organize our graph types.
There are two major groupings of data types: uan ta ve and ualita ve. Quan - ta ve data can be measured with numbers, for example, distance, dollars, speed, and
me. Qualita ve data is non-numerical informa on, usually descrip ve text like “yes or no,” “sa s ed or dissa s ed,” or longer uotes or passages from interviews and surveys.
We can further break down each major data type into subcategories. On the ualita ve side, we have nominal and ordinal scales. Nominal scales are used to label
variables and don’t have an order or uan ta ve value. In a data set of animal types, the order of lion, ger, and bear has no meaning (aside from the song, of course). In ordinal scales, order does ma er, but the exact size in comparison between val- ues is unknown. Consider a survey that asks people to select between 1. Strongly Agree; 2. Agree; 3. Disagree; and 4. Strongly Disagree. These choices can be ordered, but the di erence between 1 and 2 is not necessarily the same as the di erence between 3 and 4.
On the uan ta ve side, data can be either discrete or con nuous. Discrete data are whole numbers (integers) that cannot be subdivided. Despite na onal averages, no one has exactly 2.3 children. Con nuous data are numbers that can be broken down into smaller units, like weight, height, and temperature.
Con nuous data can be further broken down into two major scales: interval and ra o. The di erence is what we can and cannot calculate. With interval scales, we know both the order and the exact di erences, but they do not have a true zero value. This means we can add and subtract data measured in interval scales, but we can’t mul ply or divide. A classic example is temperature in degrees Farenheit: The di erence between 10 and 20 degrees is the same as between 70 and 80 degrees, but we can’t say that 20 degrees is twice as hot as 10 degrees, because 0 degrees is an actual value, not absolute zero.
a o scales have all of the characteris cs of all the other scales plus they have an absolute zero, which means we can do all of our mathema cal calcula ons.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 47
Weight is a good example of a ra o scale—a person who weighs 200 pounds is twice as heavy as someone who weighs 100 pounds, and 0 pounds is the absence of weight.
DATA EQUALITY & RESPONSIBILITY
These guidelines lay out the basic approaches to effectively visualizing our data. While this is not a book about data analysis—how and where to get data, how to analyze underly- ing statistical properties, and develop statistical models—whenever we work with data it is important to recognize that visual content can have a large influence on how people use data and make decisions. As data communicators, it is therefore our responsibility to treat our work and our data as carefully and objectively as possible. It is also our responsibility to recognize where our data may suffer from underlying bias or error, or even implicit bias that data creators may themselves not even be aware of.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
48 � PRINCIPLES OF DATA VISUALIZATION
There are many ways in which the data we use may be biased or not representative. In their book, Data Feminism, Catherine D’Ignazio and Lauren Klein describe how standard practices in data science reinforce existing power inequalities. They explore how data has been used for both good and evil—to expose injustice and improve health and policy out- comes, for example, but also to surveil and discriminate. By asking who is producing the data and for whom it is being produced, we can be better stewards of our own data and our own visualizations.
Many fields are squarely built on a model of the world in which men are the only—or maybe just the most important—participants. In Invisible Women, author Caroline Criado Perez reveals the hidden places where inequality in even basic data resides. There are straightforward examples, like how the average smartphone is 5.5 inches long—too big for most women’s hands and pants pockets. Or how the average temperature in many office buildings is five degrees too cold for women because the formula to determine the ideal temperature was developed in the 1960s based on the metabolic resting rate of a forty- year-old, 150-pound man. There are more insipient examples as well, like how women in Britain are 50 percent more likely to be misdiagnosed following a heart attack, or how car crash test dummies are based on the male body, so even though men are more likely to get into car accidents, women involved in collisions are almost 50 percent more likely to be seriously injured.
In a similar vein, the era of big data, machine learning, and artificial intelligence use more and more unseen algorithms and statistical techniques. We often know little about the data that feed these algorithms and how the models themselves may perpetuate inequality. Mathematician Cathy O’Neil explores this in her book Weapons of Math Destruction, from teacher quality, creditworthiness, and recidivism risk, algorithms can develop and reinforce discriminatory models of public policy.
When it comes to data visualization specifically, we must be mindful of the under- lying biases and inequality in how we present our results. As just one example of how data and visualizations have been used to discriminate, consider this map of Rich- mond, Virginia, produced in 1937 by the Home Owners’ Loan Corporation (HOLC), a federal agency tasked to appraise home values and neighborhoods across the United States. As Richard Rothstein writes in his book, The Color of Law, “The HOLC created color-coded maps of every metropolitan area in the nation, with the safest neighbor- hoods colored green and the riskiest colored red. A neighborhood earned a red color if African Americans lived in it, even if it was a solid middle-class neighborhood of
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
FIVE GUIDELINES FOR BETTER DATA VISUALIZATIONS � 49
single-family homes.” Systematic discrimination is and can be generated by how we use and misuse our data.
Finally, in addition to cultural differences that might arise from, say, using certain colors in different cultures, we should also be mindful of the language, shapes, and images in our visuals. Are we using language and images that are inclusive? When do we need to provide historical and social context for problems people are facing? As with developments in acces- sibility, diversity, and inclusion (see Chapter 12), these are all challenges with which the data visualization field is always wrestling.
This redlining map of Richmond, Virginia, demonstrates how data and data visualiza on can be wielded to further systema c discrimina on. Informa on Studies scholar Sa ya Umoja Nobel
argues that modern internet search engines and other algorithms are enac ng new ways of dis- crimina on and racial pro ling, crea ng a modern form of “technological redlining.”
Source: Na onal Archives.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
DATA VISUALIZATION LESSONS FROM THE CORONAVIRUS PANDEMIC
The manuscript for this book was delivered to the publisher in March 2020, just as the coronavirus (COVID-19) pandemic was traveling around the world. As it spread, it induced massive poli cal, economic, and societal changes, and it brought new terms into our lexicon, like “ a ening the curve.” In late February 2020 The Economist published a version of the graphic by data journalist Rosamund Pearce, based on original work by the Centers for Disease Control and Preven on. Graphics like this built awareness and facilitated ac on, most notably on the rela vely new concept of “social distancing”.
But for every graph that informed and educated, there were many others that mis- represented data or spread misinforma on. One pie chart, for example, inexplicably summed contagion rates among eleven separate diseases to 100 percent and added a separate note that the “numbers for COVID-19 remain a rough es mate because
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
the long incuba on period means we s ll have no idea how many people have been infected.” This is not responsible data visualiza on.
The unprecedented spread of the coronavirus gave us an opportunity to use real- me data that could be used to be er understand the virus and its spread. But one reason why many graphs and charts around COVID-19 are problema c is because too many of us assume we have ade uate knowledge in a par cular subject area. Public health professionals, epidemiologists, and physicians have the training, insight, and experience with the health care system and modeling disease transmis- sion to provide useful data and informa on. For the rest of us, without exper se in these areas, our visualiza on work—even as well-inten oned as it might be—can make things worse.
We o en create—or are asked to create—visualiza ons in subject areas in which we are not experts. Some mes this is an opportunity to explore di erent visualiza on forms and func ons and try new tools. Other mes, though, we may be out of our depth. We may not fully understand our data. Even if we have read the data dic onary or considered the data collec on methods, we may not know enough about how the data were modeled or simulated or the reliability of their collec on methods.
Under ordinary circumstances, visualiza on exercises might consider issues such as unemployment rates or housing op ons or the distribu on of wealth and not life- threatening events like a viral pandemic. In these cases, we must be especially aware of how our work might be misunderstood and how it may change the thinking or behavior of our readers.
The converse of the above is also true. An epidemiologist may know a lot about modeling disease spread, but he or she may not understand how best to visualize that modeling, explain jargon, and annotate important data points. Here, it is incumbent upon the scien st to reach out to data visualiza on experts and graphic designers to ensure their visualiza on work is accessible to readers.
There is a be er way forward. Instead of thinking our limited knowledge is su - cient to weigh in on every topic and every dataset, we should strive to collaborate. In the case of COVID-19, not knowing enough may lead to deadly outcomes. If we think of ourselves as journalists and seek out domain speci c experts, we can work to build teams, groups, and organiza ons that can deliver be er data, be er visualiza ons, and be er decisions.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
52 � PRINCIPLES OF DATA VISUALIZATION
NEXT STEPS
Now armed with basic guidelines and rules of perception, you are almost ready to start adding more graphs to your data visualization toolbox. But there’s one more thing you should consider before you start encoding your data with bars, lines, and dots: the purpose of your graph.
In what format do you need to present your data to your reader or user? Do they need a static graph where you present your argument or will an interactive visualization help them explore the data and come to more and deeper conclusions? In the next chapter, we discuss the different forms and functions for visualizations and then turn to the many ways we can visualize our data.
Schwabish, J. (2021). Better data visualizations : A guide for scholars, researchers, and wonks. Columbia University Press. Created from rutgers-ebooks on 2022-09-14 14:20:29.
C op
yr ig
ht ©
2 02
1. C
ol um
bi a
U ni
ve rs
ity P
re ss
. A ll
rig ht
s re
se rv
ed .
