So I've been reading this problem and it appears to be simple, but for some reason I'm having a difficult time, one being that I've heard of a priori POWER before, but not probability. But I digress....

Here is the problem

Group 1 has a mean of 25 and group 2 has a mean of 30 and these groups have a pooled variance of 8. I want replicate the experiment with 20 subjects in each group.

My question is: What is the a priori probability that there will be a significant difference in the the replication?

Do I need to convert the scores to a z-score or something? If I could just get some help on how to get started, that would be extremely helpful.

Thanks.