Possible Duplicate:
Best practice when analysing pre-post treatment-control designs
Is it acceptable to use ANOVA to analyze pre-test/post-test differences in two groups, or would it be better to use repeated measures ANOVA? These data can also be solved using a paired-samples t-test as well, right?
If it is just one repeat (i.e. pretest and posttest) the paired t test or the Wilcoxon signed rank tests would be the appropriate tests. The choice would depend on whether or not the normality assumption is severely violated.
If you have repeated measures data on two groups, and the distribution of your dependent variable is approximately normal, with homogeneous variance, and you are interested in knowing whether the pattern of change in scores differs between groups, then a repeated measures ANOVA would be appropriate. The effect of interest would be the two way interaction between the between-subject factor (group) and the within subject factor (time), which indicates that the pattern of change over time differs between the groups. If you have repeated measures on your subjects, it is important to use a repeated measures design as this will increase the power and accuracy of your hypothesis tests.
