

Logistic model with LC variable as pr... 

Message/Author 

Jon Heron posted on Friday, August 18, 2006  8:15 am



Hi, I've derived a 5class LCA based on 5 binary observed variables I would like to use this 5level factor as a predictor in a logistic regression model. If I include this binary outcome as a 6th variable in my usevariable list then I get a set of odds ratios as if the classvariable were the outcome and my binary variable the predictor, which is equivalent to the unadjusted estimate I seek thanks to odds ratios being reversible. However, I want to incorporate other covariates into this model as confounders so then I can no longer employ the method above. How can I model a multivariable logistic model in this way? many thanks Jon 


The covariates can be specified to influence both the latent class variable (with 5 classes) and your binary outcome (the 6th outcome). From this estimated model you can compute odds ratios. 

Jon Heron posted on Monday, August 21, 2006  12:15 am



I see, many thanks J 

Back to top 

