Message/Author 

Jon Heron posted on Thursday, July 29, 2010  2:54 am



Dear Bengt/Linda, I'm working on teaching material in which I derive latent classes from six repeated binary outcomes and then demonstrate the different ways one might relate these classes to a distal outcome (cts) and a covariate (categorical) This is motivated by Clark, S. & Muthén, B. (2009). Relating latent class analysis Petras, H & Masyn, K. (2009). General growth mixture analysis I compare modal class (entropy ~0.9), pseudoclass (Mplus + also Stata), probweighting in Stata and finally a 1 stage approach (although I don't fit covariate AND outcome at the same time). Outcome: Good agreement in estimates across the board, including the 1stage where the measurement model is pretty unaffected by the inclusion of the cts measure. Probweighting much lower SE's but the rest do OK. Exposure: 2stage results all in good agreement. Probweighting slightly lower SE's. 1stage model only leads to minor tweaks in terms of class profiles and class distribution. However, regression estimates much higher than all the 2stage results. I am a little stuck now. I usually teach probweighting but I want to show the options and that conclusions could depend on one's choice. Do I conclude that (in this ex) all 2stage models underestimate parameters or perhaps my model may be misspecified? many thanks 


It would be misspecified if the class formation changes much when adding the covariates and doing a 1step run. 

Jon Heron posted on Monday, August 02, 2010  1:25 am



Thanks Bengt, that begs the question which I expect is v hard to answer  what is an important change in class formation? From examining class profiles alone it appears (at least to me) that very little has changed  distribution and interpretation of classes is maintained. Had I not compared my results with pseudoclass I would not have suspected a problem. :S 


I would include direct effects from the covariates to the latent class indicators (in addition to c ON x). So u1 on x1xq, then u2 on x1xq, etc to search for significant direct effects. And including such effects, then again see if the class formation is similar. 

Jon Heron posted on Wednesday, August 04, 2010  12:36 am



Thanks Bengt that sounds like a test for measurement invariance / uniform DIF within a MIMIC model. Is that essentially what we have here, albeit with a categorical latent var? 


That's right. I think it would often be useful. 

Jon Heron posted on Wednesday, August 04, 2010  9:46 am



That's good to know. A stepwise approach has shown bias i.e. a main effect for two of my six items. Class distribution has changed slightly and covariate effects on C have changed considerably, in the direction of those obtained from pseudoclass but agreement between 1stage and 2stage is still poor. I feel these new biasadjusted estimates are more valid than the original 1stage estimates and if anything I am surprised that they moved *towards* the 2stage results since they too are not adjusted for bias. Think it's time to rub my eyes and have a beer... 

Back to top 