EFA error message PreviousNext
Mplus Discussion > Exploratory Factor Analysis >
Message/Author
 Emaan Lehardy posted on Monday, May 11, 2015 - 3:44 pm
Hi,

I am trying to run an EFA with 117 items and a sample size of 333. I keep receiving the following error message:

*** FATAL ERROR
THERE IS NOT ENOUGH MEMORY SPACE TO RUN Mplus ON THE CURRENT INPUT FILE.
NOTE THAT THE MAXIMUM MEMORY USAGE BY Mplus 32-BIT IS LIMITED BY THE OPERATING SYSTEM.
REFER TO SYSTEM REQUIREMENTS AT www.statmodel.com FOR MORE INFORMATION ABOUT THIS LIMIT.
NOTE THAT THE OPERATING SYSTEM HAS NO RESTRICTION ON MEMORY USAGE FOR Mplus 64-BIT.

I'm not sure how to fix the problem and would greatly appreciate any help!
 Bengt O. Muthen posted on Monday, May 11, 2015 - 6:24 pm
Change to a 64-bit computer. If your items are categorical, you may want to study our FAQ:

Estimator choices with categorical outcomes
 Emaan Lehardy posted on Tuesday, May 12, 2015 - 12:19 pm
Thank you for responding!

I'm new to SEM and MPlus and greatly appreciate any help. My research team developed a 117 item measure using a Likert scale and administered it to 666 participants. We were originally going to do a split sample EFA/CFA but have decided to do an ESEM with the entire sample.

The administrator in our computer lab informed that all the computers are 64-BIT. However, I am still getting the same error message above regarding Memory Space.

Could anything about my code be responsible for this error? I have included it below and greatly appreciate any guidance and your time.


TITLE: ESEM 7 Factor

DATA: FILE IS ESEM data-whole sample.dat;
NOBSERVATIONS ARE 666;

VARIABLE: NAMES ARE all 117 items (<-- the message board would not let me list them all)

MISSING ARE ALL (999);

ANALYSIS: ESTIMATOR = MLR;
ROTATION = geomin(0.0001);

MODEL: f1-f7 by all 117 items (<-- the message board would not let me list them all) (*1);

OUTPUT: SAMPSTAT STAND RESIDUAL MODINDICES(10)TECH1;
 Bengt O. Muthen posted on Wednesday, May 13, 2015 - 9:28 am
Try deleting your request for MODINDICES which requires a lot of space with so many items and factors. And if that isn't sufficient, try using NOCHI and NOSERR in the Output command.
 Emaan Lehardy posted on Thursday, May 14, 2015 - 11:24 am
Thank you so much for your suggestions. They were very helpful and I was able to run the model.

Which resources can you recommend for me to read through as a beginner to ESEM that would help with interpreting the results of an ESEM? For example, if an item loaded significantly onto two or more factors, do you rely on the theory of your model to decide which factor it should load onto? We are interested in creating sub-scale scores for our measure and would need to decide which factor should retain the item.
 Bengt O. Muthen posted on Thursday, May 14, 2015 - 5:48 pm
Please see the list of ESEM applications on our web page

http://www.statmodel.com/ESEM.shtml
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: