- #1
bonfire09
- 249
- 0
I have to test treatments B,C,D and E where ##H_0:\mu_B=\mu_C=\mu_D=\mu_E## vs ## H_1: not H_0##. Use ##\alpha=0.05## given this problem.
Now ##\mu_A## is not included in the hypothesis so I am trying to figure out how to go about this problem. I was thinking of using orthogonal contrasts and this is what I came up with.
##C_1-C_4## are my 4 orthogonal contrasts (all pairwise) for the ##5## treatments. The part that confuses me is what to do with ##\mu_A## since it was never included in the null hypothesis? Also I'm guessing that I test C1-C4 to see if I can violate the null hypothesis but what if each of my contrasts support the null hypothesis? There should be a better way of doing this problem I think. I am not sure if I was to ask this question here, but since this deals with statistics I asked it here. Thanks.
Edit- I just noticed that my contrasts are not linear independent. I will have to come up with new ones.
Now ##\mu_A## is not included in the hypothesis so I am trying to figure out how to go about this problem. I was thinking of using orthogonal contrasts and this is what I came up with.
##C_1-C_4## are my 4 orthogonal contrasts (all pairwise) for the ##5## treatments. The part that confuses me is what to do with ##\mu_A## since it was never included in the null hypothesis? Also I'm guessing that I test C1-C4 to see if I can violate the null hypothesis but what if each of my contrasts support the null hypothesis? There should be a better way of doing this problem I think. I am not sure if I was to ask this question here, but since this deals with statistics I asked it here. Thanks.
Edit- I just noticed that my contrasts are not linear independent. I will have to come up with new ones.
Last edited: