ABSTRACT
We are surrounded by data that shows an inherent variability. The issue addressed in this chapter is how we may describe this variability. Traditional statistics focuses on the proportions of observations lying in particular sections of the range of possible values. Corresponding plots and measurements focus on this theme, for example, the classical histogram. When we look at a single observed value, 235.1 say, it tells us very little on its own. For it to be meaningful, we need to know how it relates to the rest of the data. If we are told that all the data lie between 230 and 240 and that the observation is 12th out of 40 in increasing order, then we begin to get more of a picture of the situation. In the approach of this book we will be emphasising the ordered positions of data and the proportions of the data lying on either side of individual observations. This view draws to the fore a further set of plots and measures that add to the understanding of the data. They do not replace the others; they supplement them, giving a broader view of the data. Experience has always taught that the more ways one looks at a set of data the greater the understanding provided, and conversely the less likely the erroneous deduction.
