|
![](http://www.statmodel.com/discussion/icons/csc/1px.gif) |
Factor structure with negative correl... |
![Previous](http://www.statmodel.com/discussion/icons/csc/navgen_arrow_left.gif) ![](http://www.statmodel.com/discussion/icons/csc/1px.gif)
|
Message/Author |
|
|
My understanding is that the "factor structure" portion of the EFA output shows the correlations between the indicators and the factors. I have numerous indicators correlating negatively with one of my factors. How is this possible? Please note that I am using quartimin rotation, a survey weight, clustering on a level-2 ID variable, Estimator=WLSMV because I have some indicator variables, and I have specified that those indicator variables are categorical. |
|
|
One other question related to my analysis: is it possible to generate standardized factor loadings? If so, how? |
|
|
Negative values in the factor structure points to negative factor loadings. EFA is standardized. |
|
Back to top |
|
|