Course Survival Analysis

Organisator Biometris, Production Ecology and Resource Conservation (PE&RC)

wo 29 april 2015 tot do 30 april 2015 09:00

This 2-day course gives an introduction to survival analysis methods and how to apply these to biological data. After following the course, participants should be able to apply basic survival analysis methods to biological data, perform a simple analysis on a data set, and explore more advanced survival analysis literature themselves.

All relevant information can be found using the link on the right hand side.

TO REGISTER FOR THIS COURSE: Do note that this course is only offered once every two years! The deadline for early bird registration is April 1st.

Course Contents

Survival analysis was originally developed in medical statistics, to analyse effects of factors on survival times, (hence its name). The factor are usually referred to as covariates. The purpose of such analyses is for instance to find out whether certain treatments reduce the incidence of death. The methods are especially designed to take so-called "censored data" into account, when death occurs due to other causes than the studied disease or when patients survive until the end of the study.

Survival analysis methods can be applied in a variety of biological contexts, whenever data consist of time until occurrence of a certain event for individually followed subjects. Examples are many types of behavioural data, times until recapture, latency data. Here, we give an example of two caterpillars that are placed on a plant at the start of the observation period. One of the two caterpillars is leaving during the observation period resulting in a "failure" time or in this case time until leaving the plant. The other caterpillar stays on the plant during the whole observation period of 5 weeks. This results in a so-called "censored" observation: the information that this observation contains is that the real "failure time" is longer than 5 weeks.

The course consists of two days during which we will introduce the concept of survival analysis and show how to apply the methods to biological data. Main subjects are how to handle censored data, estimation of Kaplan-Meier survivor curves, the Log-Rank test for testing differences between survival curves, and Cox' regression model for estimating and testing effects of covariates (and interpretation of the statistical results). Lectures will be alternated with time for practicals, where participants apply the methods with statistical software. During the second day participants will analyse a data set (a provided one or their own data). Furthermore, participants are invited to bring their own data sets for discussion and (when feasible) analysis.

Book: we advise the participants to buy the book "Survival analysis: a self-learning text (2nd edition)" by David G. Kleinbaum, Mitchel Klein. This book is published by Springer and can be ordered by or

It is advised to read the first two chapters of the book in advance.