I've read some more studies, and it still is very far from clear to me that spending time on explicit vocab instruction in non-specialized lexicons is more valuable than "natural reading." I'll try below to explain where the uncertainties lie, and why the studies I've read have not quelled my doubts. I apologize for getting into the boring weeds of these scholarly articles and their trails of footnotes, but looking for evidence that vocab instruction improves comprehension brings me back to 2003: it reminds me of looking for evidence of Saddam's nuclear arsenal. Supporters of explicit vocab instruction insist that proof is there, and they have created endless daisy chains of references to others who insist that proof is there, but in the end I couldn't find it.
It is just amazing how often assertions about the value of vocabulary instruction are made without offering good evidence. I talked about this problem in relation to the What Works document in my last post, but I see it over and over again. In one of our readings for today, Karen Bromley offers as one of her "Nine Things Every Teacher Should Know About Words and Vocabulary Instruction" the following remarkable statement in bold face print as her seventh essential truth we all should know:
"Direct instruction in vocabulary influences comprehension more than any other factor."
Taken at face value, this is just absurd. We wonder: Is explicit vocab instruction a more important factor than how much a student reads? Is explicit vocab instruction a more important factor than the family a student comes from? Of course not. But then, not only does Bromley immediately backtrack from her statement ("Although wide reading can build word knowledge, students need thoughtful and systematic instruction in vocabulary as well"--and again, Jay Gatsby might appreciate that "as well"), she also cites research that does not seem to back up her claim.
Her first (and best) reference is Blachowitz and Fisher's 2004 article, which does contain a section on "The Research on Vocabulary Instruction." This section, however, primarily offers evidence for home environment and wide reading as important factors, and then concludes that explicit instruction must be needed to fill in the gap for students who don't come from literate families or read a lot.
Why, we wonder, shouldn't we try to work on the wide reading part of it? (And the socialists among us might suspect that reducing inequality and poverty might help too). But that's not considered, so we are left only with explicit instruction, for which Blachowitz and Fisher offer only lukewarm support. The closest the article comes to arguing that explicit instruction can fill the gap is when it says, "studies support the idea that good vocabulary instruction can teach students the words they need to know to learn to read (Beck, Perfetti, & McKeown, 1982; Biemiller, 2001)." Blachowitz and Fisher's article eventually concludes with the underwhelming sentence: "Research indicates that effective vocabulary instruction can make a difference."
Nevertheless, the article did offer those two citations: Beck et al, 1982; Biemiller, 2001. So I looked at those.
Biemiller's article is entirely armchair theorizing, with no data of its own, as far as I could tell. That left the Beck et al. article from 1982. That article is, mirabile dictu, the same one that my workshop instructor gave me today after I questioned whether explicit vocab instruction could actually improve comprehension. That this somewhat lame article (the "Curveball" of vocab instruction?) has been the mainstay of explicit vocab instruction promoters for 30 years is, again, just amazing.
The Beck et al. article, which contains data from a controlled trial of a 5 month long unit of in-depth vocab instruction in 104 words, has two large problems. One problem is that control group in their experiment was only given "a traditional textbook curriculum." It seems very possible that this traditional textbook curriculum was totally worthless. I would like to know more about what that curriculum was, and I'd prefer to see explicit vocab instruction competing against a well-implemented free-choice reading program.
The second problem is that the Beck et al. experiment only taught 104 words, and only tested the students on passages that contained those very words. This may perhaps be relevant for other disciplines, but it does not seem very impressive for English. Of course if you teach kids words that occur in a passage, they will be able to understand the passage better; the question is whether teaching kids words (or roots, or morphemic awareness or whatever) will help them read passages that are not hand-selected to contain those words.
In a follow-up paper the next year, Beck et al. refined their experiment, and trumpeted it as a real breakthrough. According to their introduction, "studies that have attempted to improve comprehension through vocabulary training have brought equivocal results"; but their study had finally proved a link. That Beck et al. were evidently so proud of proving that five months of instruction on 104 words could improve students' comprehension of passages containing those very 104 words is just mind-boggling to me.
What all this leaves me with is a feeling of befuddlement. I am pretty sure that wide reading over many years will improve both vocabulary and comprehension, and to support that claim there is abundant empirical and anecdotal evidence which even the explicit instruction promoters acknowledge. Wide reading over many years: is that impossible? Schools have near-total control over students for six hours a day over many years. Then why in the world can we not make sure that our students are reading for at least an hour a day, every day, in every grade? Why are we going to such enormous lengths to try to prove that a mechanical process is as good as an organic one? Explicit vocab instruction, like so many things in our curriculum, is like a vitamin pill, a nutritional supplement. Why would we want to give our students endless vitamin pills, rather than just feeding them wholesome real food? Why would we spend decades and decades trying to formulate nutritional supplements? Or, for another analogy, explicit vocab and skills instruction looks to me like infant formula. OK, we can get better at making infant formula, but it's still probably never going to be quite as good as the real thing. I want a slogan. What's the educational equivalent of "Breast is Best"?