In trying to interpret my output from a Confirmatory Factor Analysis, I find myself pondering the relationship between the standardized and unstandardized factor loadings. Is there a formula in terms of the unstandardized estimates and the various standard deviations or variances that will yield the standardized estimates? I can't find it anywhere. (I'm using EQS 6.1, if it matters).



1 Answer 1


Further searching finds an answer from David A. Kenny (http://davidakenny.net/cm/1factor.htm), that seems to work:

"Conversion from the Unstandardized to the Standardized Solution: loading: multiply the loading by standard deviation of the latent variable divided by the standard deviation of the indicator error path: square root of one minus the standardized loading squared "


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.