Effects of trends on levels PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 JM posted on Tuesday, February 24, 2009 - 1:15 pm
Hi, I just want to be sure I'm understanding the output correctly. I'm conducting a parallel process grwth model. The means for level1 and level2 are positive and significant, and means for trend1 and trend2 negative and significant. This suggests that growth in both process drops overtime, right? Now, when I regress trend2 ON level1 I get the following:
est: .07 and est/se: 2.16. Does it mean that the higher the initial level1 the faster the decrease in trend2? What if the est/se was a negative score, does that mean higher level1 suggests slower decrease in growth overtime?
 Bengt O. Muthen posted on Wednesday, February 25, 2009 - 3:14 pm
This situation is often a source of consternation. Here, it means that when a person's level1 value also tends to be high, his/her trend2 slope value tends to be high. Note that the latter "high" does not mean the trend2 slope is necessarily positive. Given that the trend2 mean is negative it is perhaps more likely that for a person with high level1 value, the trend2 slope is less negative, that is, the person declines less steeply.

If the 2 growth factors have a negative relationship, the higher the level1 the lower trend2, that is, in this case the steeper negative the trend2 slope.
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: