If you’ve collected data in your higher education research on engaged learning, chances are you’d like to present it to your readers in an effective, visually pleasing, and impactful way. The problem is that data visualization is a complex topic, and the default graph you get from putting your data into Excel is probably leaving you a bit depressed (for good reason). But it’s worth the bother — people love data visualizations. When done well, graphs can communicate your research more clearly and faster than words.

In this blog post, I’m going to share some of the tips from Cole Knaflic’s Storytelling with Data, which is one of my favorite books about data visualization. I’ll show an example of a not-great graph and make it better by applying each of Knaflic’s four (relatively) simple steps. We’ll also discuss these ideas in relation to academic writing about engaged learning specifically.

Step #1: Choose the right type of visualization

Knaflic says, “There are many different graphs and other types of visual displays of information, but a handful will work for the majority of your needs. When I look back over the 150+ visuals that I created for workshops and consulting projects in the past year, there were only a dozen different types of visuals that I used” (chapter 2, para 1). Yes, you’ve likely seen some really fancy and original data visualizations, but usually one of those simple graphs that you’ve encountered a million times is going to be the best choice. It may not be sexy, but you know that your readers will understand it (which is our ultimate goal!).

The most common types of data visualizations for higher ed research (and the types that most frequently appear in our books) are bar graphs (vertical, horizontal, and stacked), line graphs, scatterplots, and slopegraphs. Tableau has a good primer on which type of graph is appropriate for your data and what you are trying to display.

For my example, I’m going to show food insecurity over time (using data from the USDA). Since change over time is continuous data, a line graph will make the most sense to my readers. The data points are connected by a line, which illustrates the intensity of change from year to year. Here’s the default graph I get from Excel:

Step #2: Declutter your visualization

The line graph above is very basic, but Excel’s default settings produce a cluttered and busy graph. Why? Microsoft includes every chart element that could possibly be useful to the reader, which generally is way too much information. Knaflic says, “Clutter can make something feel more complicated than it actually is. When our visuals feel complicated, we run the risk of our audience deciding they don’t want to take the time to understand what we’re showing, at which point we’ve lost our ability to communicate with them. This is not a good thing” (chapter 3, para 9).

Remove borders, gridlines, and data markers; label your data thoughtfully and simply. These actions will reduce your readers’ cognitive load and help your data stand out.

Step 3: Focus your audience’s attention

A lot of academics default to showing their data in an exploratory way — that is, just showing the data in a visualization without any explanation or analysis. Ideally, you’d like your readers to examine your data carefully and draw conclusions from it. But many readers won’t take the time to do so. And you are the expert on your data; you have examined it, analyzed it, and drawn conclusions from it. Save your readers that work by telling them what is most important. You can always provide your complete data (in an appendix or supplemental online resource) for those readers who want to do their own analysis.

In my example, I’m showing food insecurity over time. But what do I really want my audience to remember from this graph? I’d like them to see the big jump in food insecurity during the 2008 recession. So I’m going to focus my reader’s attention with a callout and strategically labeled data points. I’ll also rewrite the title to be more specific.

Step #4: Think Like a Designer

The graphs you’ve been playing with as you analyze your data are just fine for internal use, but when you’re ready to publish, you’ll need to spend some time polishing! Small adjustments to font, color, and alignment can make a huge difference in how professional a data visualization looks. Knaflic advises, “You may think to yourself, But I’m not a designer! Stop thinking this way. You can recognize smart design” (chapter 5, para 2). Take some time to tweak your visualization:

• Create visual hierarchy. Make important stuff stand out with bold text in darker colors. Make less important stuff recede with smaller text in shades of gray.
• Make the data visualization accessible with simple language and only a few colors. This is especially important if your visualization will appear in black and white.
• Pay attention to alignment. Left align all text, rather than centering it. Line everything up all neat and orderly.
• Use repetition to create a cohesive design. Use the same colors and fonts throughout the visual.
• Trust your eye! If it looks bad to you, keep working on it. Good design takes time.

Now isn’t that about a thousand times better than what we started with?

Other important considerations for academics writing about engaged learning:

• Your university probably has a data visualization expert who is ready and willing to help you with your project. Often, this person is a librarian (or at least works in the library). Find them! If you’re writing for the CEL blog or one of our book series, I’d love to work with you on your visualizations.
• Proportion and size matters when thinking about how your visualization will look in the published product. If it’s going in a book, you have limited page size to fit your visualization (often no more than 4-5 inches wide).
• Check with your publisher to see what formats they would prefer. At CEL, we prefer to receive data visualizations in the file native to the program it was created in (Excel, Illustrator, etc.).

If this post has intrigued you, I highly recommend reading Cole Knaflic’s entire book, Storytelling with Data, which is short, written for non-designers, and packed with great information! You can also check out her website. If you have questions, please let me know on Twitter @CEL_Elon.

This post is an installment of our series on academic book publishing. If you missed any of the other posts, check them out:

• the process for publishing a (SoTL) book
• developmental editing
• copyediting
• typesetting
• book cover design
• publishing open access

References

Knaflic, Cole. 2015. Storytelling with Data: A Data Visualization Guide for Business Professionals. John Wiley & Sons. O’Reilly https://learning.oreilly.com/library/view/storytelling-with-data/9781119002253/.

USDA (United States Department of Agriculture). 2019. “Food Security in the U.S.: Key Statistics and Graphics.” https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-us/key-statistics-graphics.aspx.

Jennie Goforth is the Center for Engaged Learning’s Managing Editor. She works with authors to shepherd their work from proposal through production in the Center’s Open Access Book Series. She also manages production of book websites and supplemental materials for the Stylus Publishing/Center for Engaged Learning Series on Engaged Learning and Teaching.

How to cite this post:

Goforth, Jennie. 2021, March 4. “Four Steps to Better Data Visualizations” [Blog Post]. Retrieved from http://www.centerforengagedlearning.org/four-steps-to-better-data-visualizations