# How to specify a significance test for intercept and slope between multiple latent groups

I have done a Latent Profile Analysis on 34 symptom-distress-items at beginning of psychotherapy treatment. I came up with 5 latent groups. The entropy of the model was 0.779

The average Latent Class Probabilities for Most Likely Latent Class Membership (Row) by Latent Class (Column) was:

       1        2        3        4        5

1   0.933    0.015    0.026    0.026    0.000
2   0.005    0.836    0.037    0.117    0.005
3   0.020    0.048    0.897    0.035    0.000
4   0.008    0.079    0.019    0.874    0.019
5   0.000    0.012    0.000    0.037    0.951


I have also measured treatment progress every session.

I would like to use the grouping variable from the LPA in a growth model to investigate if the 5 latent groups differ in distress at intake (intercept) and rate of improvement (slope).

I used R to calculate this using a mixed model and a contrast matrix but I did not consider the measurement insecurity of the latent groups. I figure that since the error was between 0.84-0.95 the results I got would hold anyway. My question is : If this assumption is correct?

• What do you mean by "on another measure?" – Jeremy Miles May 20 '16 at 20:11
• debugging questions are off-topic here and should be asked on Stack Overflow instead – Antoine May 20 '16 at 21:04
• I think this is a general stat question which is on-topic here. I've edited the question to make it more general and not specific to Mplus. I think the answer is "likelihood ratio test" as well, but I'm waiting for clarification from OP as to exactly what they're asking. – Jeremy Miles May 20 '16 at 22:12
• Please register &/or merge your accounts (you can find information on how to do this in the My Account section of our help center), then you will be able to edit & comment on your own question. – Silverfish May 22 '16 at 9:52
• @CJ_U, I don't understand the question any more. – Jeremy Miles May 24 '16 at 19:45