More Graphics

Abstract

The importance of displaying data graphically prior to more formal analyses has already been mentioned in Chapter 2. And in Chapter 4 the fundamental diagram for exploring relationships between pairs of variables, the scatterplot, was discussed. In this chapter, the use of a number of more ambitious graphical displays will be considered. Much of the material originates from the ‘visualization’ philosophy of William Cleveland as outlined in his two excellent books, The Elements of Graphing Data (Cleveland, 1985) and Visualizing Data (Cleveland, 1993). The approach can be summarized by the following two quotations from the most recent of these:

Visualization is critical to data analysis. It provides a front line of attack, revealing intricate structure in data that cannot be absorbed in any other way. We discover unimagined effects, and we challenge imagined ones.

There are two components to visualizing the structure of statistical data — graphing and fitting. Graphs are needed, of course, because visualization implies a process in which information is encoded in visual displays. Fitting mathematical functions to data is needed too. Just graphing raw data, without fitting them and without graphing the fits and residuals often leaves important aspects of data undiscovered.