I have the following results from a biological experiment. I measure the size of the sample (2,4,8,16,32 or 64) and count its distribution in a population of 50 (sum of counts=50). I do those measurements in the wildtype (wt) and the experimental sample (KO) in three independent experiments (replicate 1,2,3).
Am I in a setup for a two way ANOVA with replicates? How can I proceed rigorously, with the maximum power in R as I am not sure (1) my data is normally distributed and (2) how to code that in R anyway?
Thanks!
strain Replicate size count
wt 1 2 2
wt 1 4 6
wt 1 8 11
wt 1 16 12
wt 1 32 9
wt 1 64 10
wt 2 2 2
wt 2 4 7
wt 2 8 15
wt 2 16 11
wt 2 32 4
wt 2 64 11
wt 3 2 0
wt 3 4 5
wt 3 8 17
wt 3 16 9
wt 3 32 9
wt 3 64 10
KO 1 2 0
KO 1 4 0
KO 1 8 2
KO 1 16 15
KO 1 32 23
KO 1 64 10
KO 2 2 0
KO 2 4 2
KO 2 8 2
KO 2 16 20
KO 2 32 22
KO 2 64 4
KO 3 2 0
KO 3 4 0
KO 3 8 3
KO 3 16 20
KO 3 32 21
KO 3 64 6
wtandKOand there are two responses that you observe for each replicate --sizeandcount. BTW, I am assuming by replicate you mean biological replicate. If I am right, it is just ANOVA with 1 treatment, but I don't understand the reason for 3 experiments. – suncoolsu Sep 1 '11 at 18:36wtandKOand you are taking two observations on it, size and count. Right? If yes, I will post the answer. – suncoolsu Sep 3 '11 at 0:33