I performed a two-way ANOVA. I have one control group and 2 experimental groups.

One experimental group received one strategy, the other group received another kind of strategy. Now, I want to know which strategy had a better effect on the group.

Which table in the two-way ANOVA will help me understand which strategy had a better effect? (I used pre-test and post-test)

  • $\begingroup$ So the two factors in your ANOVA are pre/post and treatment, for a 2 x 3 design? $\endgroup$ – atiretoo Jul 2 '12 at 22:42
  • $\begingroup$ yes i have pre -post test and treatment $\endgroup$ – zara Jul 3 '12 at 18:20

The overall F test tells you if there is a significant difference between the groups but significance does not tell which groups differ. To do that you need to follow the test up with contrast comparisons looking at the pairwise differences between the means. You specify the contrasts and the program should compute the $p$-values for the corresponding $t$-tests. You should do some adjustment to the $p$-value because you did more than one test. A simple Bonferroni bound may be adequate.

  • $\begingroup$ so i should perform independent t-test. tnx $\endgroup$ – zara Jul 3 '12 at 18:19
  • $\begingroup$ Yes but also take account of the need to adjust the p-values for mutliplicity. $\endgroup$ – Michael R. Chernick Jul 3 '12 at 18:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.