Graphs and Charts: Frequently Asked Questions

Original author: Jorge Camoes
  • Transfer
New to data visualization? Then keep the answers to frequently asked questions.

What is a chart?

Just open your eyes and incredible amounts of data will penetrate your brain in the same second. Data is processed instantly and now you already understand the world around us. Forms and structures appear and you can already separate the flies from the cutlets.
Figuratively, a graph is a great tool whose advantage is that you can reflect numbers in space. As soon as the dots begin to line up one after another, you will see a pattern that you would hardly have discovered just by looking at the numbers. At the same time, when you create a diagram, your main concern should be to simplify the detection of this diagram. Everything else is no longer needed. Everything else is just marketing;)

And why should I use charts?

Yes, because data is a value, and you must extract as much from it as possible. Because graphs allow you to efficiently process information. Because they make it clear that you will not understand using other methods. And just because they save time.


How to create professional looking charts?

Do not imitate beautiful diagrams from magazines, the fact that they are placed there does not mean “professional looking”. As a rule, they have more form than meaning, a sort of meaningless lure. If you use charts that help make a decision, clear them of trash, let them tell their story :)

But my boss loves floating 3D pie charts in PowerPoint ...

I know very well that this can be a problem. Play with him by the rules and you will be safe (or not :) If you violate it, you will be promoted or fired ...
In short, if you want to change this corporate culture by mocking the data - go ahead! People are usually not stupid, they just do not know how to make beautiful. Enlighten. Let them feel the difference with what they know and what you can do. Be patient and persistent. People change. But slowly.

What programs to use?

If you work in a corporate environment, you probably can’t avoid Excel, or similar spreadsheet editors (Powerpoint should also be present there, but try to avoid it). Do not use any dumb default settings in Excel and then this can be a good starting point.
If you are design-oriented, you could try Illustrator or a good combination of design and programming such as Processing ( translator's note: NodeBox is even better, I wrote about it). Or forget the editing work with SAS statistical package, SPSS or R . Try SpotFire for interactive analysis of large databases.

Give me a little hint to improve the graphics now

Simplify

In terms of?!

Remove all the rubbish from the charts, think only about the important (imagine that your budget on charts is tightly cut). Add more free space, after which the graphics will sparkle with meaning. Clear, clean graphics are beautiful.

What about books and authors?

Not a single book had such an influence as the remarkable work of Edward Tufty, The Visual Display of Quantative Information (his books are pleasures for the eyes, so feel free to leave his book in the living room ...). Tufty combines minimalism with understandable terms (such as dirty charts , data-to- graph ratios, data density) to create a good foundation for understanding. Use his best practices to check the design of your charts. Discover other English-language authors: Stephen Few , Jacques Bertin , William Cleveland, John Tukey , Stephen Kosslyn or Colin Ware .

Anything online?

Edward Tufty leads the forum ( Translator's note: all resources are in English) On the Stephen Few website you can also find a forum and some interesting examples in the before / after style . It's hard to find a good use of good visualization in the media, but the New York Times is a good example. Well, and finally, a wonderful list of online resources. Blog

translation original

Also popular now: