Reading Comprehension

Today's reading assignment is to read Daniel Willingham's article, The Usefulness of Brief Instruction in Reading Comprehension Strategies. Here's the money graf:

We can summarize what we know from the last 25 years of research on reading comprehension strategies fairly concisely:

• Teaching children strategies is definitely a good idea.
• The evidence is best for strategies that have been most thoroughly studied; the evidence for the less-studied strategies is inconclusive (not negative) and, therefore, there is not evidence that one strategy is superior to another.
• Strategies are learned quickly, and continued instruction and practice does not yield further benefits.
• Strategy instruction is unlikely to help students before they are in the third or fourth grade.

Modern reading instruction is all about teaching reading comprehension strategies, even though these strategies are "learned quickly, and continued instruction and practice does not yield further benefits." Willingham says:

I don’t believe that students continue using these strategies into adulthood. Literate adults do not construct story maps as they read the morning paper, nor do they pose and answer questions for themselves. They do, however, understand that the goal of reading is to obtain meaning, and they monitor their own comprehension; that understanding is likely what remains with the tenth-grader who was taught a set of reading strategies in fourth grade.

\
That's point three from the list above. Educators seem to stop reading after point two.

Point four is that the teaching of reading strategies shouldn't begin until after students have learned to decode fluently, i.e., after about the third grade.

Strategies require attention and space in working memory (e.g., Cain, Oakhill, and Bryant, 2004; Calvo and Castillo, 1998). Students who are still learning to decode fluently do not have enough working memory space available to implement strategies. Their working memory is occupied by decoding. A natural conclusion is that there is not much point in teaching reading strategies before students have gained that fluency—for most students, that will be in the third or fourth grade.

I'll save my commentary on the wacky decoding instruction favored by educators for another post. Suffice it to say, educators believe that decoding is quest for meaning, not word identification, and, as a result, teach decoding like it's merely another reading strategy. It's not.

So what should effective reading instruction be about?

Teaching reading strategies is worthwhile, but we should bear in mind that knowledge of strategies is only a small part of what makes an effective reader. A good reader also decodes fluently, has a broad vocabulary, and has wide-ranging background knowledge.

Formal reading instruction should be about teaching decoding skills in the early grades then briefly teaching a few reading strategies. But that's not enough to make good comprehenders.

What's also needed is the development of a broad vocabulary and wide-ranging background knowledge -- two things that aren't getting done in most schools.

The need for a broad vocabulary should be self-evident. It’s hard to understand the meaning of a sentence if you don’t know the meaning of the constituent words. There are times when you can deduce the meaning of an unknown word from the context, but you need to understand most of the text to be able to puzzle out the meaning of the unknown word. Your ability to do so drops rapidly, however, as the number of unfamiliar words increases (Laufer, 1997).

Background knowledge also has profound effects on reading comprehension. Have another look at the three classes of strategies in the table—two of them rely on or are facilitated by background knowledge. Encouraging students to relate the text they are reading to background knowledge won’t help much if the students don’t have the relevant background knowledge. Less obvious is the fact that relating sentences to one another also often requires background knowledge.

My assessment of the state of reading instruction is that the decoding piece is frequently done poorly (this alone causes failure in many students) and the reading strategy piece is greatly overemphasized (and taught too early). The other two piece of the puzzle, vocabulary and background knowledge, are underemphasized as part of the growing trend to minimize the instruction of content knowledge in school.

You can't miss two of the four pieces of the puzzle and then be surprised when you fail to create good readers.