ANOVA Design and Analysis

#1
Hello,

This is my first post so I hope my question is appropriate for the group.

I have performed an experiment where we have blood samples from 18 donors (subjects). The blood from each subject is split in 2 and subjected to a treatment (factor A, 2 levels). Each of these samples is then split into 3 sub-samples and subjected to another treatment (factor B, 3 levels). There is no replication (n=108 observations). I have treated this like a split plot design where subject is random. I then look at the factor A and factor B main effects and the interaction between these.

My question is: Is this an appropriate way of viewing the design?

I can use JMP or R in the analysis. Can someone suggest a linear model for this experiment.

Thanks in advance for the help,
Penn
 

bugman

Super Moderator
#2
Have a look at the R documentation for the packages Lme4 and nlme. I also strongly recommend that you read at least chapter one of Mixed Effects Models in S and S-Plus by Pinhiero and Bates. There is an example of rat data that looks very similar to your design and should help.