| Message/Author |
|
|
| Anonymous posted on Wednesday, February 21, 2001 - 7:13 am
|
|
|
| I got a standardized loading exceeding -1.0 for one of 34 dichotomous items using EFA procedure of Mplus. It seems extraordinary for me, and thus I'm not sure whether it's a correct estimation. |
|
|
|
|
| Factor loadings greater than one can occur. For example, this may be a result of negative residual variances or an otherwise misspecified model. |
|
|
| Lin Gu posted on Thursday, July 29, 2010 - 2:19 pm
|
|
|
| I remember somewhere in this discussion space a published article was referred on this topic. Would you please cite this article again here? Thanks. |
|
|
|
|
Try the following link:
www.ssicentral.com/.../HowLargeCanaStandardizedCoefficientbe.pdf |
|
|
| Joop Hox posted on Tuesday, August 20, 2013 - 1:12 pm
|
|
|
| Why does Mplus 7 no longer produce standardized factor loadings for EFA? |
|
|
|
|
| Not sure what you mean - EFA results are already standardized by the fact that we analyze a correlation matrix and fix factor variances at 1. And for ESEM, one can request standardization. |
|
|
| Akbarzadeh posted on Sunday, June 28, 2015 - 11:28 pm
|
|
|
If we change scale of observed variable(s), what happen for our results in EFA?
for ex. if weight(kg) change to weight(gr) |
|
|
|
|
You results will not change if you multiply or divide your variables by a constant.
Please do not post the same question multiple times. I am deleting your other posts. |
|
|
| Back to top |
