Different results by adding a third group?

Uschi

New Member
#1
Hi,

In my study I first only compared two different groups (1 and 2) on a number of variables. For one variable in particular, these groups were significantly different from each other (found that on t-test and ANOVA).

Now I added a third group to the sample. In each of the groups N=21, so they are all of the same size.

When I do an ANOVA now, I find a main effect of group but when I look at the post hoc analysis, I only find a significant difference between group 1 and 3 and 2 and 3. I do not find a significant difference between group 1 and 2 anymore, which was the case when I did the analyses with only two groups.

The means for group 1 and 2 are less different from each other than the means of group 1 and 3 and 2 and 3. Does this mean that with two groups there was a significant small difference but by adding the third group which are A LOT more different from the other two groups the difference between the first two groups is not really 'relevant' anymore?

I hope my problem is clear. Is this a normal thing?