Zeb Gordon
Professor Faull
Humn 270
3/27/18
Assignment #4
Data visualizations are far deeper than they appear. Much like a computer, the final resulting picture is the process of many tiny inputs and decisions which are all heavily involved in the power of the outcome. Most important of these decisions is the narrative of the data. Understanding that data is key to presenting it well, and many factors, from audience, to genre, to the creator itself can affect how this narrative is formed.
One important aspect of weaving a convincing and interesting narrative is understanding your target audience. Like anything made for public consumption, understanding who will be looking at this visualization is important for tailoring it to fit something relatable to them. Once you understand this who, adapting small details to make it more palatable for the viewer can drastically improve the perceived quality of the visualization. These things can be as sweeping as entire cultures. We in the West read left to right, and therefore are far more comfortable with visualizations going left to right (Segel and Heer, 2). Understanding key things as small as those can drastically change the reception of a visualization. Understanding your target audience can also affect how
According to Segel and Heer there are seven genres of visual narratives. They have created seven genres of narrative visualization: magazine style, annotated chart, partitioned poster, flow chart, comic strip, slide show, and video. Each of these styles comes with pros and cons and can be combined in ways to maximize the effectiveness of them. When choosing a genre, it is important to understand its pros and cons. Knowing them can help to accentuate your data to tell your narrative in the way that you want to be interpreted. These pros and cons fall into categories. These are also described by Segel and Heer, saying “Choosing the appropriate genre depends on a variety of factors, including the complexity of the data, the complexity of the story, the intended audience, and the intended medium.” (Segel and Heer) The final piece to the puzzle is author vs reader driven experiences. This descriptor essentially describes how focused the narrative is on leading the reader through the material. These factors can be easily seen in the example provided by them. The “Steroids or not, the Pursuit is On” poster described by Segel and Heer is part portioned poster and part flow chart (Segel and Heer). This data, while interesting and serious to some, is not as formal as say, a business proposal. This leads the designer to more casual and static genres, such as a portioned poster. Then considering the audience, who is likely going to already understand the subject matter and will be taking a cursory glance, the designer can incorporate the visually leading aspects of a flow chart, which is a very reader driven method as it allows the reader to explore the visualization. This is opposed to the budget forecast, which is a much more business-related visualization. Here the audience wants to be led through a clean visualization that only has the goal of relaying the information. Therefore, an annotated graph is well suited for this. Annotated graphs present information well, if a touch uninspired, which is perfect for this visualization.
Visual sequence is composed by two factors. As described by Segel and Heer, these come from visual narrative tactics and narrative structure. Visual narrative tactics are the visual portion of sequence. This visual guide is composed of three parts: visual structuring, highlighting, and transition guidance (Segel and Heer, 7). Visual structuring helps the viewer to gain their bearings in the visualization and be naturally progressed through it. Lima’s fascination with trees is a great example. Trees are a natural and easy visual guide that assists the viewer through the narrative of the visualization. Highlighting needs no explanation. Its simply the changing of color to direct attention. This method is incredibly easy to see in daily life. The final piece is transition guidance. This is just moving the scene seamlessly to not confuse the viewer. In a static image this could be an arrow, or it could be an animated transition like in a power point. All of these facets are part of just the visual aspect of narrative sequence. The logical side is just as in depth. Again, Segel and Heer describe 3 forms: ordering, interactivity, and messaging. While all of these are easier for the layman to understand at first glance, they are just as important from a logical perspective. From all of these aspects we can synthesize that sequence is key to one thing, and that is keeping the viewer engaged and understanding. Visualization can often times be overwhelming, and it is the job of sequencing to lead viewers through that clutter. Additionally, we can observe the previously mentioned genres more deeply to analyze how the data the represent differs due to their narrative sequence. Segel and Heer created an incredibly useful chart that summarizes how these genres operate. To highlight these differences, using starkly different genres is best. Three genres that use very different strategies are the video, comic strip, and the annotated graph. The video genre is well known and finds its strengths in being able to show exactly what the designer wants the viewer to see at each step of the visualization. Because of that, it relies heavily on visual narrative sequencing tactics, such as well edited cuts, transitions, close ups, motion, and character direction. However, it also completely lacks interactivity, which can cause the viewer to lose interest. Comic Strips also lack this interactivity but make up for with an understood relaxed nature of the visualization, as well easy to follow visual transitions and linear narrative. However these can often contain large lie factor numbers due to the comic nature, as seen in Tufte many examples. Many times their exaggerated nature causes incidental lie factor. As said by Tufte, “Perhaps graphics that border on cartoons should be exempt from the principle” (Tufte, 73). The last example is the annotated graph. The annotated graph is a common visualization mainly used for its interactivity and ease of communication. It uses labeling and messaging well to explain the visualization and is easy to follow thanks to its linearity. This form also has the change of lie factors with poor annotations and markings however, as seen in Tufte’s example of traffic deaths (Tufte, 74).
The designer plays the role of the spyglass in the visualization. The designer allows you to see what they want you to see. Tufte describes six principles of visualization: representation of numbers, clear labeling, show data variation, use standardized units with money, the number of variables should not be more than the dimensions, and graphics must not quote data out of context(Tufte,77). Breaking any of these results in a skewing of data. With this many principles to follow, it is not easy for any impartial body to create a visualization. And simply by being human, it is impossible to be unbiased. The expert on biased visualizations is Tuft and he has many examples of lie factors that may be incidental, but none the less damage the visualization. In his example on page 70, where dollar purchasing power is compared by a graphic dollar, the areas simply don’t accurately represent the power. While the creator was attempting to be clever, he simply can’t be unbiased due to error and human bias.
In conclusion, narrative to a visualization is nearly as important as what’s being told. Its genre and presentation can even morph the data into revealing different conclusions. Understanding how these narrative devices affect the visualization can greatly improve the creation of and interpretation of visualizations and is an important thing to understand for anyone involved in the humanities.