Message/Author |
|
Mark Burton posted on Saturday, June 28, 2014 - 8:00 am
|
|
|
Hello, I am trying to describe the relationship between three dichotomous variables. I believe the best way to do this is to use hierarchical log-linear modeling where I test the three-way interaction. Do you agree? If so, can you point me in the right direction in regards to how to run this analysis in mplus? Thank you for your help. |
|
|
See the Version 7 User's Guide, ex7.15. Loglinear modeling in Mplus can be done on both observed and latent categorical variables. |
|
Mark Burton posted on Thursday, July 10, 2014 - 4:59 am
|
|
|
Thank you for your response. If I am only concerned with observed variables, do I need to add any model paramaterization information (i.e. MODEL commands)? Also, what would my analysis command be if I am not interested in mixture modeling? Thank you very much for your help. |
|
|
Have CLASSES = c (1); along with the other language of Example 7.15 if you don't want mixture modeling. |
|
Back to top |