Mondego Science

P-values are everywhere. They are in the papers we read and in the presentations we hear. We are usually also expected to include them in our own data analysis. However, the concept of P-values can be difficult and counter-intuitive. As a result, we often interpret and report P-values without fully understanding what they really do and do not represent. This is dangerous, as it leads to heaps and heaps of data misinterpretation and dodgy science. For example, people often falsely believe that the P-value represents the probability that the null hypothesis is true, and this misunderstanding leads to an over-confident interpretation of the data.

Over the past couple of decades, there has been an increasing push to use P-values less, or perhaps not at all, and the concept has become quite controversial. However, for the moment, P-values still remain ubiquitous in science. As a result, it is crucial that we know how to interpret them properly, whether we're looking at somebody else's results or our own.

This short online course provides an intuitive introduction to what a P-value is, and how it is related to the concept of statistical tests. We will also briefly discuss some of the issues with P-values, and how to make sure these don't lead to problems in our work.

This is a pre-recorded course - you can work through it at your own pace. It is made up of short video lectures, hands-on practical activities, and quizzes. As long as you complete all the activities (even if you make mistakes!), you will receive a certificate at the end. We will use an approach based on intuition and logical thinking rather than formal maths. There are therefore no pre-requisites in terms of maths or statistics skills. There is also a discussion forum where you can ask questions when-ever you get stuck, and you should get a reply quite quickly. Don't hesitate to use this possibility - these are tough concepts but if you ask for help, we will get there!

For the practical activities, you can use what-ever statistics tools you like. However, solutions will be provided in R and MS Excel.

For those of you who have taken or will take either Analysis of RNA-seq data or Modern Statistical Thinking for Biologists with Mondego Science, be aware that the content of this self-paced course is largely covered in those two longer courses. So it may be a bit redundant for you. However, it can be a good way to revise.