Message/Author 

Sarah Bragg posted on Friday, January 09, 2015  1:13 pm



Good afternoon, I was trying to run a bifactor model in Mplus, but I have run into an error and did not know if you would have any suggestions. Below is my input syntax: ANALYSIS: type = general; ESTIMATOR = ML; MODEL: General BY ITEM1* ITEM3 ITEM4 ITEM7 ITEM9 ITEM10; Optimism BY ITEM1* ITEM4 ITEM10; Pessimism BY ITEM3* ITEM7 ITEM9; General@1; Optimism@1; Pessimism@1; General WITH Optimism@0; General WITH Pessimism@0; Optimism WITH Pessimism@0; Output: STDYX; The system generates the following error: THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL. PROBLEM INVOLVING THE FOLLOWING PARAMETER: Parameter 24, ITEM10 To fix the problem, I tried setting Item10@0 in the syntax under the MODEL heading. When I do that, the model works perfectly. However, is there a better way to fix this problem, without assuming 0 error on Item 10? Thank you very much for your time. Sincerely, Sarah Bragg 


The model looks correctly set up. Perhaps a large negative residual variance occurs. We would need to know more to diagnose this. Please send data, input, output, and license number to Support@statmodel.com. 

fulya baris posted on Wednesday, February 01, 2017  1:10 am



Hi Dr Muthen I want to estimate person(theta) and item parameters( a, d) for bifactor model Does mplusAutomation can do it ? I coulndt find anything about it would u help me pls ? 


Yes. Just specify a bifactor model using the regular BY approach. And ask for factor scores (see UG)  they give you the estimated theta. Mplus also offers exploratory bifactor analysis  see UG examples. 

fulya baris posted on Wednesday, February 01, 2017  11:15 pm



Thanx for your answer Dr. Muthen I examined your Mplus examples about bifactor I saw model results.. thera are factor scores u mean that they are person parameter( theta) but I still couldn't find discrimination (a) and difficulty (a) parameters in your examples. 


Bifactor modeling lets every item be influenced by 2 factors. This usually is not described in terms of discrimination and difficulty parameters. See the FAQ on our website: IRT parameterization using Mplus thresholds 

Back to top 