2
$\begingroup$

I am designing my study, but I am a little stuck in which test I eventually should use. I have a between-subject design with 6 conditions (let's say A, B, C, D, E, F), with each having 6 responses (let's call these a, b, c, d, e, f), on a 7-point scale:

IV (condition) -> DV
A -> a, b, c, d, e, f
B -> a, b, c, d, e, f
C -> a, b, c, d, e, f
D -> a, b, c, d, e, f
E -> a, b, c, d, e, f
F -> a, b, c, d, e, f

So, each condition has 6 dependent variables. I don't want to test difference between the conditions, but only the effect of the condition on the different DV's. For example, when someone is in condition A, I would expect them to score higher on DV a than on DV b, c, d, e, and f. When someone is in condition B, I would expect them to score higher on another DV, and so forth.

Later on, I expect also some moderator effects, but I guess that's only adding interaction terms to the model.

Can I use separate tests for each condition? Or should I test it in one big model? What kind of test would be most suitable for this. Most tables that should help you choose a statistical test only show options with 1 DV.

$\endgroup$
3
  • $\begingroup$ Do you expect what happens under each condition to be independent of what happens under every other condition? $\endgroup$
    – P Schnell
    Mar 17 '14 at 21:52
  • $\begingroup$ Yes, I expect what happens under each condition to be independent. $\endgroup$
    – user40037
    Mar 17 '14 at 22:12
  • $\begingroup$ Would a MANOVA be the right choice here? $\endgroup$
    – user40037
    Mar 18 '14 at 8:06
1
$\begingroup$

Can you please clarify you variable and their measurement scale?

6 conditions (let's say A, B, C, D, E, F) represents 6 input variables or 6 categories of one input variable.

Each having 6 responses (let's call these a, b, c, d, e, f), on a 7-point scale. Is it one outcome variable measured on 7-point likert scale or 6 outcome variables (each measured on 7-point likert scale)

Techniques which can be used:

6 inputs, 6 outcomes - multivariate multiple regression or MANCOVA

1 input (with 6 categories), 6 outcomes - multivariate regression or MANOVA

6 inputs, 1 outcome - multiple regression

1 input (with 6 categories), 1 outcome - simple regression with dummy variable or ANOVA

$\endgroup$

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.