Message/Author |
|
|
I would like to know if and how it is possible to estimate by MPLUS a multilevel Partial Credit Model or, that's the same, a random intercept ordinal logistic model for ADJACENT CATEGORIES (not for cumulative proportional odds). Thank you very much! Silvia |
|
|
No, that's currently not implemented. I am not sure if the Mplus Model Constraint feature could accomplish this, but it would probably be complicated. |
|
|
Hi have different items with 2 categories and others with 4 categories. I've used TAM in the past, and that environment is possible to treat a 2 category item as an special case while fitting the partial credit model. I've been trying to replicate this excercise with no sucess, by specifying CATEGORICAL = i1 i2 i3 (gpcm) i4 (gpcm) i5 (gpcm) ; Am I doing something wrong? Or this feature is not implemented in MPLUS v8.4 ? |
|
|
Take a look at Section 4 https://www.statmodel.com/download/MplusIRT.pdf There are different parameterizations for the GPCM. Check out the IRT parameterization section of the output. If you can't figure it out send the example as well as the TAM results to support@statmodel.com |
|
Back to top |