Multiple-Sample Path Analysis PreviousNext
Mplus Discussion > Structural Equation Modeling >
 Vincent Costa posted on Thursday, November 26, 2009 - 8:06 am

I have a dataset with 9 continuous observed variables assessed across 3 groups that differ in their sample size.

To add an additional level of complexity each of these variables is sampled within an individual at 4 points in time (1 s apart) so there is a degree of auto correlation. Is there an appropriate way to build this into the model?

I would like to run a multiple sample path analysis but am unable to find an appropriate example/code to get started in specifying the analysis.

Could someone point me in the right direction or provide suggestions?
 Linda K. Muthen posted on Thursday, November 26, 2009 - 8:18 am
You can do a growth model if your interest is the development of the measure across time. If not, multivariate modeling of the variables takes care of the lack of independence.

See Topic 4 starting at Slide 50 for information on multiple group growth modeling. A multiple group path analysis would be Example 3.11 with the GROUPING option.
 Vincent Costa posted on Wednesday, December 02, 2009 - 6:59 am
Thanks for the information.

With respect to using the GROUPING option in the context of Example 3.11 what other syntax would be needed to set the constraint that the two groups have the same parameters?

Also is there limit on the number of groups? Could I say specify 30 groups.
 Linda K. Muthen posted on Wednesday, December 02, 2009 - 7:41 am
Equality labels are used to constrain the parameters to be equal across groups. See Chapter 13 where multiple group analysis is discussed including the use of equality constraints.

It is possible to have 30 groups.
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message