Mplus Discussion >> Random Intercept EFA (RI-EFA)

Topics
Last Day
Last 3 Days
Last Week
Tree View

Edit Profile


Random Intercept EFA (RI-EFA)

Mplus Discussion > Exploratory Factor Analysis >

Message/Author

Kathryn Russell posted on Thursday, June 07, 2018 - 7:04 am

I am working with a measure of Big Five personality traits. The indicators are on an ordinal scale, so categorical in MPLUS.

Following the steps outlined by Aichholzer, 2014, I am attempting an RI-EFA to help adjust for response bias. The author includes the MPLUS syntax in the manuscript which I followed.

When I run a straight EFA, CFA, or ESEM - I have no problems. But in this RI-EFA, I am getting an error message and I am not getting model fit statistics. Can you help me identify what I am doing wrong?

The error is:
"THE MODEL ESTIMATION TERMINATED NORMALLY

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 716, F5 BY FFPIC81

THE CONDITION NUMBER IS -0.928D-18."

I can send syntax in a following post

Kathryn Russell posted on Thursday, June 07, 2018 - 7:08 am

ANALYSIS: ESTIMATOR = WLSMV;
ROTATION = QUARTIMIN;

MODEL: F1-F5 BY FFPIC01 FFPIC02 FFPIC03 FFPIC04 FFPIC05
FFPIC06 FFPIC07 FFPIC09 FFPIC11 FFPIC12
FFPIC14 FFPIC15 FFPIC16 FFPIC17 FFPIC18
FFPIC20 FFPIC21 FFPIC22 FFPIC23 FFPIC24
FFPIC26 FFPIC27 FFPIC28 FFPIC30 FFPIC31
FFPIC32 FFPIC33 FFPIC34 FFPIC35 FFPIC36
FFPIC37 FFPIC38 FFPIC39 FFPIC40 FFPIC41
FFPIC42 FFPIC43 FFPIC44 FFPIC45 FFPIC46
FFPIC47 FFPIC48 FFPIC51 FFPIC53 FFPIC54
FFPIC55 FFPIC56 FFPIC57 FFPIC58 FFPIC59
FFPIC60 FFPIC61 FFPIC62 FFPIC63 FFPIC64
FFPIC65 FFPIC66 FFPIC67 FFPIC68 FFPIC69
FFPIC70 FFPIC71 FFPIC72 FFPIC73 FFPIC74
FFPIC75 FFPIC77 FFPIC78 FFPIC79 FFPIC80
FFPIC81 FFPIC82 FFPIC83 (*L);

RI BY FFPIC01 FFPIC02 FFPIC03 FFPIC04 [...rest of variables...]
FFPIC75 FFPIC77 FFPIC78 FFPIC79 FFPIC80
FFPIC81 FFPIC82 FFPIC83@1;

RI WITH F1-F5@0;

Bengt O. Muthen posted on Thursday, June 07, 2018 - 4:20 pm

Your statement

RI BY FFPIC01 FFPIC02 FFPIC03 FFPIC04 [...rest of variables...]
FFPIC75 FFPIC77 FFPIC78 FFPIC79 FFPIC80
FFPIC81 FFPIC82 FFPIC83@1;

does not fix all loadings at 1, only the last one. You have to either use a list or repeat @1.

We ask that postings be limited to one window.

Kathryn Russell posted on Monday, June 11, 2018 - 6:55 am

Thank you! I will change my syntax accordingly.

My apologies for posting in two windows. I will be sure to post in only one in the future.