

Fitting a multivariate latent trait m... 

Message/Author 

Derek Xu posted on Sunday, March 28, 2010  5:46 pm



Hello, I am a beginner and try to fit a multivariate latent trait model in Mplus: Prob (Y_gki <= c_l  Z_gi, U_gi) = r_gk + (1  r_gk)/(1 + exp((alpha_kl + beta_k*Z_gi + delta_k*U_gi))) where g = 1,2 indicating groups, k = 1,..., 7 indicating 7 instrument items, i indicating the subject number in each group, so Y_gki is the response for item k from subject i in group g; each item has a Likert scale with responses 05 Z_gi = (Z1_gi, Z2_gi) is 2 dimentional latent variable and assumed to follow bivariate normal distribution It's known that Z1 relates to first 3 instrument items and Z2 relates to the other 4 items. So beta_k can be determined. U_gi is observed covariate. The parameter of interest is r_gk and its MLE is needed. May I know if Mplus can fit such a model? If yes, which example/section of handouts/paper shall I follow? Thanks a lot!! 


I think r is a guessing parameter. The current version of Mplus does not estimate a threeparameter IRT model. 

Derek Xu posted on Monday, March 29, 2010  10:35 am



Thanks for your response, Linda. Do you know if there any other approaches to fit such a model? 


I don't know of any program that includes the guessing parameter and covariates. Check to see if any of the IRT programs put out by Scientific Software can do this. 

Back to top 

