Perhaps this is an easy issue, but I wanted to throw it out anyway.

I'm working on a dataset where there are two groups, both having pre- and post-test scores. I want to see if one group has a significantly greater improvement gain from pre- to post-test than the other.

The way I'm looking at it is as follows: to compute a new variable in SPSS which is the difference from pre- to post-test and then run an ANOVA on this computed variable to assess between group differences in improvement. Is this the best approach or should I be analyzing the pre- and post-test scores directly?

perhaps you can use the difference scores and use t-test?

Perhaps this is an easy issue, but I wanted to throw it out anyway.

I'm working on a dataset where there are two groups, both having pre- and post-test scores. I want to see if one group has a significantly greater improvement gain from pre- to post-test than the other.

The way I'm looking at it is as follows: to compute a new variable in SPSS which is the difference from pre- to post-test and then run an ANOVA on this computed variable to assess between group differences in improvement. Is this the best approach or should I be analyzing the pre- and post-test scores directly?


I suggest use the pre and post test scores and then do a hypothesis test on the POWER of the study.

I suggest use the pre and post test scores and then do a hypothesis test on the POWER of the study.

How would you do that? The power analysis that is. I am also interested in this...Thank you.