Chart Design Considerations (10-15 min)
Last class we talked about when and what charts to used based on different kinds of scenarios and data we might have. Let’s quickly review from Miller’s chapter about some general considerations when it comes to axis construction, scales, design choices, etc.
- Is the title clear about what the variables are, what their relationship is, what is being measured, etc.? Is the topic distinguishable from other charts or tables used in the paper?
- Is the chart self-contained? Could it survive outside of the paper? That is, could someone interpret it (mostly!) and get a sense of what it means without looking at the text?
- To help self-containment: at least some context is present (Ws), units or categories of all variables clear, legend provided if needed, terms/abbreviations/symbols are defined, data sources are provided.
- Does the organization of the chart coordinate with the narrative of the paper? E.g., theoretical, alphabetical.
- Consistent design choices (e.g., color, line style, spacing)
- Axis scale consistent within chart and with other charts? Deviations okay if good reason for it.
- Do the axes have intervals that are easy to read and useful for the analysis interpretation?
- Would data labels help?
- Is there too much? Is it easy to read?
- Does the axis scale include zero if zero is plausible value?
- Accessibility: color is not only way to make meaning, good font size to read from far away, not overly complicated wording in title (e.g., lots of clauses), etc.
Activity: On 9/12, we talked about how we can use language in different ways to signal evaluations of numbers. For instance, in the examples from that lesson, we talked about how the same statistic was framed in different ways–one to tell a story about overall trends in the 2017 labor market and one about missing expectations for December 2017 specifically.
We tried our hand at writing about a specific statistic from this article about recycled plastic. You’ll notice that there are three charts in this article. Consider the above points from Miller above. What do you think these charts do well in terms of the pattern or relationship they are trying to visualize? Are they easy to read? Why or why not? How about the design and accessibility of these charts? What could be revised if anything?
Journal 7 and “Emotion” (20 min)
What is emotion? What is objective? What is a “neutral” argument? What does that mean?
Why not just hand them the spreadsheet?
What is an argument? What is emotion?
But reality is, that the data will always be persuasive because the creator of the visualization will pick the variables that make a certain aspect standout, as that is their job. Although they may intend to be neutral and just offer some data to the reader, that is not how it ends up working out most of the time. From my personal standpoint, I know that when I produce a data visualization, I am most definitely trying to get a point across to the reader. However, I would not say that I am trying to make them feel an “emotion.”
Knowledge is always partial…a trick to make you believe that you can see everything, all at once, from an imaginary and impossible standpoint.” I agree with this statement because no chart or table is ever able to encapsulate the full picture of what’s going on. I’m currently taking a film class at the moment, where we’ve learned that every frame in a film is a deliberate choice of what to include, and thus what to exclude out of the frame as well. I don’t think the choice of data in a visualization is any different. So in a sense, I think an attempt by an author to pretend like he or she didn’t purposely pick data for a visualization is misleading. Instead, I think authors should embrace their deliberate choice of data if they truly want to be transparent in their argument.
Why not use emotion?
shouldn’t the goal of an author’s work be to produce material that will make a lasting impact on the reader? As long as the ‘embellishment’ of the visualization doesn’t make it misleading, I think emotion should always be used as a tool to attract the reader’s attention. If a chart or table is made blander in an attempt to seem more impartial or unbiased, then I think the purpose of that chart or table is being defeated.
Emotion should not be used to trick an individual but can and should be used to help get a message across.
What is context?
If data cannot be presented neutrally, do we have to treat it with complete relativity? Obviously not. There is, of course, an inherent truth value to data. To use the gun violence example from Periscopic used by D’Ignazio and Klein, there is a correct number of gun deaths that occurred between 2010 and 2013. However, as Nikki Stevens points out in a comment, Periscopic has certainly missed or excluded certain gun related deaths. This is where transparency comes in. By being open about the data collection, limitations, and potential holes in analysis, the data, while not being neutral, can still be informative and correct without having to sacrifice the trust of the reader.
The inclusion of clarification and real-world context is necessary to maintain integrity in these situations. This is where emotion can come in to be a bridge between what a data visualization looks like and what it is actually saying to the reader. The clarifying rhetoric that should always follow visualization lessens the tendency to make “mental shortcuts to make judgements” and not see the full picture.
Minimalism!!
The purpose of the stolen years’ graphic is a very unique and interesting idea with a powerful premise yet looking at the graphic doesn’t help me visualize the stats behind it at all. How they describe the statistics beforehand to setup the chart makes more sense to me than the chart itself. When the description of the chart makes more sense than the chart itself, a problem exists.
Sci/Tech Writing?
Personally, it would be hard for me to imagine the “Monstrous Costs” graph in a formal, government report. It’s interesting that, although all people react to the more creative, design-focused visuals better than the simple ones, it would seem out of place to include something so memorable in a technical or scientific piece of writing. After all, it’s still people reading technical and scientific pieces, but I’d imagine that including something like “Monstrous Costs” would take away some of the trust that the reader has in the writer’s expertise and professionalism.
The current state of the scientific and academic communities is one that strongly believes in the “unemotional” and “neutral” presentation of data, i.e. standard histograms, pie charts, bar graphs, etc.. Therefore, even if unique data presentations are just as valid, and have the added bonus of potentially being easier for non-scientific audiences to understand and remember, they are not allowed in scientific papers (or at least, not that I have seen). This then perpetuates the cycle of standardization of data presentation, making it harder to break the mold.
Telling someone about uncertainty just does not seem as effective as making them feel uncertainty. It is age-old writing advice: show, don’t tell.
Emotion never goes away. You cannot make an appeal purely to logic, purely to emotion, or purely to one’s own credibility. All three are always there, it is just a matter of how you leverage them. Thoughts?
how might activating emotion – leveraging it, rather than resisting emotion in data visualization – help us learn, remember, and communicate with data?
It is defined by what Harding calls “strong objectivity” which acknowledges that regular-grade, vanilla objectivity is mainly made by mostly rich white guys in power and does not include the experiences of women and other marginalized groups. (context?)
Infographic activity (30 min)
What qualifies as an infographic? What are its characteristics? How are they different from things like bar graphs that you can generate from Microsoft Excel or a table that you can make in a word processor?
Are there conventions as a genre you can begin to see as you compare these four infographics?
Think about these questions as you and your group look at the following infographics:
Caffeine in food and drink industry (scroll down for the larger graphic).
Acquisition strategies of tech companies
Let’s try to define some conventions for this genre and what is possible when composing an infographic. What should we keep in mind in regard to the following?:
Color?
Typography?
Arrangement?
Size?
Use of text?
Use of images?
Motion?
Where does an infographic go? For what purpose? For what kind of rhetorical situation (problem, constraints, audience)? What media can utilize it?
Why would you do an infographic and not a traditional table or chart? What is the difference between an infographic and such a thing? How about a bar graph with some nice design elements: infographic or not?
Activity: Go back to the visual representations in the recycling article. Would you call them infographics? Which ones and why or why not? Go back and revise one of them–infographic or not–and make some changes in light of what we have talked about so far.
Quick Review / Questions for Proj (5-10 min)
-Let’s go over prompt one more time.
Next Time (2-5 min)
-Turn in draft of Sci/Tech Writing project by 11:59pm this Thursday (11/7). In class on 11/7, you will bring 2 print copies of your in-progress draft to class for a peer response activity.
-We will also be using the free version of Tableau for a quick workshop on some basic functionality to play with for your Data Visualization project due at the end of the term.
-Speaking of, before class, this is IMPORTANT (If you do not have Windows or Mac for your operating system, you can’t download the program and will work with a partner on Thursday):
- Go to https://public.tableau.com/en-us/s/
- Enter email address and click “download the app” right next to that.
- Make sure you have the correct version for Windows or Mac.
- Might take about 10 min to download.
- Run the application to download the program (will have to restart computer when done, I’m pretty sure.