1
$\begingroup$

In an experimental study, the aesthetic value (independent var) of a picture is manipulated into (low and high quality). "Low Quality" -- to make it look less appealing; and "High Quality" to make it look "very attractive."

I want to examine the effect of "Low" and "High" Quality images on the perception of Mood (dependent var 1) and Affective reation (dependent var 2) of participants. The same participants are exposed to the low quality and high qaulity images. I presume it is considered a one way MANOVA within-subject tests design.

(i) I want to examine if a variation of aesthetic level (low or high) will affect the two dependent var - mood perception and affective reaction.

I am planning to run a MANOVA (since we have two dependent var), and report the "effect size" using Cohen's value, p-value and F-Statistics. What I am not sure is whether we need to run a regression to see how the independent var (two levels: low and high) would affect the two response var and report the value r and r-squared (variability shared among the two dependent var). Please advise how do I proceed.

Kindly confirm if I need to conduct a regression analysis or a MANOVA.

$\endgroup$
0
$\begingroup$

In the classical experimental design theory, ANOVA and regression are often implemented hand-in-hand. By referring to any traditional textbook of experimental design (for example: Experiments: Planning, Analysis and Optimization by Jeff Wu), you'll always find ANOVA followed by regression.

Fundamentally, ANOVA answers the question of whether there is a relationship between the dependent and independent variables and if yes, which of these explanatory variables are significant and should be included in the model.

On the other hand, regression "describes" the relationship between dependent and independent variables and allows you to make predictions for incoming data points that were not observed during the training phase of your model.

$\endgroup$
1
  • $\begingroup$ Thank you for the response. Do I understand that to understand the idea of "to examine the effect of " I will need to conduct both an ANOVA (to see if p-value is significant).>> Not sure how to find the relationship between IV and DV in ANOVA? Secondly, we conduct a regression analysis to postulate the coefficient r, and R-square? Kindly confirm $\endgroup$ – user39531 Sep 20 '16 at 0:40

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.