0
$\begingroup$

I have an experimental (EG, N = 50) and a control group (CG, N = 210) which have to report ratings of an object at five different time points (t1-t5).

I have a direct hypothesis, that both groups will report increasing scores throughout time points and that the EG will have a higher increase in scores throughout the time points than the CG.

Intuitively I ran two separate one-way repeated ANOVAS for each group to see if ratings did differ at different time points within each group and then compared average ratings between groups at each time point with a t-test. Is this process correct or do I have to use a two-factor mixed ANOVA (2x5)? I am unsure what to do, as I have read in some discussions here, that mixed ANOVA should not be used with direct hypotheses. What do you suggest?

And I came across the option of planned contrast, but I am not sure if this would be the answer to my question. Should I conduct a Two-Factor mixed ANOVA with planned contrasts or is the analysis as i ran it okay?

Thanks for your help and have a nice weekend!

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

If I am correct then you have a 2 X 5 design with EC vs CG as between-subject factor and time (t1 - t5) as repeated, and the to-be-tested hypothesis is that the positive linear increase over time points is steeper in the EG than CG group. I see no reason why this would not be testable with a single contrast (i.e., a positive difference in slopes). If you use SPSS or STATA an similar analysis is covered in Example 12 of http://pareonline.net/getvn.asp?v=23&n=9

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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