On plotting OSL equivalent doses

Abstract

This article is motivated by some recent discussion of the use of so-called ―probability density‖ plots of OSL equivalent doses. Such graphs are not advocated in the statistics literature. I try to explain what they are doing, why they are easy to mis-interpret and why they are not to be recommended. I include discussion of the meaning of dose frequency distributions, statistical research on the problem of estimating a frequency distribution when observations from it have added errors, and the possible role of dose histograms, in addition to radial plots, as data displays.