# Two samples t-test and ANOVA

B) When doing a two way ANOVA After calculating the sum of squares column(between), the sum of squares error(within), and the sum of squares block. 1. Are there any other sum of squares? 2. What is the degree of freedom for MSB (block)? I know in one way ANOVA degree of freedom for MSC and MSE is k-1 and n-k respectively. 3. Also in one way ANOVA F-ratio= MSC/MSE, so considering there are more sum of squares in 2 way ANOVA how do I calculate the F-ratio after all the mean of the sum of squares have been obtained?

To elaborate (I forgot to add I'm not using Spss, I'm trying to calculate this by hand

So first I find the SSC then the SSE followed by the SSB(block) and the SST(total). Are there any more sum of squares? Then when calculating the mean of the sum of squares

MSC = SSC/k-1

MSE = SSE/n-k

MSB= SSB/? <--- what is the Df here?

Also, After this step how do I compute the F-ratio considering there are more sum of squares than in one way ANOVA? In one way ANOVA F-ratio= MSC/MSE, how about for two way ANOVA? Briefly put, I need you to tell me how many sum of squares do I have to compute, what are they, and the step in 2 way ANOVA after you compute the sum of squares. It's a bonus if you also add the steps of repeated measures ANOVA.

• You need to edit this to expand. For instance you say "Or is there another formula?" but we do not know what formula you have in mind so we cannot say whether there is another one. Similaryl for question B. – mdewey Feb 2 '17 at 14:23
• I have edited it. Hope this is more clear. – future.shrink Feb 2 '17 at 14:56