How do I know if my figure is too complicated?
From the post:
One of the key things every statistician needs to learn is how to create informative figures and graphs. Sometimes, it is easy to use off-the-shelf plots like barplots, histograms, or if one is truly desperate a pie-chart.
But sometimes the information you are trying to communicate requires the development of a new graphic. I am currently working on a project with a graduate student where the standard illustration are Venn Diagrams – including complicated Venn Diagrams with 5 or 10 circles.
The post goes onto give some good suggestions for evaluating a figure for being “…too complicated.”
One of the comments takes up the refrain:
Simplicity is good, but you shouldn’t sacrifice the possibility of depth of understanding in order to save your readers from the pain of having to think.
True, but when we make a diagram, don’t we usually make it of our current, rather complete understanding? We omit the near misses, torn up drafts, slips of the pen, consequential and otherwise, and present the reader with a polished final product.
Which bears little resemblance to how we got there, but looks impressive now that we have arrived.
I don’t think readers mind thinking, but on the other hand, we should not expect readers to be smarter than we are.
There isn’t any reason, other than habit, that we can’t say: “When I saw graphic X, which is incorrect but it made me think of…” and make our publications more closely resemble how we actually do research.