I use the open source program R for doing statistics. In one study, I use the package SEM for doing a “Structural Equation Model” and I was just not able to get something specific done (factor scores for latent variables). Well, what do you do? I sent a mail to the R help mailing list and got a reply “What you want to get done is currently not possible with SNA. But it is a reasonable request.”

A few weeks later I receive a personal reply from the author of the SEM package, John Fox, a sociology professor at McMaster University in Ontario, Canada. He wrote:

Dear Tomek and Sebastian,

Both of you inquired recently about getting factor scores for latent variables in the sem package in R. I’ve added a function named fscores() to
the development version of the package, which will compute either factor-score coefficients or the factor scores themselves.
I’ve added a function named fscores() to the development version of the package, which will compute either factor-score coefficients or the factor scores themselves.

The development version of sem is hosted on R-Forge, where packages are built daily. You should be able to install sem from R-Forge tomorrow
(Tuesday), via the command install.packages(“sem”, repos=”http://R-Forge.R-project.org”).

Regards,
John

This is one reason for embracing the open source development model. Way to go, and… thanks John.