Hellow I need to know what the difference between a pairwise t-test and a pairwise ANOVA is. Preferably explained in laymen terms.
$\begingroup$
$\endgroup$
1
-
$\begingroup$ If null hypothesis $H_0:\alpha_1=\alpha_2=\cdots=\alpha_n$ is rejected (where $\alpha_i$'s are effect of different level) then we want to check the null hypothesis $H_{0i}:\alpha_i=\alpha_{i'}$(where ($\alpha_i,\alpha_{i'}$),$i \neq i'$ are all posible pairs).Are you mean this by pairwise ANOVA? $\endgroup$– ArghaCommented Aug 31, 2012 at 4:06
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
If for example you are doing a one-way analysis of variance with 3 or more groups the pairwise comparisons of the group means are called pairwise t tests. I do not know of anything called pairwise ANOVA. Perhaps someone has made up that nomenclature to mean the same thing as the pairwise t test. The analysis of variance F test tests for an overall difference between the groups.
answered Aug 30, 2012 at 20:20