A Novice's Confession on Data Visualization

By Seow Ling Ong, MSW | September 5, 2017
Research Associate, ETR

Colorful charts. Gorgeous graphs. Incredible infographics. Yes, everybody loves data visualization, and we can’t enough of it. But if you’ve never worked with dataviz, how do you get started?

Here is my own story—the confession of how a self-professed “data-only” person who is not a graphic designer ended up venturing into the colorful world of dataviz.

Less is More

In the world of research, older reporting styles tended to go the route of frequency tables and showing all the data for all response options. We would highlight some key findings, but the general ethos was, “We must be fair and objective. We do this by showing all of the data.”

The problem was that non-researchers and their audiences (for example, school administrators and parents within a school district) often couldn’t make a lot of sense out of the frequency tables. Data overload kept them from picking up the essential outcomes of the research.

Making a graphic interpretation of the data compels you to sit down and ask yourself some questions.

• “What is my reason for collecting this data?”
• “What does my audience want to know?”
• “What are the take-home messages that will be both most authentic and most useful for these stakeholders?”

The work is done behind the scenes and the result for the client is a clean, uncluttered graphic that distills the main points.

How Important Is 10%?

Here are a couple of examples. On one project, our team examined attitudes about making a specific public space non-smoking. Among people who used that space, 67% strongly agreed that smoke-free was a better choice. Only 10% disagreed. While we could have presented responses all along our scale, the striking finding was that two-thirds of our respondents supported establishing a non-smoking environment.

While all of the data was available to our clients in our final report, the figure we emphasized in our data visualization was the 67% measure of strong support. The 10% figure was not the important take-home.

In another situation, we were measuring student performance in a school. While most students were performing well or adequately, we found that 10% of students were performing below grade level. This was vital information for the school. They needed to focus on who those students were, set aside resources to assist them and improve their school success. In this case, our data visualization emphasized the 10% figure, making sure that every child was supported in achieving academic success.

Make It Bold

Ready to step into this wild world? Don’t be afraid to make your graphic BIG, BOLD and BEAUTIFUL! This is not the time to be coy and shy about the results. You want to get people excited and talking about your creation. Think of that one unforgettable, powerful data point, give it lots of attention, and share it with others!  Here’s one that I keep in my arsenal of inspirations!

There are many other reasons for including dataviz in your tool box—it can help us absorb information more easily, identify patterns rapidly, and gain a better understanding of data over time. There are many tips out there on how to get it done. If you are like me, knowledge doesn’t always transfer to behavior right away, and that’s okay. It took me two years to actually jump in and draft my first infographic! So I would like to share some “how-to’s” because taking the first step is often the hardest.

“Do-Re-Mi-Fa”

Jump in, choose a software (this is what I started with), template or favorite dataviz expert (here’s one of mine), and just DO it.

Then, yes, you guessed it, REpeat! Practice makes perfect!

Imitation is the sincerest form of flattery. Print out some templates or infographics that you like and MImic away.

Here’s one more part of the “jumping in” that’s been important to me: identify a mentor who’s FAmiliar with dataviz. We have an internal work group here at ETR where we can share interesting articles and dataviz examples. I’ve also been inspired by several of my more experienced colleagues, including Pam Drake, BA Laris and Jennifer Silver Herman.

Have you tried dataviz? Do you want to? Do you have a favorite program? Do you have tips of your own? I’d love to hear about dataviz in your own work.

Seow Ling Ong, MSW, is a Research Associate at ETR. She is experienced in creating comprehensive evaluation plans, developing survey instruments, managing databases, analyzing data and reporting. She can be reached at seowling@etr.org.