Comparing regression slopes in a with... PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Katharina Lotz-Schmitt posted on Thursday, June 21, 2012 - 5:25 am
Hello!

I am struggling with the following problem:

I want to compare two regression slopes in a 1 x 2 within subject design.

My question is: Is the regression slope of y1 on x1 different to the regression slope of y2 on x2?

I'm not very good with MPlus, I've just started using it. How can I compute this analysis?

I would be very thankful for help!

kathalo
 Linda K. Muthen posted on Thursday, June 21, 2012 - 2:55 pm
y1 ON x1 (p1);
y2 ON x2 (p2);

MODEL TEST:
0 = p1 - p2;

This assumes the variables are on the same metric or comparing the coefficients make no sense.
 Katharina Lotz-Schmitt posted on Monday, July 09, 2012 - 4:47 am
Thank you so much!

I have got one more uncertainty:

I thought it would be reasonable to control for y2 and x2 ind the first regression and for y1 and x1 in the second regression.

So I get:

model: y1 on x1 (p1)
x2
y2;

y2 on x2(p2)
x1
y1;

But if I do so, a

"THE DEGREES OF FREEDOM FOR THIS MODEL ARE NEGATIVE. THE MODEL IS NOT IDENTIFIED. NO CHI-SQUARE TEST IS AVAILABLE. CHECK YOUR MODEL."-

Error occurs.

What is wrong?

Does it make more sense just do control for the other predictor, instead of controlling for both criteria and predictor?


Thank you for your help,

Katharina
 Linda K. Muthen posted on Monday, July 09, 2012 - 12:00 pm
Please send the output and your license number to support@statmodel.com.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: