I have used a 3 step approach in MPLUS to first estimate a latent class model (LPA) and then used the most likely class (taking into account the misclassification) and put them in a growth model.
I used a 5 class solution for the LPA.
I would like to test if there are significant differences in intercept and slope between the 5 groups but I can’t figure out how to specify this in the model?
Help would be much appreciated!
Below are the code that I used.
TITLE: Step 3 approach
DATA:
FILE = 5_Classes.dat;
VARIABLE:
NAMES = I1-I34 Core1-Core29 p1-p5 n;
MISSING=*;
Classes = c(5);
nominal=n;
usevar are Core1-Core10 n;
ANALYSIS: TYPE = MIXTURE; STARTS = 0;
Model:
%OVERALL%
i s | Core1@0 Core2@1 Core3@2 Core4@3 Core5@4 Core6@5 Core7@6 Core8@7
Core9@8 Core10@9;
%C#1%
[N#1@6.287];
[N#2@0.906];
[N#3@1.947];
[N#4@2.912];
i s | Core1@0 Core2@1 Core3@2 Core4@3 Core5@4 Core6@5 Core7@6 Core8@7
Core9@8 Core10@9 ;
%C#2%
[N#1@-1.226];
[N#2@3.803];
[N#3@1.134];
[N#4@1.825];
i s |Core1@0 Core2@1 Core3@2 Core4@3 Core5@4 Core6@5 Core7@6 Core8@7
Core9@8 Core10@9;
%C#3%
[N#1@9.238];
[N#2@10.417];
[N#3@13.727];
[N#4@10.625];
i s | Core1@0 Core2@1 Core3@2 Core4@3 Core5@4 Core6@5 Core7@6 Core8@7
Core9@8 Core10@9;
%C#4%
[N#1@-0.997];
[N#2@1.008];
[N#3@0.245];
[N#4@3.536];
i s | Core1@0 Core2@1 Core3@2 Core4@3 Core5@4 Core6@5 Core7@6 Core8@7
Core9@8 Core10@9;
%C#5%
[N#1@-5.724];
[N#2@-4.079];
[N#3@-13.746];
[N#4@-2.958 ];
i s | Core1@0 Core2@1 Core3@2 Core4@3 Core5@4 Core6@5 Core7@6 Core8@7
Core9@8 Core10@9;
