|
|
 |
| Replicating a CFA using summary data |
 
|
| Message/Author |
|
|
| jmquinn posted on Wednesday, June 04, 2014 - 12:41 pm
|
|
|
Good afternoon,
I am attempting to replicate a published study using their supplied correlation matrix and means and standard deviations, and I keep getting the following error message after specifying this input:
Data:
FILE IS "matrix3_2014.txt";
TYPE IS CORR MEANS STDEVIATIONS ;
NOBSERVATIONS = 122;
variable:
NAMES ARE V1-V6;
USEVARIABLES ARE ALL;
model:
Y1 BY V1-V3;
Y2 BY V4-V5;
V6 on Y1 Y2;
Y1 WITH Y2;
*** ERROR
One or more variables have a variance of zero.
Variable Variance
VOC 0.146
LSK 0.380
BURT 0.660
**WRAT 0.000
**LC 0.000
RC 0.889
Since this is summary data, why am I getting this error message? Thank you for your help! |
|
|
| jmquinn posted on Wednesday, June 04, 2014 - 12:43 pm
|
|
|
| I apologize, I was trying to blind the variable names at the bottom. Those correspond to V1-V6. |
|
|
| jmquinn posted on Wednesday, June 04, 2014 - 1:37 pm
|
|
|
good news, I figured out my issue.
For those who read this at a later time and wonder the same thing: try to use a full correlation matrix instead and specify FULLCORR. that fixed it for me, though there may have been another way to fix it using the partial matrix. |
|
|
| Back to top |
|
|
