1. Know your data & purpose. Know how each variable was measured and have some hypotheses about what you might see.
♦ It is helpful to make sure your data is clean and error free.
♦ Decide whether you are trying to find patterns or tell a story to the reader (Cole Nussbaumer Knaffic)
2. Pick a chart type that best accomplishes your purpose, given the data you have.
♦ Each chart type uses different visual properties to encode the data, like points, lines, bars, or boxes
♦ Chart Choosers can help with this. Sketch out your visualization on paper before trying to create it on a computer.