This opening chapter provides an introduction to the subject of data visualization and the intention behind this book.
We start things off with some context about the subject. This will briefly explain why there is such an appetite for data visualization and why it is so relevant in the modern age against the backdrop of enhanced technology, increasing capture and availability of data, and the desire for innovative forms of communication.
After this introduction, we then look at the theoretical basis of data visualization, specifically the importance of understanding visual perception. To help establish a term of reference for the rest of the book, we'll then consider a proposed definition for this subject.
Next, we introduce the data visualization methodology, a recommended approach that forms the core of this book, and discuss its role in supporting an effective and efficient design process.
Finally, we consider some of the fundamental data visualization design objectives. These provide a useful framework for evaluating the suitability of the choices we make along the journey towards an accomplished design solution.
The following is a quotation from Hal Varian, Google's chief economist (http://www.mckinseyquarterly.com/Hal_Varian_on_how_the_Web_challenges_managers_2286):
The ability to take data—to be able to understand it, to process it, to extract value from it, to visualize it, to communicate it—that's going to be a hugely important skill in the next decades.
Data visualization is not new; the visual communication of data has been around in various forms for hundreds and arguably thousands of years. Popular methods that still dominate the boardrooms of corporations across the land—the line, bar, and pie charts—originate from the eighteenth century.
What is new is the contemporary appetite for and interest in a subject that has emerged from the fringes and into mainstream consciousness over the past decade.
Catalyzed by powerful new technological capabilities as well as a cultural shift towards greater transparency and accessibility of data, the field has experienced a rapid growth in enthusiastic participation.
Where once the practice of this discipline would have been the preserve of specialist statisticians, engineers, and academics, the globalized field that exists today is a very active, informed, inclusive, and innovative community of practitioners pushing the craft forward in fascinating directions. The following image shows a screenshot of the OECD 'Better Life Index', comparing well-being across different countries. This is just one recent example of an extremely successful visual tool emerging from this field.
Image from "OECD Better Life Index" (http://oecdbetterlifeindex.org), created by Moritz Stefaner (htpp://moritz.stefaner.eu) in collaboration with Raureif GmbH (http://raureif.net)
Data visualization is the multi-talented, boundary-spanning trendy kid that has seen many esteemed people over the past few years, such as Hal Varian, forecasting this as one of the next big things.
Anyone considering data visualization as a passing fad or just another vacuous buzzword is short-sighted; the need to make sense of and communicate data to others will surely only increase in relevance. However, as it evolves from the next big thing to the current big thing, the field is at an important stage of its diffusion and maturity. Expectancy has been heightened and it does have a certain amount to prove; something concrete to deliver beyond just experimentation and constant innovation.
It is an especially important discipline with a strong role to play in this modern age. To help frame this, let's first look at the data side of things.
Take a minute to imagine your data footprint over the past 24 hours; that is, the activities you have been involved in or the actions you have taken that will have resulted in data being created and captured.
You've probably included things such as buying something in a shop, switching on a light, putting some fuel in your car, or watching a TV program: the list can go on and on.
Almost everything we do involves a digital consequence; our lives are constantly being recorded and quantified. That sounds a bit scary and probably a little too close for comfort to Orwell's dystopian vision. Yet, for those of us with an analytical curiosity, the amount of data being recorded creates exciting new opportunities to make and share discoveries about the world we live in.
Thanks to incredible advancements and pervasive access to powerful technologies we are capturing, creating, and mobilizing unbelievable amounts of data at an unbelievable rate. Indeed, such is the exponential growth in digital information, in the last two years alone, humanity has created more data than had ever previously been amassed (http://www.emc.com/leadership/programs/digital-universe.htm).
Data is now rightly seen as an invaluable asset, something that can genuinely help change the world for the better or potentially create a competitive goldmine, depending on your perspective. "Data is the new oil", first voiced in 2006 and attributed to Clive Humby of Dunnhumby, is a term gaining traction today. Corporations, government bodies, and scientists, to name but a few, are realizing the challenges and, moreover, opportunities that exist with effective utilization of the extraordinary volumes, large varieties, and great velocity of data they govern.
However, to unlock the potential contained within these deep wells of ones and zeros requires the application of techniques to explore and convey the key insights.
Flipping to the opposite side of the data experience, we also identify ourselves as consumers of data. As you would expect, given the volume of captured data, never before in our history have we been faced with the prospect of having to process and digest so much.
Through newspapers, magazines, advertising, the Web, text messaging, social media, and e-mail, our eyes and brains are being relentlessly bombarded by information. In a typical day, it is said we can expect to consume about 100,000 words (http://hmi.ucsd.edu/howmuchinfo_research_report_consum.php), which is an astonishing quantity of signals for us to have to make sense of.
Unquestionably, a majority of this visual onslaught flies past us without consequence. We see much of it as noise and we zone out as a way of coping with the overload and saturation of things to think and care about.
What this shows is the necessity to be more effective and efficient in how data is communicated. It needs to be portrayed in ways that help to get our messages across in both an engaging and informative way.
If data is the oil, then data visualization is the engine that facilitates its true value and that is why it is such a relevant discipline for exploiting our digital age.
One of the most compelling arguments for the value of data visualization is expressed in this quote from John W Tukey (Exploratory Data Analysis).
The greatest value of a picture is when it forces us to notice what we never expected to see.
Through visualization, we are seeking to portray data in ways that allow us to see it in a new light, to visually observe patterns, exceptions, and the possible stories that sit behind its raw state. This is about considering visualization as a tool for discovery.
A well known demonstration that supports this notion was developed by noted statistician Francis Anscombe (incidentally, brother-in-law to Tukey) in the 1970s. He compiled an experiment involving four sets of data, each exhibiting almost identical statistical properties including mean, variance, and correlation. This was known as "Anscombe's quartet".
Sample data sets recreated from Anscombe, Francis J. (1973) Graphs in statistical analysis. American Statistician, 27, 17–21
Ask yourself, what can you see in these sets of data? Do any patterns or trends jump out? Perhaps the sequence of eights in the fourth set? Otherwise there's nothing much of interest evident.
So what if we now visualize this data, what can we see then?
Image published under the terms of "Creative Commons Attribution-Share Alike", source: http://commons.wikimedia.org/wiki/File:Anscombe%27s_quartet_3.svg
Through the previous graphical display, we can immediately see the prominent patterns created by the relationships between the X and Y values across the four sets of data as follows:
the general tendency about a trend line in X1, Y1
the curvature pattern of X2, Y2
the strong linear pattern with single outlier in X3, Y3
the similarly strong linear pattern with an outlier for X4, Y4
The intention and value of Anscombe's experiment was to demonstrate the importance of presenting data graphically. Rather than just describing a dataset based on a selection of some of its key statistical properties alone, to make proper sense of data, and avoid forming false conclusions we need to also employ visualization techniques.
It is much easier to discover and confirm the presence (or even absence) of patterns, relationships, and physical characteristics (such as outliers) through a visual display, reinforcing the essence of Tukey's quote about the value of pictures.
Data visualization is about a discovery process, enabling the reader to move from just looking at data to actually seeing it. This is a subtle but important distinction.
Data visualization is not easy. Let's make that clear from the start. It should be genuinely viewed as a craft. It is a unique convergence of many different skills and requires a great deal of practice and experience, which clearly demands time and patience.
Above all, it requires a deep and broad knowledge across several traditionally discrete subjects, including cognitive science, statistics, graphic design, cartography, and computer science.
This multi-disciplinary recipe unquestionably makes it a challenging subject to master but equally provides an exciting proposition for many. This is evidenced by the field's popular participation, drawing people from many diverse backgrounds.
If we look at this subject convergence at a more summary level, data visualization could be described as an intersection of art and science. This combination of creative and scientific perspectives represents a delicate mixture. Achieving an appropriate balance between these contrasting ingredients is one of the fundamental factors that will determine the success or failure of a designer's work.
The art side of the field refers to the scope for unleashing design flair and encouraging innovation, where you strive to design communications that appeal on an aesthetic level and then survive in the mind on an emotional one. Some of the modern-day creative output from across the field is extraordinary and we'll see a few examples of this throughout the chapters ahead.
The science behind visualization comes in many shapes. I've already mentioned the presence of computer science, mathematics, and statistics, but one of the key foundations of the subject comes through an understanding of cognitive science and in particular the study of visual perception. This concerns how the functions of the eye and the brain work together to process information as visual signals.
One of the other most influential founding studies about visual perception emerged from the Gestalt School of Psychology in the early 1900s, specifically in the shape of the Laws of Perceptual Organization (http://www.interaction-design.org/encyclopedia/data_visualization_for_human_perception.html).
These laws provide an organized understanding about the different ways our eyes and brain inherently and automatically form a global sense of patterns based on the arrangement and physical attributes of individual elements.
Here, we can see two visual examples of Gestalt Laws.
On the left-hand side is a demonstration of the "Law of Similarity". This shows a series of rows with differently shaded circles. When we see this our visual processes instantly determine that the similarly shaded circles are related and part of a group that is separate and different to the non-shaded rows. We don't need to think about this and wait to form such a conclusion; it is a preattentive reaction.
Images republished from the freely licensed media file repository Wikimedia Commons, source: http://en.wikipedia.org/wiki/File:Gestalt_similarity.svg and http://en.wikipedia.org/wiki/File:Gestalt_proximity.svg
On the right-hand side is a demonstration of the "Law of Proximity". The arrangement of closely packed-together pairs of columns means we assume these to be related and distinct from the other pairings. We don't really view this display as six columns, rather we view them as three clusters or sets.
At the root of visual perception knowledge is the understanding that our visual functions are extremely fast and efficient processes whereas our cognitive processes, the act of thinking, is much slower and less efficient. How we exploit these attributes in visualization has a significant impact on how effectively the design will aid interpretation.
Consider the following examples, both portraying analysis of the placement of penalties taken by soccer players.
When we look at the first image, the clarity of the display allows us to instantly identify the football symbols, their position, and their classifying color. We don't need to think about how to interpret it, we just do. Our thoughts, instead, are focused on the consequence of this information: what do these patterns and insights mean to us? If you're a goalkeeper, you'll be learning that, in general, the penalty taker tends to place their shots to the right of the goal.
Image republished under the terms of "fair use", source: http://www.facebook.com/castrolfootball
By contrast, this second display's attempt to portray the same type of data presentation causes significant visual clutter and confusion. Rather than using a simple and relatively blank image like the previous one, this display includes strong colors and imagery in the background. The result is that our eyes and brain have to work much harder to spot the footballs and their colors because the data layer has to compete for attention with the background imagery. We are therefore unable to rely on the capabilities of our preattentive visual perception (determined by the Law of Similarity) because we cannot easily perceive the shapes and their attributes representing the data. This delays our interpretative processes considerably and undermines the effectiveness and efficiency of the communication exchange.
Image republished under terms of "fair use", source: http://www.mirror.co.uk/sport/football/euro-2012-where-italy-will-place-their-penalties-907506
This is just a single, simple example but it does reveal the significance of understanding and obeying visual perception laws when portraying our data.
When we design a visualization, we need to take advantage of the strengths of the visual function and avoid the disadvantages of the cognitive functions. We need to minimize the amount of thinking or "working out" that goes into reading and interpreting data and simply let the eyes do their efficient and effective job.
Through the pioneering studies and development of theories acquired and refined over many years by the Gestalt School of Psychology as well as influential academics and theorists like Jacques Bertin, Francis Anscombe, John W Tukey, Jock McKinlay, and William Cleveland, we now have a greater understanding of how to achieve effective and efficient visualization design.
There is still a great amount of empirical evidence to gather, studies to conduct, and firm answers to unearth, but the wealth of knowledge available to us is a significant help to remove an undue amount of instinct in our design work.
It is important now to consider a definition of data visualization. To do this, we first need to consider the main agents involved in the exchange of information; namely, the messenger, the receiver, and the message. The relationship between these three is clearly very important, as this illustration explains:
On one side we have a messenger looking to impart results, analysis, and stories. This is the designer. On the other side, you have the receiver of the message. These are the readers or the users of your visualization. The message in the middle is the channel of communication. In our case this is the data visualization; a chart, an online interactive, a touch screen installation, or maybe an infographic in a newspaper. This is the form through which we communicate to the receiver.
The task for you as the designer is to put yourself in the shoes of the reader. Try to imagine, anticipate, and determine what they are going to be seeking from your message. What stories are they seeking? Is it just to learn something new or are they looking for persuasion, something with more emotional impact? This type of appreciation is what fundamentally shapes the best practices in visualization design: considering and respecting the needs of the reader.
The important point is this: to ensure that our message is conveyed in the most effective and efficient form, one that will serve the requirements of the receiver, we need to make sure we design (or "encode") our message in a way that actively exploits how the receiver will most effectively interpret (or "decode") the message through their visual perception capabilities.
From this illustration we can form the following definition to clarify, at this early stage, what we mean by data visualization:
The representation and presentation of data that exploits our visual perception abilities in order to amplify cognition.
Let's take a closer look at the key elements of this definition to clarify its meaning; these are as follows:
The representation of data is the way you decide to depict data through a choice of physical forms. Whether it is via a line, a bar, a circle, or any other visual variable, you are taking data as the raw material and creating a representation to best portray its attributes. We will cover this aspect of design much more in Chapter 4, Conceiving and Reasoning Visualization Design Options and Chapter 5, Taxonomy of Data Visualization Methods.
The presentation of data goes beyond the representation of data and concerns how you integrate your data representation into the overall communicated work, including the choice of colors, annotations, and interactive features. Similarly, this will be covered in depth in Chapter 4, Conceiving and Reasoning Visualization Design Options.
Exploiting our visual perception abilities relates to the scientific understanding of how our eyes and brains process information most effectively, as we've just discussed. This is about harnessing our abilities with spatial reasoning, pattern recognition, and big-picture thinking.
Amplify cognition is about maximizing how efficiently and effectively we are able to process the information into thoughts, insights, and knowledge. Ultimately, the objective of data visualization should be to make a reader or users feel like they have become better informed about a subject.
The definition that I've put forward here is not dissimilar to the many others articulated by authors, academics, and designers down the years. It is not intended to offer a paradigm shift in our understanding of what this is all about. Rather, it represents a personal perspective of the discipline influenced by many years of experience teaching, practicing, and constantly studying the subject.
The fact that data visualization is such a dynamic and evolving field, with this unique conjunction of art and science shaping its practice, means that a single, perfect, and universally-agreed definition is always going to be difficult to construct. However, this proposed definition should at least help you develop an appreciation of the boundaries of data visualization and recognize when something evolves into a different form of creative output.
The following is a quote from Stephen Few from his book Show Me the Numbers:
"The skills required for most effectively displaying information are not intuitive and rely largely on principles that must be learned."
More and more of us are becoming responsible for the analysis, presentation, and interpretation of data. This naturally reflects the explosion in access to data and the value attributed to potential insights that are contained.
As I've already stated, where once this was typically a specialist role, nowadays the responsibility for dealing with data has crept into most professional duties. This has been accelerated by the ubiquitous availability of a range of accessible productivity tools to handle and analyze data.
This means visualization has become both a problem and an opportunity for the masses, which makes the importance and dissemination of effective practice a key imperative.
The quote from Stephen Few will resonate with many of you reading this. If you were to ask yourself "Why do I design visualizations in the way I do?", what would be your answer? Think about any chart or graphic you produce to communicate information to others. How do you design it? What factors do you take into account? Perhaps your response would fall in to one or more of the following:
You have a certain design style based on personal taste
You just play around until something emerges that you instinctively like the look of
You trust software defaults and don't go beyond that in terms of modifying the design
You have limited software capabilities, so you don't know how to modify a design
You just do as the boss tells you—"can you do me some fancy charts?"
For many people, the idea of a conscious data visualization design technique isquite new. The absence of any formal coaching, at almost any level of education, in the techniques of visualization means until you become aware of the subject, you have probably never even thought about your visualization design approach.
Before discovering this subject, my own approach to presenting data was certainly not informed by any training or prior knowledge. I'd never even thought about it. Taste and gut-feel were my guiding principles alongside a perceived need to show off technical competencies in tools like Excel. Indeed, I'd like to take this opportunity to apologize for much of my graphical output between 1995 and 2005 where striking gradients and "impressive" 3D were commonplace. The thing is, as I've just said, I didn't realize there was a better way; it simply wasn't on my radar.
In some respects, the reliance on instinct, playing about with solutions that seem to work fine for us, can suffice for most of our needs. However, these days, you often hear the desire being expressed to move beyond devices like the bar chart and find different creative ways to communicate data.
While it is a perfectly understandable desire, just aiming for something different (or even worse, something "cool") is not a good enough motive in itself.
If we want to optimize the way we approach a data visualization design, whether it be a small, simple chart or a complicated interactive graphic, we need to be better equipped with the necessary knowledge and appreciation of the many design and analytical decisions we need to make.
As suggested previously, instinct and taste have got us so far but to move on to a whole new level of effectiveness, we need to understand the key design concepts and learn about the creative process. This is where the importance of a methodology comes in.
The design methodology described in this book is intended to be portable to any visualization challenge. It presents a sequence of important analytical and design tasks and decisions that need to be handled effectively.
As any fellow student of Operational Research (the "Science of Better") will testify, through planning and preparation, and the development and deployment of strategy, complex problems can be overcome with greater efficiency, effectiveness, and elegance. Data visualization is no different.
Adopting this methodology is about recognizing the key stages, considerations, and tactics that will help you navigate smoothly through your visualization project.
Remember, though, design is rarely a neat, linear process and indeed some of the stages may occasionally switch in sequence and require iteration. It is natural that new factors can emerge at any stage and influence alternative solutions, so it is important to be open-minded and flexible. Things might need to be revisited, decisions reversed, and directions changed. What we are trying to do, where possible, is find the best path through the minefield of design choices.
Some may feel uncomfortable at the prospect of following a process to undertake what is fundamentally an iterative, creative design process. But I would argue everyone should find value from working in a more organized and sequenced way especially if it helps to reduce inefficiency and wasted resource.
The design challenges involved in data visualization are predominantly technology related; the creation and execution of a visualization design will typically require the assistance of a variety of applications and programs. However, the focus of this methodology is intended to be technology-neutral, placing an emphasis on the concepting, reasoning, and decision-making.
The variety, evolution, and generally fragmented nature of software in this field (there is no single tool that can do everything) highlights the extra importance of reasoned decision-making, regardless of the richness and power individual solutions can offer.
Another key point to remark on is to emphasize, if it wasn't already clear, that data visualization is not an exact science. There is rarely, if ever, a single right answer or single best solution. It is much more about using heuristic methods to determine the most satisfactory solutions.
On that note, the content of the methodology intentionally avoids any sense of dogmatic instruction, preferring to focus on guidelines over explicit rules; sometimes an ounce of chaos, a certain license to experiment, a leaning on instinct, and a sense of randomness can spark greater creativity and serendipitous discovery.
The methodology is intended to be adopted flexibly, based on your own judgment and discretion, by simply laying out all the important things you need to take into account and proposing some potential solutions for different scenarios.
Finally, as I stressed with my definition of the subject earlier, I'm not suggesting this is a ground-breaking new take on the creative process. It is merely a personal interpretation based on experience and also exposure to the many brilliant people out there who share their own design narratives. It is, though, consistent with how most established observers of the subject would recommend you undertake this task. Moreover, it is an approach that I fundamentally believe works and it has genuinely helped me improve my own work since I've adopted it more deliberately, allowing me to cut through projects with the efficiency and elegance I've always yearned for.
Before we launch in to the first stages of the methodology in Chapter 2, Setting the Purpose and Identifying Key Factors, it is important to acknowledge a handful of key, overriding design objectives that should provide you with a framework to test your progress and the suitability of your design decisions.
Whereas the methodology will introduce a number of key thoughts and decisions at each stage of the process, these objectives transcend any individual step and highlight the intricate issues you have to handle throughout your process.
The key objectives are as follows:
The following is a quote from Frank Lloyd Wright:
"Form follows function—that has been misunderstood. Form and function should be one, joined in a spiritual union."
The first objective brings us immediately face-to-face with the age-old debate of form versus function or style over substance. As Frank Lloyd Wright proposed, all the way back in 1908, these are aspects of design that should be combined and brought together in harmony, not at the sacrifice of one or the other. There's room and a need for both.
It is a very difficult balancing act to achieve, as I've already alluded to in the discussion about art and science, but our aim should be to hit that sweet-spot where something is aesthetically inviting and functionally effective.
The designer and author Don Norman (http://www.jnd.org/dn.mss/emotion_design.html) talks about how we're more tolerant about things that are attractive and more likely to want them to perform well. Indeed, there is a school of thought that suggests how we think cannot be separated from how we feel.
Norman goes on to describe how well-executed aesthetics can naturally create favorable emotional and mental responses, but emotional affection can also come from the experience of good usability and the accomplishment of insight. Fundamentally, attractive form enhances function and the function portrays beauty through its effect.
Throughout this book, we will see examples of designs that have succeeded in creating elegance in form and in function. The following image is taken from an animated wind map developed by Fernanda Viégas and Martin Wattenberg. It is a beautiful piece of work, exceptionally well designed and executed but it also serves its purpose as a way of informing users about the wind patterns, strength, and directions occurring across the United States. This is form and function in spiritual union:
Image from "Wind Map" (http://hint.fm/wind/) created by Fernanda Viégas and Martin Wattenberg
The general advice, especially for beginners, is to initially focus on securing the functional aspects of your visualization. First, try to achieve the foundation of something that informs—that functions—before exploring the ways of enhancing its form. The simplest analogy would be build the house before decorating it, but I wouldn't want to create too much separation between the two as they are often intrinsically linked. Over time, you will be much more confident and capable of synthesizing the two demands in harmony. We shall discuss this in more depth in Chapter 4, Conceiving and Reasoning Visualization Design Options.
The following is a quote from Amanda Cox (http://vimeo.com/29391942), who works as a graphics editor at the New York Times:
"We're so busy thinking about if we can do things, we forget to consider whether we should."
In many ways, the central idea behind the methodology is encouraging you to determine that everything you do is thoroughly planned, understood, and reasoned.
This particular objective is about recognizing and responding to the scoping information that you will gather at the start of the methodology, to ensure that everything undertaken thereafter serves the purpose of our work and the needs of the audience.
Here, we should consider the idea of deliberate design, which means that the inclusion, exclusion, and execution of every single mark, characteristic, and design feature is done for a reason.
When we reach the stage of designing, concepting, and construction, you should be prepared to challenge everything; the use of a shape, the selection of a color pallet, the position of a label, or the use of an interaction.
In this next example, when displaying a section of a tree-hierarchy work by data illustrator, Stefanie Posavec, every visible property presented is used to communicate data, whether it be the use of color, the arc lengths of the petals, the position and sequence of stems; nothing is redundant and everything is deliberate.
Image from "Literary Organism" (http://itsbeenreal.co.uk/index.php?/wwwords/literary-organism/), created by Stefanie Posavec
It is also important to make sure that any visual property that is included, but does not represent data, such as shading, labels, colors, and axes among other properties, should only be included to aid the process of visual perception, not hinder it.
Furthermore, for interactive and animated visualizations, remember Amanda Cox's quote—"just because you can, doesn't mean you should." Don't succumb to the belief (like I did for many years) of thinking a visualization is a platform solely to showcase your technical competence.
Cluttering visualizations with fancy interactive features is a trap that is easy to fall into and leads to projects that look nice or are impressive technically but fail to serve their intended purpose. Instead, they interfere with the efficiency and effectiveness of the information exchange thus demonstrating a failure to synthesize form and function.
The following is a quote from Edward Tufte (http://adage.com/article/adagestat/edward-tufte-adagestat-q-a/230884/):
"Overload, clutter, and confusion are not attributes of information, they are failures of design."
When you next happen to be in a town or city center, take a look around you and observe how often people are confused by and struggle with the basic operation of correctly opening and entering doors into a store. Notice how the accessibility and function of a door—the simple act of opening and walking through it—is often impaired through a lack of intuitive design.
The method of opening a door should be straightforward, but often the aesthetics of features such as stylish door handles means we pull when we should push and we push when we should pull. This is a flaw in the intuitiveness and logic of the design, a failure in perceived affordance—it doesn't do what it looks like it should do.
This idea is an important concept to translate into visualization. As we have already outlined, we are trying to exploit the inherent spatial reasoning and pattern recognition functions of visual perception. We don't want people to have to spend unnecessary time thinking about how to use or how to read and interpret something.
When you are creating a visualization, you are integrating visual design with a subject matter's data. The former is the window into the latter, and it is the design and execution of this window that creates the accessibility.
But it is important to create a distinction between accessibility and immediacy. The speed with which you are able to read or interpret a visualization should be determined by the complexity of the subject and the purpose of the project, not by the ineffectiveness of design.
Sometimes subjects are fundamentally simple and the portrayal of the data is straightforward and intuitive. This in turn means the reader's task of interpreting the data should be relatively easy.
On other occasions, a data framework might be more complex. Your challenge will be to respect the complexity and avoid simplifying, diluting, or reducing the essence of this subject. This might mean something is not immediately easy to interpret. Some visualizations will require effort to be put in, forcing the reader to undertake a certain amount of experiential practice in order for the eye and mind to essentially become trained in reading the display.
Think of it being like muscle memory, but for the eye and the brain. We are so used to reading bar charts and line charts that they have become entrenched and programmed into our interpretative toolkit. But when we are faced with something new, something different or seemingly complex, its not always immediately clear how we are supposed to handle it.
In the following example, we see a demonstration of what is quite a complex data framework. This is an image of a legend that was used to explain how to read an innovative visualization to portray three separate indicators of a movie's success. On the left-hand side of the image is the aggregate reviews (the higher the value, the better) and on the right-hand side of the image are both the budget and gross takings (the bigger the gap, the better):
Image from "Spotlight on Profitability" (http://www.szucskrisztina.hu), created by KrisztinaSzucs
It is an unusual representation of data, not something as preprogrammed as the bar or line chart, and so it takes a short while to learn how to read and interpret the resulting shapes formed by the movie data shown across piece. This is absolutely legitimate as an effective approach to visualizing this data so long as the efforts that go into learning how to read it eventually leads the user to understand it.
Take another example, which portrays the key events in a couple of soccer matches showing completed passes (green lines), shots (blue triangles), and goals (red dots) as shown in the following image:
Image from "Umbro World Cup Poster" (http://www.mikemake.com/Umbro-s-World-Cup-Poster), created by Michael Deal
Once the reader has mastered the understanding of what each shape and its position means, these displays provide a powerful and rewarding insight in to the key incidents and the general ebb and flow of each game.
In simple terms, so long as you can avoid all the negative characteristics that Edward Tufte mentions at the top of this section, you should succeed in giving people an accessible route in to the data. Make sure that the efforts needed from the reader or user to understand how to use and interpret a visualization are ultimately rewarded with a worthy amount of insight gained.
Visualization ethics relates to the potential deception that can be created, intentionally or otherwise, from an ineffective and inappropriate representation of data. Sometimes it can be through a simple lack of understanding of visual perception.
In the following diagram, we see a 2D pie chart and a 3D version. When the eye interprets a graphic like this, what it is actually doing is perceiving the proportion of visible pixels:
Image from "The Curious Incident of Kevins in Zurich…and other stories" (http://www.researchobservatories.org.uk/EasysiteWeb/getresource.axd?AssetID=38334) by Alan Smith.
On the left-hand side of the diagram, we see a blue segment representing 82 percent and an orange segment representing 18 percent. These are the actual values. However, when we introduce a third dimension on the right—incidentally, a dimension which is purely decorative and has no relationship with data values—our eyes are deceived because we are not capable of easily adjusting our interpretation of the values across this isometric projection. With the introduction of the extra dimension and the visible height of the pie itself, we now perceive 91 percent of the visible area as blue and only 9 percent orange. This is clearly a hugely distorted reading of the values.
Another similar example comes from a Wikipedia fundraising campaign from a few years ago and a progress bar depicting the status of their efforts; as shown in the following screenshot:
Image published under the terms of "Creative Commons Attribution-Share Alike", source: https://donate.wikimedia.org/
As with the pie chart, for a bar chart we perceive the visible pixels as being representative of the values. The label indicates a total of $0.8M USD had been raised (10.7 percent towards target) but if you calculate the actual length of the bar displayed, this occupies 24.6 percent of the overall bar length. Once again, a significant distortion of the truth.
This next example is a demonstration of where aesthetics and style completely hijack a visualization. Here, we have a still showing a 3D bar chart that swooshes impressively onto the screens of those watching soccer on TV in the UK:
But what have we here? There is a yellow Drawn bar representing the value 1 and this appears to be more than half the length of a red Aston Villa bar representing 4. How can that be?
The designers of this visual have chosen to include the category labels within the bar's length, thus completely distorting the values being represented. Now, this is possibly one of the least interesting statistics you'll come across, and I'm assured the world will not stop turning as a result of this graphical misdemeanor, but it should demonstrate the pitfalls of decoration and overly stylized design.
Obeying visualization ethics is clearly an objective for any project, but really it is just about basic, good practice, respect for your readers, and attention to detail.
In this chapter, we have learned about the context of the digital era and the role data visualization can play in helping us make greater sense of the huge volumes of captured data we have access to in today's world.
We have discussed how more and more people are getting involved in activities that require visualization techniques, but the skills required to accomplish this effectively go beyond instinct and require careful learning and practice.
The methodology presented in this book will provide a strategy for designers to develop these techniques through good practice. It will help them navigate through the key decisions that are required throughout the creative process.
Finally, to commence the design thinking, we have learned about some important overriding objectives that should provide a useful assessment of the effectiveness of your visual solution throughout its creation.
In the next chapter, we will commence the data visualization methodology by exploring the first stage of any design challenge: establishing the project's purpose and identifying its inherent key influencing factors.
