Analysis of Repeated Measurements
The course is focussed on statistical methods used when there are several measurements per experimental unit. This is often the case in medical and biological research, e.g. if patients are treated over a period of time. Appropriate analysis methods need to take account of the fact that repeated measurements on the same experimental unit are not independent.
This course requires statistical knowledge at the level of "Visualization of Lab Data" or equivalent, as well as elementary knowledge of Excel. Other statistical packages will be used as needed, but no special knowledge is required.
What will you learn ? When several measurements are taken on each observational unit (e. g. animal, patient) over a period of time, the term "repeated measures" is used. A typical situation is a clinical trial to study the effect of a drug on circulatory disturbance, when the heart rate for a group of patients is measured one, two, and three hours after the drug has been taken.
Standard statistical methods are not suitable for the analysis of such experiments, because they do not take into account the existing correlation between observations on the same unit. They must be accordingly modified. Which method is finally chosen depends essentially on the questions that the experiment should answer.
In this course you will learn how to correctly analyse repeated measures data in order to obtain meaningful results. After an introduction in graphical techniques and simple methods refined analysis of variance models and regression approaches are presented.
All methods are taught on a basic level. High value is set on practical exercises on the PC. For the introduction the software Excel is used, then other software will be demonstrated. Participants have the opportunity to analyse their own data.
Who should attend ?
- For scientists in research and development who need to analyse repeated measures data.
- Basic statistical knowledge is assumed (at least at the level of the course "Visualisation of Lab Data" ).
What topics are covered ?
| ||Simple graphical displays: Boxplot, Scatterplot, Line plot |
Comparison of two samples: t-test
Comparison of several samples: analysis of variance
Linear regression: main idea
| Elementary Methods|
| ||Comparison at individual times (time-by-time analysis) |
Response feature analysis
| Advanced Methods (ANOVA-based)|
| ||Univariate analysis of variance (ANOVA) |
Models with more than one error term (Split plot models)
Multivariate analysis of variance (MANOVA)
| Advanced Methods (Regression-based)|
| ||Models with correlation structure for observational units |
Profile curves, Growth curves
Any questions ?
- Course duration: 2 days
- Participants: max. 12 (one PC per participant)
- Costs: including complete course documentation, coffee and lunch: see registration form
- Dates: see registration form
- Further information: see contact page