|
|
 |
| Bivariate Process of Mixtures |
 
|
| Message/Author |
|
|
| Chanes posted on Thursday, October 13, 2005 - 12:20 pm
|
|
|
Dear Dr Muthen
I am trying to make sense of the following model, where i want to generate a two-way table for cross-classifying two outcomes based on mixtures. In a previous step i have identified for each outcome the number of classes (2 for each outcome). I try to understand the solution to this model that i enclose. First, in the output a latent class pattern is reported for each combination of classes (e.g., 1-1, 1-2, 2-1, and 2-2). I know that a solution with both outcomes simultaneuosly will be different from the analyses for separate outcomes. However, the counts do not match with the solution that i get when the analyses were conducted separatedly and i was expecting some resemblance in the results from both analyses. I don't know if MPLUS is redefining internally the labels for the classes so that i've got the classes missplaced when the model inlcuded the two outcomes simultaneously. This is very confusing.
MODEL:
%overall%
ca#1 WITH cb#1;
i1 s1 | af14@0 af15@1 af16@2 af17@3 af18@4;
i2 s2 | mf14@0 mf15@1 mf16@2 mf17@3 mf18@4;
i1 WITH i2-s2@0;
s1 WITH i2-s2@0;
MODEL ca:
%ca#1%
[i1*3.162 s1*.168];
%ca#2%
[i1*3.086 s1*-.044];
MODEL cb:
%cb#1%
[i2*2.526 s2*-.005];
%cb#2%
[i2*2.918 s2*.254];
Thanks |
|
|
|
|
| Please send the full outputs from the two analyses conducted separately and the output with two categorical latent variables along with the data and your license number to support@statmodel.com so we can take a look at it. |
|
|
| Back to top |
|
|
