Latent Class Output Interpretation with Dichotomised Data

Dear All,

Would appreciate some feedback regarding the output of a latent class model obtained using the 'randomLCA' package.

For background, I have six categorical variables: 'A', 'B', 'C', 'D' , 'E', 'F'. Each categorical variable takes a value 0-5 (it's a rating score for health states).

To ensure that my data is suitable for randomLCA, I dichotomised the data: 0 = {0,1,2}, 1 = {3,4,5}. However, this is where my confusion arrives regarding the output.

From the table below: is there a 98.6% probability that A is zero or one in class 1 etc?

             A    B     C    D   E    F
Class  1 0.9859 0.9935 0.9905 0.9982 0.7736 0.6984
Class  2 0.1497 0.9891 0.2405 0.9072 0.7459 0.4284
Class  3 0.5863 0.4976 0.7716 0.9532 0.5949 0.6716
Class  4 0.0081 0.0731 0.0000 0.1576 0.2067 0.2492

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.