They say a picture is worth a thousand words. In terms of getting your message across quickly, that’s true. Here are three rules to developing your skills in representing data visually.
Know your data.
You must understand your data before you share it. Is it numeric? Geospatial? Text? Time series? All of the above? Numeric data usually works best in bar charts, line graphs are great for changes over time, choropleths work great with geospatial data.
Does your data have outliers, things that just don’t make sense? Spreadsheets often have data entry typos, so make sure your data is clean and accurate before you create a visualization. Bad data = bad visualizations.
Make sure people who look at your visualization know that you used reliable data to create it by including a reference to the source of the information.
Know your message.
Your visualization answers a question about the data. You need to decide what that question is, and then determine the best way(s) to present the answer visually.
Here’s a fantastic example of a visualization that answers the question “How did America vote in the last presidential election?” It makes it easy to see not only how people voted but also where they live and how their votes were clustered. Most importantly, it follows the third rule.
Keep it simple.
When charting your data, it’s tempting to use all the tools in your toolbox. But the best visualizations are simple. Not everyone is an expert at reading charts, so it’s best to make it as easy as possible on your audience.
For example, using lots of color makes charts pretty, but 1 in 12 men and 1 in 200 women (about 4.5% of the total population) is color blind, so your pretty colors don’t help them understand your visualization at all.
Also remember that most people get their news (including charts) on mobile devices these days, so your visualization needs to work on a screen as small as a smartphone’s.
To the right is a great example of how someone took a reasonably simple chart about how teens use Snapchat, Instagram, Facebook, and Twitter, and made it even simpler to read.
As you can see from all the links and examples in this post, there’s a ton of information available on the internet. Like all things on the internet, some is good information and some is not so good. Here are some helpful links to get you started.
- FlowingData: This site, run by Nathan Yau, has information on visualizations, statistics, infographics, source data, design, etc. It costs money to be a member ($11 per month or $72 per year), but there’s lots of great free content too, like these 7 basic rules for making charts and graphs. This is my favorite site about data visualization, and I have Yau’s Famous Movie Quotes as Charts hanging on the wall in my office.
- Consumer Financial Protection Bureau Design Manual, Data visualization: The CFPB has some great rules for how they present data visually.
- Makeover Monday: Smart people take public data sets and make cool visualizations from them. My favorite is The World as 100 People.
- xkcd: some of the comics on this site have some coarse language, but there are also some brilliant data visualizations. Here’s one called the Timeline of Earth’s Average Temperature.
- Stephen Few’s Perceptual Edge: Few has written books and tons of articles on how to visualize data. He’s smart and opinionated. He also really dislikes pie charts. These articles are long, but they are worth reading.
