Parallel processes LGCM interpretation PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Anamaria Brailean posted on Thursday, June 08, 2017 - 11:59 am
Hello,

Considering an LGCM model with two scenarios:

Scenario 1: the mean intercept is positive and the mean slope is negative.
Option a) the correlation between the mean intercept and mean slope of the outcome is positive. Does this indicate that those with higher levels of the intercept decline less?
Option b) the correlation between the mean intercept and mean slope of the outcome is negative. Does this indicate that those with higher levels of the intercept decline more?

Scenario 2: the mean intercept is positive and the mean slope is positive.
Option a) the correlation between the intercept and slope of the outcome is positive. Does this indicate that those with higher levels of the intercept increase more?
Option b) the correlation between the intercept and slope of the outcome is negative. Does this indicate that those with higher levels of the intercept increase less?

Would the above interpretations apply similarly when I look at:
1) The correlation between the mean intercept of an outcome and the mean slope of the same outcome
2) The effect of the mean intercept of one outcome on the mean slope of another outcome
3) The effect of a baseline covariate on the slope of an outcome
4) How does the interpretation of a significant association between intercept and slope change when the mean slope is not significant?

Many thanks for your advice!
 Bengt O. Muthen posted on Saturday, June 10, 2017 - 12:16 pm
Yes on all the questions under the 2 scenarios.

And Yes on 1) - 3). The answer to 4): It doesn't change.
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: