We're reworking now the placement test I spent the end of last year doing. It's a bit odd, coming back now to work I did in the spare room at home, during the Tyneside winter, as I sit at my desk in Tripoli.
However, at last I have access to the internet here, and have decided to go back to the test's original source, the Common European Framework. Googling, I came across this article from the Grauniad in 2004. In it, North comments that the six CEF levels "are not the product of acquisitional hierarchies from second language acquisition (SLA) research." Indeed they are not, because, "Unfortunately SLA research has so far only produced a partial, contradictory glimpse of what an acquisitional hierarchy might look like."
This is where we are going to. When we understand the order in which language is learned, we can teach it accordingly. It's late in the day, but I might shift my SLA research to this area for the paper I have to produce before August.