|
|
LCA and LPA with categorical distal o... |
|
Message/Author |
|
|
Hello, I have an model with a 4 class solution that I would like to relate to a categorical distal outcome which is comprised of 5 categories. I had thought to handle this using DCAT. Mplus returns the chi-square solution as expected with 4 df when comparing each group. However, I am interested in whether it is possible to do something like pairwise comparison. That is, I would like to know whether or not Group 1 significantly differs from Group 2 on category A, B, C, but D or E. Is it possible to do this? I apologize for putting this in a new thread, but I had not seen this question posted elsewhere. Thank you! |
|
|
It is possible to do this using the manual 3-step approach, see Section 3 https://www.statmodel.com/download/3stepOct28.pdf or the manual BCH method, see Section 3 https://www.statmodel.com/examples/webnotes/webnote21.pdf where in the last estimation step you can specify the needed test in Model Test or you can form as new parameters the difference between probabilities of the specific categories, see pages 772-773 in the User's Guide. |
|
Back to top |
|
|