One of the things I've begun to notice recently is that I can generally distinguish between kids who were taught to read using a whole language approach and those taught to read using phonics. Almost invariably, the kids who were taught using whole language have considerably more difficult breaking words apart and examining their component parts. I tend to see this much more prominently when I tutor French or Italian -- often a student will read the first couple of letters in a word and then simply guess what the rest of it says, which is an absolute disaster in French -- but I see it when I tutor the SAT as well, albeit in a more roundabout way.
For example, one Blue Book sentence completion contains the the answer choice "deferential," which is a word that most of my students are unfamiliar with. What's interesting, though, is how they react to it. Usually I ask them if they can relate it to a word they know, and typically they can't think of anything, but recently one of my students said that it looked like "different." That one threw me a little. On one hand, my student was absolutely right: "defer" and "differ" do sound similar. Unfortunately, they have nothing to do with one another. And that, in turn, made me wonder about the whole idea of asking students to relate unfamiliar words to words they already know. The underlying assumption of that strategy is that students already know what parts of words they should and should not focus on, that they can distinguish between "sounds similar" and "related in meaning." And that assumption, as I've discovered, is not necessarily a valid one.
The SAT and Phonics