I admit it without any shame, I love this time of year… no “bah humbug” here, and Martha Stewart’s got nothing on me…well, except a few billion dollars.
Just like millions of other people around the world I love the dazzle and sparkle of the lights of the season. Every last one of them.
When it comes to holiday lights there are basically two camps…multi-colored or a single color, usually white. And it really doesn’t matter which camp you fall into because it’s purely a matter of preference…bright fun colors or the serenity and symmetry of a single color of lights. Both are good and both work to make things festive.
And that is the function of these lights — to be decorative and to dazzle us — pure and simple.
The use of color in your charts and graphs, however, serves a much different function. When you use color in your data displays it should never be because you want them to look bright and colorful, or because you want to entertain and dazzle your viewers. Rather, color in your displays of healthcare data should only be used when encoding differences in meaning.
I know, you’re probably thinking … “killjoy…bah humbug to you, too.” Oh well, such is my curse.
But here is the deal — your job is to create charts and graphs that impart information and support informed decision making. It is not to entertain or dazzle anyone with bright colors.
Consider the following graph:
When we see all of the different colors on the bars it causes our brain to ask “why are these things different?” And so, when color is used for no other purpose than to make a display “decorative” as in this example, it causes the viewer to use a lot of mental processing power, which takes away from their ability to see the message you wish to convey — the message you want the viewer to understand “at-a-glance.”
Now consider the same graph and how much easier it is to compare the overtime hours for each month. You don’t have to spend any time or energy searching for the meaning (that doesn’t exist) of the colors of the bars.
That doesn’t mean that you never use different colors in your graphs, you can and should when you need to show that something is different.
For example, the use of a contrasting color in a graph can be helpful if you need to highlight a specific piece of information, such as comparison data or targets — because it makes that information stand out from the rest-it serves to encode a difference in meaning.
Now it is easy to see each month’s overtime hours versus budget.
And remember, approximately 10% of all males and 1% of all females are colorblind and cannot discern differences in certain color combinations like green and red. So steer clear of those combinations and choose something that everyone can see, such as the blue and orange I have shown above.
If you follow this simple rule and use color to encode differences in your displays of data and not to try and jazz things up (that is the job of holiday lights and crayons) you will be certain to impart your message clearly and compellingly.
In a few days it will be time to un-decorate our home from the holidays and pack everything away until next year, which I am certain will fly by (yet again) in a blink of Rudolph’s shiny red nose. But for now I am going to enjoy the lights of the season, drink a whole lot of leftover eggnog and take a long, long, long winter’s nap.