Sarah Bragg posted on Friday, January 09, 2015 - 1:13 pm
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;
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?
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 bi-factor 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.