Message/Author 

rechard posted on Wednesday, October 11, 2006  6:55 pm



I have some questions about multilevel LCA/LCR, similar to Example 10.3 in User's Guide. (1) Example 10.3 assumes c#1 and c#2 are related through the common factor f. If I don¡¯t believe this, and delete ¡°f by c#1 c#2¡±, how will Mplus fit the model? Also, how to get output of variance (and/or correlation) of the two random effects in this case? (2) Is it possible in Mplus that we assume the two random effects are correlated bivariate normal and estimate the model by maximum likelihood with numerical integration? If so, how? Thanks very much. 


1. If you don't use f, and instead say: c#1 c#2; c#1 with c#2; then you get variances and the covariance estimated for the two correlated, biv normal random means. So ML uses 2dimensional numerical integration on the between level. 

Paul Widdop posted on Wednesday, August 06, 2008  2:56 am



Morning, I am running example 10.7 for my data with the following script...... Variable: Names are swimming snooker football newoutdo wintersp water cycling fitness golf tenpin jog cricket rackets2; Missing are all (9999) ; USEVARIABLES = swimming  rackets2; CATEGORICAL = swimming  rackets2; CLASSES = cb(5) cw(4); WITHIN = swimming  rackets2; BETWEEN = cb; CLUSTER = lad_code; ANALYSIS: TYPE = TWOLEVEL MIXTURE; PROCESSORS = 2; STARTS =100 10; MODEL: %WITHIN% %OVERALL% %BETWEEN% %OVERALL% cw#1cw#3 ON cb; MODEL cw: %WITHIN% %cw#1% [swimming$1rackets2$1]; [swimming$2rackets2$2]; %cw#2% [swimming$1rackets2$1]; [swimming$2rackets2$2]; %cw#3% [swimming$1rackets2$1]; [swimming$2rackets2$2]; %cw#4% [swimming$1rackets2$1]; [swimming$2rackets2$2]; ........But when I run the model I keep recieving the following error message.. *** ERROR in Model command Unknown variable(s) in an ON statement: CB ... If anyone on th eforum can offer advice that would be great. Many thanks Paul 


I can't see the problem. Please send the files and your license number to suppport@statmodel.com. 

Joan W. posted on Thursday, March 05, 2009  12:58 pm



Dear Drs. Muthen, I am using LCA on six math subtest scores with a sample of around 300 children. Those children were from 20 different classrooms. I read from the Mplus manual that Type=COMPLEX can be used to enable corrections to standard errors and chisquare test of model fit for nested data structure. If my major interest is in number of classes identified instead of significance of path coefficient, is it still important to include Type=COMPLEX? What are the expected consequences when TYPE=COMPLEX is not enabled? Thanks. Joan 


Without TYPE=COMPLEX using only the CLUSTER option and not the WEIGHT option, parameter estimates will be correct, standard errors will be underestimated, and chisquare will be overestimated. Note that a minimum of 3050 clusters is recommended for TYPE=COMPLEX and TYPE=TWOLEVEL. If you are not concerned with standard errors and chisquare, you can look at the loglikelihood and BIC to determine the number of classes. They are not affected by nonindependence of observations due to clustering. TECH11 and TECH14 will not be correct. 


Hello, I am running a multilevel latent profile analysis and I would like to change the reference class for the betweenlevel part of the analysis. Is there a way to do this? Thanks, Ginger 


Are you saying that you have a betweenlevel latent class variable? Perhaps you need to send your output to Support. 


HiyesI have a betweenlevel latent class variable. I will send my output to Support. Thanks, Ginger 

Back to top 