A common basic question that arises in the course of visualizing data is: what is the best type of chart to use? First, think about what it is you want to show. Is it a relationship between two variables, a distribution, a comparison? The answer to this question will lead you to the right suite of charts that might be appropriate. Andrew Abela (www.ExtremePresentation.com) has developed a chart chooser to help with this exercise:
Once you've answered the question of what you want to show and determined the right suite of charts that might be appropriate, the right answer to the question what is the right chart for my situation? will always be the same: whatever will be easiest for your audience to read. There's an easy way to test this, which is to make your visual and share it with a friend or colleague. They don't need any context; actually, it's better if they don't have any (this puts them in a position similar to your audience, who will always be less familiar with what you want to communicate than you are). Ask them to talk you through what they see: where they focus, what observations they make, what questions they have. This will help you to see if you are on the right track when it comes to whether what you want to communicate is coming across, or in the case that iteration is needed, where to focus your effort.
The meta-lessons here are 1) choose a visual display that's appropriate for the data and information you are trying to show, and 2) seek feedback, there is tremendous value in getting a fresh perspective from someone less familiar with the data than you are to help you iterate for success. A related tip is to start with a blank piece of paper, which can be helpful as you try to determine the visual display that will work best as part of your early iterating process.
Do you have other resources or tips for choosing the right type of visual display? Leave a comment with your thoughts!