In this blog post, I talked about researchers' fear of leaving information out of their…
Sometimes, Data Soapbox (i.e., me) provides pro bono assistance to members of our “support team” (in this post, my husband). Last week, my engineer spouse asked for input on the best way to visualize a boatload (technical term) of data from testing a piece of equipment.
He had a preliminary graph of the data in Excel. It looked something like this:
I asked my husband to tell me more about the data, and what his team wanted to be able to say with it. To paraphrase our conversation:
Husband: We’ve promised that our equipment can process 20 tons of stuff a day. We need to back up that claim.
Me: Does anyone need to know about the variation in amount processed from day to day?
Me: Does anyone need to know the average number you hit, or the average amount by which you surpassed the target?
Me: Can you just say that you met the 20-ton threshold 86% of the time?
As with all research communication, the core question was, “What message needs to be conveyed to the target audience?”
Sometimes a message is best communicated through charts, graphs, or diagrams. This could include when you need to show a comparison, a process, how something has changed over time, or where things are located.
But sometimes all your audience needs is a number:
Q: Will my plants freeze overnight?
A: Probably not. The low temperature is forecasted to be 39 degrees.
Q: Did this job candidate perform well in school?
A: Yes, they graduated with a 3.9 GPA.
Q: If I write a check for $300, will it bounce?
A: No. Your current bank balance is $1,693.01.
Q: Did the Yankees wipe the floor with the Red Sox last night?
A: Absolutely. The Yankees won, 17 to 0.
Notice that the audience doesn’t need the low temperatures for every day in the past month or the GPAs of everyone in the school. But when we have a lot of data at our fingertips and an easy tool to graph them, it’s tempting to default to a dataviz.
Unfortunately, when we show an audience more information than they need, we run the risk that they will get distracted and miss the message. Also, when we visualize information that could be shared in a simple sentence, we waste valuable real estate on the screen or page. That space could instead be used to share additional insights, give eyeballs more white space, or save paper/ink/bytes.
What other sorts of insights can stand on their own, without a data visualization? Share in the comments below!