Noelle Olsen

The past two weeks of class have been dedicated to fine tuning our data visualization techniques. The first part consisted of bringing in a chart or graph from our research and presenting our data in a whole new way. Wilmelie Cruz Marrero reported on what we learned during that week in a previous blog post, and I will reference some of her examples used later in this post. Rather than focusing on the analytical tools used, we discussed the aesthetic elements. Jane Thomas from great tutorial on YouTube which I would recommend for anyone looking for an introduction to data visualization.

A quick review of data visualization tips:

• Remove chart junk/clutter/excessive words
• Use active titles with the most important words first
• Be color-wise and colorful, certain colors like red can be used to emphasize a main point
• Use mixed media (e.g. adding a picture of a species instead of using its scientific name)
• Utilize multiple softwares/programs (e.g. both R or Excel and Adobe Illustrator)
• Eliminate unnecessary data—not all data is needed to show main point!

The recommendation that came up time and time again when we went over my classmates’ graphs was to reduce clutter. Streamline. Take away x, y, and z. Remove these words. Eliminate redundancy. This reminded me of the application of feng shui principles when organizing and decorating rooms. In order to enhance the flow of energy in your room, you should be mindful of colors, beware of too many sharp corners, remove clutter, etc. On the same note, too many elements or lines in your graph often turns your graph chaotic, confusing your audience, muddling your main point, and decreasing the overall flow of your diagram. Colors can either work for or against you: we talked a lot about how red can be used to emphasize a point (whether you intend it to or not) or represent something bad (e.g. anoxic conditions), so use cautiously.

My “before” and “after” graphs and challenges with coding:

I completed my statistical analysis and graph using R programming and eventually found myself being limited by the language of R itself. Luckily, there are plenty of online resources for editing graphs in R. Because R requires running code before the graph changes will appear—as opposed to manually fixing the lines or colors in Adobe Illustrator—I seriously made 50+ versions of the graph (and still wasn’t 100% satisfied). Initially, I was being stubborn and wanted to create my graph solely in R. Once I was on the verge of giving up, I decided to edit my latest R version using Adobe Illustration. Eureka! Adobe Illustrator allowed to make the finishing touches I was looking for in R (particularly, re-writing the y-axis labels).

My critiques:

• Change vertical units to %
• Include 50% line/dashed line @ 50% line
• Make dots smaller
• Move legend outside of box
• Use arrow lines to point to where 50% maturity occurs on x-axis
• More informative title
• Remove outline of chart frame

What I changed:

• Made my title active and more informative
• Changed my y-axis to read % maturity instead of proportion mature
• Reformatted the line at 50% maturity and added arrow line point at width at 50% maturity
• Edited my legend to be less distracting
• Removed outer box on graph to open up and spread out data points

And the awards for most improved go to….*drumroll*

1. Keota Silaphone: Keota did a great job incorporating her critiques and advice from our first class and created something really special! Her use of color to coordinate with the different plants mixed with the perfectly circular, chronological diagram create a harmonious and inviting diagram. Where there was once a lot of confusing clutter and plain text, there is now a great flow of information that is not only visually appealing, but also dynamic!

2. Kate Gillespie: I had to do a double take when I first saw Kate’s latest version of her phylogenetic tree! Kate had one of the biggest data challenges, because genomics, well, are so complicated! Another great example of utilizing color to reduce clutter, increase readability, and make the diagram more inviting. Kate’s latest version is a truly admirable piece of data visualization art full of flow and finesse.

Some final thoughts…

If I’m being completely honest, I never thought twice about putting the “before” figure in a PowerPoint presentation. After going through this graph with my class, however, I never want to go back to the original! I would absolutely recommend asking people who aren’t directly involved in your research to look over your graphs for recommendations. There is almost always room for improvement, so make sure you are open to suggestions and critiques (and don’t take it personally)! Remember to remove as much “chart junk” as possible to streamline your message and let your data speak for itself. You want your audience to be able to grasp the main point of your data as easily and comfortably as possible. Harmonize the elements in your graph to achieve great results!