While the first half of "The Visual Display of Quantitative Information" addressed data visualization in the practical realm, the second half of the book approached it from the theoretical. The chapters addressed the process of creating visualizations from the particular (eliminate Moire Vibrations, avoid grids) to the philosophical (maximize data density to working with large data sets). I wanted to point out a few particularly interesting points:
- Quartile Plots
The "box plot" is a standard sight in statistics classes, but Tufte provides an alternative way of depicting the same idea - all in the name of cutting down on non-data-ink. The alternative, a quartile plot, astounded me with its visual simplicity, while still effectively conveying all of the information.
- Different levels of depth found in graphics
"Graphics can be designed to have at least three viewing depths:
1) what is seen from a distance, on overall structure usually aggregrated from an underlying microstructure
2) what is seen up close and in detail, the fine structure of the data
3) what is seen implicitly, underlying the graphic - that which is behind the graphic"
- Shrink Principle
The idea of the shrink principle is that effective graphics can be shrunk way down and still retain their information. Tufte also included an illustration from famed visualist Bertin, who demonstrated several techniques of shrinking data while maintaining the given relationship between variables. I find this beautiful.
No comments:
Post a Comment