That sounds obviously counterproductive. Did I misunderstand?
But I also think there's another, far more useful purpose for educating people suffering from stereotype threat about stereotype threat: they can then put a label to what is keeping them from getting their work done when asking their peers, mentors, or managers for help. And then their peers, mentors, and managers can pay attention to situations that might provoke stereotype threat, try to lessen them, and otherwise show a bit of compassion for the person suffering from it.
I say this personally: learning about stereotype threat cleared a huge communication roadblock I'd been having for years with managers and peers, and as soon as we were on the same page we were actually able to address some of the issues contributing to it.
For example, I'm self-taught, and while I have many years experience now and have done some significant projects, I still second guess myself quite a bit. (This is often called "Imposter Syndrome", see e.g. http://en.wikipedia.org/wiki/Imposter_syndrome.) I found it tremendously helpful to discover that people I've looked up to for years have felt the same way, and had helpful suggestions for getting over the internalized anxiety.
Everybody has nagging fears like this. Learning how to see them for what they are and to avoid making things worse for other people benefits everybody.
People underperform because they're worrying about problem X (confirming stereotypes). Let's refer to this as problem Y (stereotype threat).
Education about problem X would be counterproductive; the article is suggesting education about problem Y.