Introduction to Analysing Repeated Measures Data
Training session with Dr Helen Brown, Senior Statistician, at The Roslin Institute, March 2016.
************************************************
These training sessions were given to staff and research students at the Roslin Institute. The material is also used for the Animal Biosciences MSc course taught at the Institute.
************************************************
*Recommended YouTube playback settings for the best viewing experience:  1080p HD
************************************************
Content:
Example – Clinical Trial
- Hypertension trial of 3 treatments
- Four fortnightly post treatment visits

Analysis Model – Fixed effects
- Fixed effects:
--- Treatment
--- Time
--- Treatment.Time
--- Baseline (pre-treatment)
- Treatment.Time interaction is optional:-
Included
--- Treatment differences available at each time point
--- Overall treatment effects are an unweighted average of treatment effects at each time point
Omitted
--- Treatment differences not available at each time point
--- Treatment differences are a weighted average of treatment effects at each time point
----- Weights relate to variances at each time point (and hence to number of observations)

Analysis Models – Covariance patterns considered
Model Covariance pattern
1. Compound symmetry
2. First order autoregressive
3. Toeplitz
4. General
5. Separate compound symmetry for each treatment group
6. Separate Toeplitz pattern for each treatment group