🍃 Leaf

Data Visualization

A graphical representation of information that turns raw numbers into meaningful insights. We use it not only to make data easier to understand, but also to expose hidden facts that would be unlikely—or practically impossible—to get noticed otherwise. Due to the human nature, visuals happen to be the most effective way of communication, they are meant to do their job for the reader fast and easy, in a pleasant manner.

A picture is worth a thousand words.

Visualizing data provides a clearer vision of reality. Humans have share their visions through draws on walls and carving stones since thousands of years. Nevertheless, it was William Playfair (1759-1823) who changed how we convey information. He pioneered the field before it even was recognized as such. An engineer and economist who innovated with time series, line/bar/pie charts, and more, to explain non-evident facts and drive better decision making.

Reasons To Visualize, Nature and Modernity

We humans are visual creatures. We discover, learn, and memorize more effectively through images than other medium. It’s easy for us to perceive visual alignment, proportions, size, and patterns, it’s an innate strength. In fact we do that constantly, instantly, and unconsciously all of the time. We are highly specialized for visual processing. Our eyes are the best sensorial organs due to their capabilities, e.g.: scope range; detectable intensities; resolution size; change sensitivity; detail discrimination. Also, our neural networks and the brain prioritize visual stimuli, processing it faster than other senses. Furthermore, the stimuli we remember the most is the visual one, this effect is known as Pictorial Superiority Effect. Due to these inherent traits humans have, our attention and behavior are highly dependant on and reactive to images.

Nevertheless, and paradoxically, blind people can also get benefit from data visualization due to the huge simplification and thoughtful organization of information—if we do a good job.

The more visual the input becomes, the more likely it is to be recognized—and recalled. (John Medina [Biologist], Brain Rules, 2008)

Yet another reason is that modern societies produce and consume more data than ever, and there is a tendency for that to increase through time, with is often problematic for people in general. Critical, heavily reliant on data events occur everyday on mayor fields such as education, health care, finance, science, and engineering. Modern life styles make people struggle when it comes to dealing with data, on a daily basis, because of the high amounts of it and the poor job we do to prioritize and filter the relevant from the non-relevant information. We could call that “data pollution,” and data visualization helps us address that problem.

From a business point of view: the amount of products, services, and their customizability is also higher than ever, so it is the difficulty of choosing among options—for both buyers and sellers. This is why, through the exposure of evident, intuitive trends that lead to accurate predictions, visualizations are a perfect tool for persuasion because, for many, to see is to trust. They convey often complex ideas in a frictionless, friendly manner that aids conviction for the readers.

Consequently, data visualizations:

Building a Visualization

Pre-Design, Goals & Boundaries

Three key aspects must be clarified before we start designing:

  1. Identify the audience and their needs.
  2. Gather the data to interpret it.
  3. Determine the appropriate message.

(Preferentially in that order.)

Identify the Audience and Their Needs

Let’s say that a visualization is an answer, then, in this first step we focus on two points of vital importance: who is asking, and what question(s), which is not necessarily evident. It is critical to know what they want—or would like—to hear as this information is the actual reason for the visual to be built, gathering it is an effectiveness requirement. If a visual is built before a discovery that came out from data and is about something no one asked for, it is presumable that we already know—or think we know—both the “who” and the “what” because otherwise we would not consider it something valuable nor necessary.

Public in general—millions of people—will come from a wide range of backgrounds and roles, and have low to none exigences. On the other hand, technicians or business stakeholders—maybe a couple of people—could have much more demanding requirements and expectations; they’ll be ready and certainly waiting for mid to high levels of complexity. Knowledge about all those aspects leads to an appropriate presentation; to use proper tone and vocabulary; to leverage cognitive loads based on their domain and level of knowledge; to consider potential biases, etc.

On the Visual’s Function

The previous context allows us be more specific about what the (overall) function of the visualization is. Two major categories—with a fundamental difference—can be enumerated:

A visual about “the known,” that which has a non-ambiguous goal; an explanatory presentation made to clarify facts about a given topic we already know and want to understand better. I.e.: it is build to work as a communication and presentation tool.

A visual about “the unknown,” is that aimed to facilitate discovery; an exploratory presentation that aids research for exposing facts, unveiling hidden insights, chain cause-consequence events, validate hypothesis, etc. I.e.: it works as an analysis and examination tool for addressing our uncertainty about something we barely suspected. Naturally, this kind of visual is more complex—and is read by people apt for such complexity as well.

We can use the overall target as a guiding big-picture, however, it’s realistic to consider that any visualization could—and certainly will—have at least, part of both “show the known” and “discover the unknown.”

Gather the Data to Interpret It

-This is the raw matter used to answer.

-That data is often parsed, filtered, grouped, organized, etc., while being interpreted.

-During this treatment is when patterns are identified,

-at this point, is not rare to find unexpected yet interesting insights that—were not part of the initial requirements, but—turn out to be worth including to add context or elaborate an explanation for a given fact.

-Research falls out of this summary’s scope, however, it would be possible—or even recommendable—to collect data on our own from either external data source banks or custom, specific surveys.

-all off this is often called “data mining,” where we the raw matter is processed for a proper consumption.

Determine the Intended Message

The concrete concept to be communicated through the visual based on the previous points …

At this point we can choose one of the two major presentation strategies:

Design, Good Design

Given the abundance of less-than-beautiful visualizations, it’s clear that the path to beauty is not obvious. (Beautiful Visualizations, How Do We Achieve Beauty?, page 6).

… intellectually enjoyable

-Good data visualization is scientific based

-do it as straightforward as possible

-we must justify the existence of any (graphical) element or (informative) content

-complexity is welcome as long as it is relevant/needed

-simple or not: irrelevant information is obstructive noise and should be avoided

-notwithstanding, less relevant yet useful information could be kept but should be de-emphasized—after considering its deletion in the first place, we must keep as few as posible of such low priority content

-a good balance is between few and too much information, on one side the presentation would be vague due to the lack of information; while on the other, it would be confusing because of data overloading—which is the most common of both cases.

///

… keep textual contents as few and integrated as possible

///

-general graphic design principles are as necessary as in any other deliverable—well design in terms of hierarchy, coherence, and simplicity.

-they will be applied for highlighting, hierarchize, associate, and dissociate items, but also for creating an aesthetically appealing presentation

-aesthetics are more important than we might think

The opposite, including large amounts of text and imagery are considered even by the average person as a bad, poor, and unattractive info/data visualization,

-if a professional (designer) gets it wrong, it’ll be perceived as something even worst;

-in general, people think design is easy, everyone can do it. Following that line of thinking a professional is required to do at least a good and nice visual, then if the visual looks ugly, is because is bad or wrong, in consequence it will lose the reader’s trust which is the last think we want.

-Aesthetics and credibility have a long known correlation.

Type of Charts

…

Reference: