Karl Bowman offers a list of the top 10 worst graphs in the scientific literature. Bowman not only critiques each graph, but also offers suggestions to improve the graph in the future.
For this list, the social sciences were not included. I imagine that we have some relatively uninformative displays in our articles, but I have not seen a comparable list for our discipline—perhaps due to the obvious disincentives that exist for compiling such a list together.
