Workshop: an introduction to contemporary statistics using R
Next week I am teaching a course on statistical data analysis using R, here in the Department of Plant Sciences, University of Cambridge. It is aimed at advanced undergraduates, graduate students, and other researchers with no or little experience in statistical analysis and/or R. As such, it is much more focused on understanding the process of statistical modelling, and the importance of plotting everything, that on R programming per se. Actually, we will use R commander for most tasks. I am aware that the extraordinary potential of R is only realised by programming, but in my experience many introductory courses on R fail because people don’t really understand the statistics lying behind the code. If you have the statistical knowledge, learning how to make it in R is really much easier. Indeed, fitting a complex mixed model in R can be done with just one line of code. Understanding it, and making it right, is much more difficult.
So, I hope that a thorough explanation of statistical models for ecological data (focusing on generalised linear models as an overarching framework), without yet distracting people with particularities of the code, can be more fruitful in the long term. I specifically want to avoid people getting scared of R because of the code, or forgetting the commands just a few days after the course and finding it difficult to use R again a few months later. In the end I will try to persuade of the advantages of going beyond R commander and GUI. Hopefully, future courses will deal with more complex models and exploiting all R power by programming.
Let’s see how it goes…