I have two different chemicals applied in 4 different concentrations into medium and control with no chemicals inoculated with 3 replications (Control y1, y2, y3; Chemical A 10 μg y1, y2, y3; Chemical A 20 μg y1, y2, y3...; Chemical B 10 μg y1, y2, y3...). I have compared differences between control and each concentration of each chemicals with one-way anova, but I also need to compare the differences between control group and treatments A and B. I was suggested to use two-way anova, but in this case the number of observations between treatments is not equal so the comparison cannot be done. Is there any way to compare treatments in general to control?
Go to youtube and search Amos. it is free.
you can choose PLS which is difficult for you.
The most easiner use SPSS.
Hello Elena,
So, you have 9 different "conditions" being compared: 1 control, 4 concentrations of chemical "A" and 4 concentrations of chemical "B".
If you envision this as a one-way design (with 9 groups/levels), then most of the relevant tests you might wish to conduct could be represented as a set of a priori contrasts.
Here are some examples:
1. Control vs. all chemical conditions: (8, -1, -1, -1, -1, -1, -1, -1, -1)
2. Chemical "A" vs. chemical "B": (0, 1, 1, 1, 1, -1, -1, -1, -1)
3. Control vs. lowest concentration of "A": (1, -1, 0, 0, 0, 0, 0, 0)
4. Control vs. highest concentration of "B": (1, 0, 0, 0, 0, 0, 0, 0, -1)
5. Lowest concentration vs. highest concentration: (0, 1, 0, 0, -1, 1, 0, 0, -1)
where the coefficients in parentheses represent the weights to apply for the respective contrasts, presuming that group 1 = control, group 2 = A/lowest concentration; group 3 = A/2nd concentration; group 4 = A/third concentration; group 5 = A/highest concentration; group 6 = B/lowest concentration..etc., and group 9 = B/highest concentration.
Good luck with your work.
