t test or anova

Question

I have a pandas data frame of the form:

r1    r2    r3    r4    r5

0    1    12    0    4
1    1    2    9    2
32   5    0    0    0
12   14   3    1    23
0    2    43    5    2
9    3    5    1    1
0    0    0    0    1
1    0    0    0    0

And I want to check if any column: r1, r2, r3, r4, r5 significantly differs from any of the other. Should I do a t test or an anova? And how would I set it up for the computation?

Answer 1

This is typical statistics problem. When you have multiple 'classes' that you assume are normally distributed you first run an ANOVA. Then, IFF (if-and-only-if) the ANOVA is significant, then run post-hoc pairwise t-tests with an appropriate correction (e.g. Bonferroni).