Digital Design for Diverse Demographics

What should someone prioritize in digital design? Nice colors, clear site navigation, maybe big headlines? I’m sure most people can remember a website they thought had some fun feature or format, or some website they hoped would be fun but just didn’t work for one reason or another. This thought process may not be a top priority for many web designers. After all, the internet and websites are so ubiquitous that they have become the background noise to daily life. However, for websites, like most everything else, first impressions matter. A website interface is what gives a user that first impression. This is the main argument in Tara Andrews and J. van Joris’ article “What Are You Trying to Say? The Interface as an Integral Element of Argument.”¹ They propose that the interface of a website is the result of a series of purposeful choices and should be understood as purposeful by the creator and the user. These choices articulate an argument about the content of the page, whether the author intends it to or not. The choices a creator makes can enhance the purpose of the page, or detract from it.

A case study from the in-class activity to explore visualizations of data on Tableau Public can be used to illustrate their point. Using Andrews and van Joris’ idea, foundational instruction from Dr. Otis’ video on Data Visualization,² and the factors laid out in the article “Reframing the Conversation: Digital Humanists, Disabilities, and Accessibility,”³ we can identify a few key features to consider in digital design.

If data is being visualized, the type of chart chosen should be relevant and able to show the data in a readable and meaningful way. Using a type of chart that does not accurately display your data will misrepresent the data and put on the user a high responsibility to decipher and understand the data. Overcomplicated charts make data intimidating. Less is more, and simpler is better in data visualization, especially online.

Data visualizations should be legible without color before color theory makes your chart a gallery piece. Once data is visible in black and white, color can be added, but rainbow color palettes should be avoided. Color theory is not a cross-cultural form of communication, and color should be used to highlight certain aspects of the data, not just to make it fun to look at. In choosing color, creators should consider the pros and cons of different palettes. Different color combinations can be difficult to read for users with astigmatism or color blindness, so features to adjust or invert the font and background colors can greatly improve user experience.

Using these three general ideas, I want to investigate the Tableau Public page “Who Lives Inside Our Bedtime Books.”⁴ The page is a study of the ratio of gender representation in 284 children’s books with animal characters across decades from before the 1950s to 2020.⁵ The page shows 9 decades represented as tall buildings with rows of windows. Each building’s windows are assigned a color representing the gender of the animal characters in the book.⁶ Blue windows signify that all the animals are male, yellow that they are all female, blue and yellow that both are represented, and green meaning the animal is “it.”⁷ Hovering over or clicking on an individual window shows which book is represented, the year it was published, a rating out of 5 stars from an unlisted source,  and lists each animal character and how they are identified.⁸

All of this is to say, even a cute, fun idea can be heavily affected by its presentation. If the interface is not accessible, some users may not get any use of it at all. The data is presented in a way that undermines the stated purpose of the site (see footnotes for comments on data accuracy), and the functionality of moving between windows to see information about the books is slow and cannot be done without a mouse. Ultimately, digital design is a complex undertaking with a lot to consider, and exploring examples like these has been very enlightening. Keeping general principles of design in mind is good, but blind spots exist. The main takeaway from the referenced sources, and from class, has to be that talking about these ideas is key to identifying possible issues and adapting. No one can make a website usable by everyone, but presenting users with options and accepting feedback can go a long way. And one more thing: data accuracy was mostly noted in footnotes, but if you’re going to create a website about children’s books, you have to actually read them.

1. Andrews, Tara L. and Zundert, van. Joris J.  “What Are You Trying to Say? The Interface as an Integral Element of Argument,” Roman Bleier, Digital Scholarly Editions as Interfaces. Norderstedt, 2018. https://kups.ub.uni-koeln.de/9106/1/02_andrews_zundert.pdf ↩︎
2. Otis, Jessica M. Data Visualization. ↩︎
3. Brett, Megan R. Otis, Jessica M.  Kelly, Mills. “Reframing the Conversation: Digital Humanists, Disabilities, and Accessibility” Debates in Digital Humanities. 2023. ↩︎
4. Lukashik, Anna. Vizuators. “Who Lives Inside Our Bedtime Books” Tableau Public. https://public.tableau.com/app/profile/vizuators.make.over.monday/viz/WhoLivesInsideOurBedtimeBooks/Dashboard
   Note: No matter what I tried, I was not able to compress to potato-ify a screenshot of the page enough to include it in this post as a still image. The link should still work. ↩︎
5. For my classmates, this is the website that broke on my university laptop and became illegible because that laptop has a non-standard pixel ratio, which is its own can of issue-worms that I did not feel like experimenting with. ↩︎
6. No data source is identified in the Tableau, and from comparing a handful of represented books to my own family collection, I can say without a shadow of a doubt that a number of these are using incorrect data. As in, “the creator clearly did not even open some of these books” levels of wrong. I don’t know what went into gathering the data, but it wasn’t reading. ↩︎
7. For the green “it” category, at least one represented animal uses it for one animal, and the 10 other animals represented are not referred to by any pronouns. Some instances of “he/him” actually refer to “they/them” characters or nongendered groups of characters. ↩︎
8. I had decided to use this as a case study before I realized how inaccurate the data is. I feel like this just further proves my point, and I don’t feel like critiquing another example, so we’re gonna go with this. At least the errors are funny when you can compare them to the actual books. ↩︎