I have performed a mixed-design ANCOVA on cortisol values (in SPSS), with 'time' as the within-subject variable, 'group' as the between-subject variable and 'overall treatment time' as a covariance variable. I found a significant interaction between 'time' and 'group' which I could interpret using contrasts (planned comparisons), indicating there is a interaction such that cortisol values in the experimental group increase significantly more than the control group between two time-points (pre- and post-intervention). This is a great result, however, I would also like to know wether this increase in cortisol values in the experimental group is a significant increase on itself (i.e. when only considering the experimental group, is the increase between these time-points significant?). This information (as far as I understand) can not be obtained from the information in the mixed-design ANCOVA. I therefore figured that to gain this insight, an additional ANCOVA is needed, using only data from the experimental group: with 'time' as the within-subject variable and 'overall treatment time' as a covariance variable.

My question is: am I correct in assuming this information cannot be obtained from the mixed-design ANCOVA, and if so, should I correct the obtained significance from the second test, and if so, how?

Thanks a lot in advance!


Your Answer

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

Browse other questions tagged or ask your own question.