Target Rotation ESEM PreviousNext
Mplus Discussion > Exploratory Factor Analysis >
 Paul Tremblay posted on Wednesday, March 01, 2017 - 12:59 pm
In regards to setting the target loadings, Asparouhov & Muthén (2009) state "Not all entries in matrix B need to be specified. For identification
purposes at least m - 1 entries have to be specified in each column for oblique rotation..."

My question is whether it is OK to specify not just m-1 but all entries as in the example below where I set all the cross-loadings to ~0.

rotation = target;
Physic by agg2 agg3 agg9 agg18 agg20 agg21 agg22 agg26 agg28r
agg1~0 agg5~0 agg8~0 agg25~0 agg29~0
agg6~0 agg11~0 agg13~0 agg14~0 agg15~0 agg17~0 agg23r~0
agg4~0 agg7~0 agg10~0 agg12~0 agg16~0 agg24~0 agg27~0 (*1);

Thank you.
 Tihomir Asparouhov posted on Wednesday, March 01, 2017 - 10:48 pm
I think so. The more targets the better the rotation identification (and possibly smaller standard errors).
Back to top
Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Options: Enable HTML code in message
Automatically activate URLs in message