Description
This course is based on the book by the same name but also incorporates material from the work in preparation, Dicing with Death. It has two main objectives. The first is to take a fresh look at statistical issues commonly encountered within drug development in a way that highlights the controversies rather than trying to achieve some false consensus. The second is to provide heuristic arguments to explain as simply as possible a number of difficult issues. It is hoped that these will be useful "ammunition" for the statistician who needs to explain complex matters to fellow scientists with a lesser interest in the mathematics of statistics. The course will stress philosophical and inferential issues and will use very little algebra. The course will be less about "how to" but more about "why" and "whether to".
Day one provides an overview of statistics from four perspectives: historical, philosophical, inferential and professional. Day two will cover a number of areas of controversy: allocation of treatments; baselines and measurement of effects; p-values and tests of significance; multi-centre trials and meta-analysis. There will also be discussion sessions on each day and a chance for course delegates to share controversies, concerns and experiences with colleagues.
The course is particularly suitable for statisticians who have frequently to interact with non-statistician colleagues and explain statistical concepts to them.
By the end of the course, if nothing else, delegates should understand why it is that frequentists think it is the thought that counts, whereas Bayesians count the thoughts.
|
