Growth: multi-group pretest/postest &... PreviousNext
Mplus Discussion > Growth Modeling of Longitudinal Data >
Message/Author
 Ilona posted on Friday, September 03, 2010 - 11:54 am
Hi,

I would like to analyze data with these conditions and sample sizes with nonrandom group assignments:

Sample (in a single grade level):
Clusters of about 20 students per classroom
Clusters of about 4 classrooms per school
Clusters of about 5 schools per district
=Total of about 400 students per district

There are two similar districts:
District 1 gets treatment (400 students)
District 2 is control (400 students)

All students in both districts get a pretest and posttest.

Purpose 1: Want to see if growth from pretest to posttest is greater for treatment district than control district overall.

Purpose 2: Want to see if growth from pretest to posttest is greater for treatment district than control district. Wish to examine this effect separately by race of students (Black/White).

And if possible:
Wish to examine differences in growth trajectories based on initial pretest score. (I.e., wish to view growth trajectory differences by pretest scores overall, and pretest scores by race.)

------------------------------------
Please advise on best method, plus references to guide me with this.

Thank you!
-Ilona
 Linda K. Muthen posted on Friday, September 03, 2010 - 3:24 pm
This is a general analysis question that is more appropriate for Multilevelnet. See the Raudenbush and Bryk book as a reference.
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: