I am in the midst of a lot of outside things at the moment and cannot contribute like I would want to the discussion. However, for background on the current discussion regarding Q, see the poll here as well as Mark Goodacre’s post.
I do not believe the methodology behind Q is measurable nor demonstrable. Rather, it begins with a presupposition based on a previous generation’s lack of concern over writing styles in the ancient world. (in a really simplified version of the Q hypothesis:) It suggests that Matthew and Luke used several sources because their verbiage, when they are in agreement, doesn’t fully agree. But, as some have dared to demonstrate, innovation (adding to, rewording, making use of in some way) was required even when borrowing a previous text.
This is why we can demonstrate the changes Matthew applied to Mark via a lot of articles by Mark Goodacre and by an impressive book on Matthew’s Judaization of Mark by Anne O’Leary. For instance, Matthew could have easily taken Matthew 6.9-15 directly from various parts of Mark. While Sanders and others would allow for undefined sources, I would suggest pointing Matthew’s unique passages, such as the Sermon on the Mount, to Deuteronomy or other important books to his community.
Q cannot be demonstrated except by extrapolation. On the other hand, the Farrer-Goulder and Goodacre theory can be demonstrated by first understanding ancient writing styles and then by showing how Matthew first expanded Mark and then by showing how Luke used both Mark and Matthew. We can discuss, using the same methodology, how John used Mark and Luke as his primary sources if you would like, but I’m afraid that may bog us down at the moment.
While Q was a valid hypothesis was a long time and the work poured into it by Q scholars (learned scholars who must have our respect) it simply is not needed when we have a firmer, and established, pattern of literary development.