Storytelling with Data
Page 7
Gestalt principles of visual perception
When it comes to identifying which elements in our visuals are signal (the information we want to communicate) and which might be noise (clutter), consider the Gestalt Principles of Visual Perception. The Gestalt School of Psychology set out in the early 1900s to understand how individuals perceive order in the world around them. What they came away with are the principles of visual perception still accepted today that define how people interact with and create order out of visual stimuli.
We’ll discuss six principles here: proximity, similarity, enclosure, closure, continuity, and connection. For each, I’ll show an example of the principle applied to a table or graph.
Proximity
We tend to think of objects that are physically close together as belonging to part of a group. The proximity principle is demonstrated in Figure 3.1: you naturally see the dots as three distinct groups because of their relative proximity to each other.
Figure 3.1 Gestalt principle of proximity
We can leverage this way that people see in table design. In Figure 3.2, simply by virtue of differentiating the spacing between the dots, your eyes are drawn either down the columns in the first case or across the rows in the second case.
Figure 3.2 You see columns and rows, simply due to dot spacing
Similarity
Objects that are of similar color, shape, size, or orientation are perceived as related or belonging to part of a group. In Figure 3.3, you naturally associate the blue circles together on the left or the grey squares together on the right.
Figure 3.3 Gestalt principle of similarity
This can be leveraged in tables to help draw our audience’s eyes in the direction we want them to focus. In Figure 3.4, the similarity of color is a cue for our eyes to read across the rows (rather than down the columns). This eliminates the need for additional elements such as borders to help direct our attention.
Figure 3.4 You see rows due to similarity of color
Enclosure
We think of objects that are physically enclosed together as belonging to part of a group. It doesn’t take a very strong enclosure to do this: light background shading is often enough, as demonstrated in Figure 3.5.
Figure 3.5 Gestalt principle of enclosure
One way we can leverage the enclosure principle is to draw a visual distinction within our data, as is done in the graph in Figure 3.6.
Figure 3.6 The shaded area separates the forecast from actual data
Closure
The closure concept says that people like things to be simple and to fit in the constructs that are already in our heads. Because of this, people tend to perceive a set of individual elements as a single, recognizable shape when they can—when parts of a whole are missing, our eyes fill in the gap. For example, the elements in Figure 3.7 will tend to be perceived as a circle first and only after that as individual elements.
Figure 3.7 Gestalt principle of closure
It is common for graphing applications (for example, Excel) to have default settings that include elements like chart borders and background shading. The closure principle tells us that these are unnecessary—we can remove them and our graph still appears as a cohesive entity. Bonus: when we take away those unnecessary elements, our data stands out more, as shown in Figure 3.8.
Figure 3.8 The graph still appears complete without the border and background shading
Continuity
The principle of continuity is similar to closure: when looking at objects, our eyes seek the smoothest path and naturally create continuity in what we see even where it may not explicitly exist. By way of example, in Figure 3.9, if I take the objects (1) and pull them apart, most people will expect to see what is shown next (2), whereas it could as easily be what is shown after that (3).
Figure 3.9 Gestalt principle of continuity
In the application of this principle, I’ve removed the vertical y-axis line from the graph in Figure 3.10 altogether. Your eyes actually still see that the bars are lined up at the same point because of the consistent white space (the smoothest path) between the labels on the left and the data on the right. As we saw with the closure principle in application, stripping away unnecessary elements allows our data to stand out more.
Figure 3.10 Graph with y-axis line removed
Connection
The final Gestalt principle we’ll focus on is connection. We tend to think of objects that are physically connected as part of a group. The connective property typically has a stronger associative value than similar color, size, or shape. Note when looking at Figure 3.11, your eyes probably pair the shapes connected by lines (rather than similar color, size, or shape): that’s the connection principle in action. The connective property isn’t typically stronger than enclosure, but you can impact this relationship through thickness and darkness of lines to create the desired visual hierarchy (we’ll talk more about visual hierarchy when we discuss preattentive attributes in Chapter 4).
Figure 3.11 Gestalt principle of connection
One way that we frequently leverage the connection principle is in line graphs, to help our eyes see order in the data, as shown in Figure 3.12.
Figure 3.12 Lines connect the dots
As you have learned from this brief overview, the Gestalt principles help us understand how people see, which we can use to identify unnecessary elements and ease the processing of our visual communications. We aren’t done with them yet. At the end of this chapter, we’ll discuss how we can apply some of these principles to a real-world example.
But first, let’s shift our focus to a couple of other types of visual clutter.
Lack of visual order
When design is thoughtful, it fades into the background so that your audience doesn’t even notice it. When it’s not, however, your audience feels the burden. Let’s look at an example to understand the impact visual order—and lack thereof—can have on our visual communications.
Take a moment to study Figure 3.13, which summarizes survey feedback about factors considered by nonprofits in vendor selection. Note specifically any observations you may have regarding the arrangement of elements on the page.
Figure 3.13 Summary of survey feedback
As you look over the information, you might be thinking, “this looks pretty good.” I’ll concede: it’s not horrible. On the positive side, the takeaway is clearly outlined, the graph is well ordered and labeled, and key observations are articulated and tied visually to where we’re meant to look in the graph. But when it comes to the overall design of the page and placement of elements, I’d have to disagree with any praise. To me, the aggregate visual feels disorganized and uncomfortable to look at, as if the various components were haphazardly put there without regard for the structure of the overall page.
We can improve this visual markedly by making some relatively minor changes. Take a look at Figure 3.14. The content is exactly the same; only the placement and formatting of elements have been modified.
Figure 3.14 Revamped summary of survey feedback
Compared to the original visual, the second iteration feels somehow easier. There is order. It is evident that conscious thought was paid to the overarching design and arrangement of components. Specifically, the latter version has been designed with greater attention to alignment and white space. Let’s look at each of these in detail.
Alignment
The single change having the biggest impact in the preceding before-and-after example was the shift from center-aligned to left-justified text. In the original version, each block of text on the page is center-aligned. This does not create clean lines either on the left or on the right, which can make even a thoughtful layout appear sloppy. I tend to avoid center-aligned text for this reason. The decision of whether to left- or right-justify your text should be made in context of the other elements on the page. In general, the goal is to create clean lines (both horizontally and vertically) of elements and white space.
Presentation software tips for aligning elements
To help ensure that your elements line up when you are placing them on a page within your presentation software, turn on the rulers or gridlines that are built into most programs. This will allow you to precisely align your elements to create a cleaner look and feel. The table functionality built into most presentation applications can also be used as a makeshift brute-force method: create a table to give yourself guidelines for the placement of discrete elements. When you have everything lined up exactly like you want it, remove the table or make the table’s borders invisible so that all that is left is your perfectly arranged page.
Without other visual cues, your audience will typically start at the top left of the page or screen and will move their eyes in a “z” shape (or multiple “z” shapes, depending on the layout) across the page or screen as they take in information. Because of this, when it comes to tables and graphs, I like to upper-left-most justify the text (title, axis titles, legend). This means the audience will hit the details that tell them how to read the table or graph before they get to the data itself.
As part of our discussion on alignment, let’s spend a bit of time on diagonal components. In the previous example, the original version (Figure 3.13) had diagonal lines connecting the takeaways to the data and diagonally oriented x-axis labels; the former were removed and the latter changed to horizontal orientation in the makeover (Figure 3.14). Generally, diagonal elements such as lines and text should be avoided. They look messy and, in the case of text, are harder to read than their horizontal counterparts. When it comes to the orientation of text, one study (Wigdor & Balakrishnan, 2005) found that the reading of rotated text 45 degrees in either direction was, on average, 52% slower than reading normally oriented text (text rotated 90 degrees in either direction was 205% slower on average). It is best to avoid diagonal elements on the page.
White space
I’ve never quite understood this phenomenon, but for some reason, people tend to fear white space on a page. I use “white space” to refer to blank space on the page. If your pages are blue, for example, this would be “blue space”—I’m not sure why they would be blue, but the use of color is a conversation we will have later. Perhaps you’ve heard this feedback before: “there is still some space left on that page, so let’s add something there,” or worse, “there is still some space left on that page, so let’s add more data.” No! Never add data just for the sake of adding data—only add data with a thoughtful and specific purpose in mind!
We need to get more comfortable with white space.
White space in visual communication is as important as pauses in public speaking. Perhaps you have sat through a presentation that lacked pauses. It feels something like this: there is a speaker up in front of you and possibly due to nerves or perhaps because they’re trying to get through more material than they should in the allotted time they are speaking a mile a minute and you’re wondering how they’re even able to breathe you’d like to ask a question but the speaker has already moved on to the next topic and still hasn’t paused long enough for you to be able to raise your question. This is an uncomfortable experience for the audience, similar to the discomfort you may have felt reading through the preceding run-on, unpunctuated sentence.
Now imagine the effect if that same presenter were to make a single bold statement: “Death to pie charts!”
And then pause for a full 15 seconds to let that statement resonate.
Go ahead—say it out loud and then count to 15 slowly.
That’s a dramatic pause.
And it got your attention, didn’t it?
That is the same powerful effect that white space used strategically can have on our visual communications. The lack of it—like the lack of pauses in a spoken presentation—is simply uncomfortable for our audience. Audience discomfort in response to the design of our visual communications is something we should aim to avoid. White space can be used strategically to draw attention to the parts of the page that are not white space.
When it comes to preserving white space, here are some minimal guidelines. Margins should remain free of text and visuals. Resist the urge to stretch visuals to take up the available space; instead, appropriately size your visuals to their content. Beyond these guidelines, think about how you can use white space strategically for emphasis, as was illustrated with the dramatic pause earlier. If there is one thing that is really important, think about making that the only thing on the page. In some cases, this could be a single sentence or even a single number. We’ll talk further about using white space strategically and look at an example when we discuss aesthetics in Chapter 5.
Non-strategic use of contrast
Clear contrast can be a signal to our audience, helping them understand where to focus their attention. We will further explore this idea in greater detail in later chapters. The lack of clear contrast, on the other hand, can be a form of visual clutter. When discussing the critical value of contrast, there is an analogy I often borrow from Colin Ware (Information Visualization: Perception for Design, 2004), who said it’s easy to spot a hawk in a sky full of pigeons, but as the variety of birds increases, that hawk becomes harder and harder to pick out. This highlights the importance of the strategic use of contrast in visual design: the more things we make different, the lesser the degree to which any of them stand out. To explain this another way, if there is something really important we want our audience to know or see (the hawk), we should make that the one thing that is very different from the rest.
Let’s look at an example to further illustrate this concept.
Imagine you work for a U.S. retailer and want to understand how your customers feel about various dimensions of their shopping experience in your store compared to your competitors. You have conducted a survey to collect this information and are now trying to understand what it tells you. You have created a weighted performance index to summarize each category of interest (the higher the index, the better the performance, and vice versa). Figure 3.15 shows the weighted performance index across categories for your company and five competitors.
Figure 3.15 Original graph
Study it for a moment and make note of your thought process as you take in the information.
If you had to describe Figure 3.15 in a single word, what would that word be? Words like busy, confusing, and perhaps exhausting come to mind. There is a lot going on in this graph. So many things are competing for our attention that it is hard to know where to look.
Let’s review exactly what we’re looking at. As I mentioned, the data graphed is a weighted performance index. You don’t need to worry about the details of how this is calculated, but rather understand that this is a summary performance metric that we’d like to compare across various categories (shown across the horizontal x-axis: Selection, Convenience, Service, Relationship, and Price) for “Our Business” (depicted by the blue diamond) compared to a number of competitors (the other colored shapes). A higher index represents better performance, and a lower index means lower performance.
Taking in this information is a slow process, with a lot of back and forth between the legend at the bottom and the data in the graph to decipher what is being conveyed. Even if we are very patient and really want to get information out of this visual, it is nearly impossible because “Our Business” (the blue diamond) is sometimes obscured by other data points, making it so we can’t even see the comparison that is most important to make!
This is a case where lack of contrast (as well as some other design issues) makes the information much harder to interpret than it need be.
Consider Figure 3.16, where we use contrast more strategically.
Figure 3.16 Revamped graph, using contrast strategically
In the revised graph, I’ve made a number of changes. First, I chose a horizontal bar chart to depict the information. In doing so, I rescaled all the numbers to be on a positive scale—in the original scatterplot, there were some
negative values that complicated the visualization challenge. This change works here since we’re more interested in relative differences than absolute values. In this remake, the categories that were previously along the horizontal x-axis now run down the vertical y-axis. Within each category, the length of the bar shows the summary metric across “Our business” (blue) and the various competitors (grey), with longer bars representing better performance. The decision not to show the actual x-axis scale in this case was a deliberate one, which forces the audience to focus on relative differences rather than get caught up in the minutiae of the specific numbers.
With this design, it is easy to see two things quickly:
We can let our eyes scan across the blue bars to get a relative sense of how “Our business” is doing across the various categories: we score high on Price and Convenience and lower on Relationship, possibly because we’re struggling when it comes to Service and Selection, as evidenced by low scores in these areas.