SEM with clustered/multilevel model PreviousNext
Mplus Discussion > Multilevel Data/Complex Sample >
 Erkan Er posted on Saturday, October 10, 2015 - 9:17 pm
I have collected data with the same instrument three times (1 month after each other) from the same population. When I test model with each data set I obtain a good model fit, with some minor differences across data sets (for example, some relationships are not significant in one data set).

Then, I tried doing clustered/multilevel SEM analysisÁ Mplus runs the code (below) successfully. I obtain three result sets for three separate models (actually they are the same model), and all the relationships among the latent variables in each model are NS. I tried clustered/multilevel model because the parameters are all estimated using all of the data, so should have smaller standard errors and I should get more power to detect your effects.

Does my logic make sense? Do you have any idea why I receive bad results when I do clustered/multilevel SEM? Any help is appreciated. I can also send my data + code if needed.


NAMES = X1-X41;
GROUPING IS X3 (1 = 1 2 = 2 3 = 3);
cluster is X1;
TYPE IS complex;

REL BY X4 X5 X7 X10;
TS BY X12-X15;


model 1:

model 2:
!the same

model 3:
!the same
 Bengt O. Muthen posted on Sunday, October 11, 2015 - 9:34 am
It is not clear to me if the analyses of each data set also use Type=Complex. And it is not clear what your cluster variable X1 represents, or what you mean by using all of the data.
 Erkan Er posted on Sunday, October 11, 2015 - 9:49 am
Here is the full code:

X1 is the student ids.

X3 is the time identifier, 1 for the first data set, 2 for the second, and 3 for the third.

All data means all data sets: collected three times from the same population by using the same instrument.

Do you think I should run separate analysis for each data set to validate my model? In this way, I can show that model fits well for three data sets.

Or should I use clustered/multilevel model (as I am attempting) to obtain a better prediction power with less measurement error?
 Bengt O. Muthen posted on Monday, October 12, 2015 - 5:56 pm
So when you say "three times from the same population", do you mean that it is the same people or different people at the different time points?

If X1 is the student id, what is clustered within student?
 Erkan Er posted on Monday, October 12, 2015 - 6:06 pm
It is the same population. Actually, nothing is clustered within student. I just meant to group students according to time points: first data collection (1), second data collection (2), and third one (3).
 Bengt O. Muthen posted on Monday, October 12, 2015 - 6:13 pm
Same population and also the same sample of individuals?
 Erkan Er posted on Monday, October 12, 2015 - 6:19 pm
There are slight differences since some participants did not complete the questionnaire in all data collections. I tested the code with a single model (without cluster or grouping), and I get good results, here is the code with single model:

Here is my complete code with cluster & grouping:

Do you have any idea why is this happening?
 Bengt O. Muthen posted on Wednesday, October 14, 2015 - 2:18 pm
If you have largely the same individuals at each of the 3 time points you should not treat time point as group in a multiple-group analysis. Multiple-group analysis needs to have independent observations from the different groups, that is, no same people.

Instead, those data should be analyzed as a longitudinal model, that is, with p variables measured at T time points you analyze a model with p*T variables.
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