Advanced Data Analysis Workshop Two day Workshop

A lot of statistical applications involve count, binary (success or failure, dead or alive, pass or fail) or categorical (never, sometimes, often) responses. In this case the standard linear regression is insufficient, and generalised linear models (GLMs) should be used instead.

Furthermore, while the standard linear regression assumes independent observations, in reality, there is often some kind of further structure in the data. Your survey subjects might be grouped into workplaces, classes or families. Your trees might be planted on plots within sites. Your soil samples might be collected from the sediments of different rivers. Or you might have repeated measures: taking the blood pressure of a patient before and after the treatment, or simply several times over the course of the study. All of these situations need to be accounted for. And the mixed effects (aka hierarchical) model can take care of that.

In this, advanced data analysis workshop, you will learn when to apply GLMs and mixed-effects models. And you will learn how to apply them to various situations using R.

Participants are expected to be familiar with the basics of classical statistical inference, linear regression and ANOVA, as well as have a basic working knowledge of R to the extent covered in the introductory Data Analysis Workshop.

The workshop will take place on Wed-Thu September 4-5. Each day will consist of a seminar session in the morning 10-12, and a lab session in the afternoon 1-3pm. To take a full advantage of the workshop, the participants are expected to commit to the entire two days. Because number of places is limited, please be considerate and only register if you are actually intend to come.

Wednesday 4th September Morning Session-10-12pm 

Ernest Rutherford  464 

Wednesday 4th September afternoon Session 1-3 pm 

Ernest Rutherford  465

Thursday 5th  September Morning Session-10-12pm 

Ernest Rutherford  464 

Thursday 5th September afternoon Session 1-3 pm 

Ernest Rutherford  465

If you have any questions, please contact the Statistical Consulting Unit at statsconsulting@canterbury.ac.nz. If you want to discuss your project specifically, just make an appointment.