This was one of my favorite projects I worked on while I was at University of South Florida as it combined things I learned as an undergraduate in Mathematics with things I learned as a professor in Visual Communication. Anscombe's Quartet is made up of four data sets that look identical under normal statistical analysis. However as the graphs seen in my power above illustrate they graph very differently. My student and I used this poster to show how this could clearly extend to multi-dimensional visualization.

The core concept is illustrated above. The four X,Y data sets from Anscombe's Quartet were used to generate the X,Y,Z data set visualize above. The XYZ set is pictures from top, front, and side to see how drastically different it looks, even though statistical analysis of the XYZ sets would not seem hugely different.

The abstract related to this paper can be found at the link below:

This poster and abstract was presented at the annual conference for the Association for Information Science & Technology (ASIST) 2013, November 1-6, 2013, Montreal, Quebec, Canada.