I am looking at two-factor factorial data where the response variable is number of cancer cells after an experiment is performed.
The two-factors are:
Factor A, which has a = 2 levels: "Conventional" (which is basically a control) and Dye Added
and
Factor B, which is the concentration of a drug Cisplatin, which has b = 6 levels: 0, 0.15, 1.5, 15, 150 and 1500 (ng/ml)
Data is given below, but it is unclear to me whether it is referring to treatment means for each factor-treatment combination or if each treatment only has one replicate:
I am asked to analyze the data above for the effects of concentration and dye.
My initial idea was to use three contrasts and test for significance:
Main Effect Contrast of Factor A
Main Effect Contrast of Factor B (using linearity in dose)
Interaction Effect between Factors A and B
However, since no sample size is given, I'm not sure how to test for significance since I have no way of computing the MSE. Are there any other ways I can test for significance of the contrasts or test for the interaction?