Introduction to Experimental Design
Training session with Dr Helen Brown, Senior Statistician, at The Roslin Institute, January 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:
‘Crossover’ design
-Several Interventions given to same individuals or units in a random order
---Eg mice preference for fluid types tested in turn
-Appropriate when interventions do not have permanent effect
-Often reduces variation in data
---Interventions compared ‘within individual’ or ‘within unit’
-A ‘within subject’ design

‘Crossover’ design: Example
-Experiment to see whether mice preferred tastes of saccharin or salt
-4 cages of mice
-Each cage studied with salt or saccharin fluid
-Outcome = % test fluid drunk in cage compared to water
-Cages randomised to ‘test fluid order’
-Note: If several treatments there will be many potential orders to randomise to

‘Repeated measures’ design
-Same mice observed at each time point
-Experimental unit = mouse
---treatments randomised to mice, NOT time
-Define objectives in advance, eg
---test treatment effect at each time point?
---test treatment effect at a particular time point?
-Care with statistical analysis
---eg do not assume 24  independent observations