Maureen noted that it is an either/or fallacy to assume that one must teach process or grammar, but the question goes beyond that. There are many, many different ways in which grammar can be taught, and even most of the people who are interested in teaching grammar, i.e., the people on this list, haven't given enough thought to the possible options and their implications. (I'll document this in my report on the survey I gave at the conference.)

By the way, I am VERY distrustful of anyone who states that "statistics show." Those are the words used to refer to all those incompetent studies which were used to "prove" that teaching grammar is ineffective. Martha destroyed many of those studies in her essay, "Closing the Books on Alchemy," and I dealt with several of the later studies. See:
http://www.sunlink.net/rpp/t001.htm
In other words, before I trust any references to statistical studies, I want specific bibliographical references so I can read the studies. In those about grammar (Mellon's. O'Hare's, etc. ― See link above.), the premises were often fallacious, methods were questionable, terms were often poorly defined.