Advanced Statistics using R

Organisers/teachers

Prof. Ido Pen (Theoretical Biology, Groningen Institute for Evolutionary Life Sciences, RUG)

Prof. Sander van Doorn (Theoretical Biology, Groningen Institute for Evolutionary Life Sciences, RUG)

Aim of the course

Learn how to master the art of advanced data analysis and graphical presentation, using the shareware program R. At the end of the course, the student is able to:

1. The student can translate specific combinations of experimental design and data into appropriate statistical models.
2. The student can program and analyze statistical models in R.
3. The student can summarize statistical analyses with appropriate tables and graphs.
4. The student can draw justified inferences and conclusions from statistical analyses.
5. The student can describe statistical methods, analyses and conclusions in a format suitable for publication.

Contents & Structure

This course offers an introduction to R and offers a review of basic statistics. Further topics that will be: general linear models (ANOVA, ANCOVA, multiple regression); generalized least squares; mixed models; generalized linear models; generalized linear mixed models; Bayesian analysis and MCMC; animal models; multivariate analysis. During the last week of the course analysis and presentation of own data set.

This course teaches advanced statistical analysis almost from the ground up. The only requirement is some familiarity with basic statistical concepts and methods, such as taught in most introductory statistics courses. Some experience with R is useful but not crucial. During the first three days, basic methods and R will be reviewed to refresh your memory. During the next two weeks, cutting-edge techniques such as GLMMs, power analyses and Bayesian MCMC models will discussed and practiced. Each day will start with a review of the exercises of the previous day, followed by lectures and new computer labs. Mathematics will be kept to a minimum, and in addition to developing analytical skills, the course also puts much emphasis on producing effective and great-looking graphs (mostly using the ggplot2 package).

The last week of the course will be dedicated to analyzing your own data, unleashing the newly learned techniques. If you have no data yet, alternative suitable data will be found elsewhere or simply created de novo with simulation models. Your methods, results and conclusions will be documented in a report which will be graded.

General Information