|
|
 |
| Multifactor item response model |
 
|
| Message/Author |
|
|
| arif özer posted on Sunday, July 15, 2018 - 6:50 am
|
|
|
Dear Authorized,
I tried to write a syntax which its output gives both multifactor CFA model coefficents and IRT parameters (2PL), but I failed.
Mplus Guide includes only one factor model for irt modes. Can you help me, what is wrong with the syntax?
TITLE: Adaptation of Vocational Development Scale
DATA: FILE IS mesgel.dat;
VARIABLE: NAMES ARE
m1 - m25;
USEVARIABLES ARE m1 - m25;
CATEGORICAL ARE m1 - m25;
ANALYSIS:
ESTIMATOR = mlr;
integration = montecarlo;
MODEL: crystallizing by m1-m6*;
exploring by m10-m12*;
deciding by m15-m19*;
preparing by m21-m23 m25*;
crystallizing with exploring-preparing;
exploring with deciding-preparing;
deciding with preparing;
[crystallizing@0]; crystallizing@1;
[exploring@0]; exploring@1;
[deciding@0]; deciding@1;
[preparing@0]; preparing@1;
OUTPUT: stdyx TECH8;
PLOT: TYPE= PLOT3; |
|
|
|
|
Mplus prints the IRT parameterization with a single factor and binary items. But you can use Model Constraint to do it yourself - in your case where each item is influenced by only one factor you can do the translation just as shown in our document
http://www.statmodel.com/download/MplusIRT.pdf |
|
|
| Back to top |
|
|
