The Simple View of Reading

By Matthew Davis, PhD

The reading program being developed by Core Knowledge under the editorial direction of Matthew Davis is making great strides. Many of you have seen a preview of the work in Reading Instruction: the Two Keys. This is the first installment in a series of articles by Dr. Davis that we will be publishing in the coming year.

Reading is a complicated operation, and yet one of the most widely supported models of reading is surprisingly simple. In fact, it is known as “the simple view of reading.” This view, which is associated with Philip Gough, William Tunmer, and other reading researchers, holds that there are two chief elements that are equally important to reading comprehension. One is decoding skills and the other is language comprehension ability.

To achieve reading comprehension, a person needs to be able to decode the words on the page and then make sense of those words. The first is made possible by decoding skills and the second by language comprehension ability. If the person can’t decode the words on the page, she won’t be able to achieve reading comprehension, no matter how much oral language she can understand. But even if the person can decode the words on the page, that in and of itself is still no guarantee of reading comprehension. If the sentences the person is attempting to read are sentences she could not understand if they were read aloud to her, then there is not much hope that she will understand them during independent reading either.

Supporters of the simple view—and there are a growing number of them among reading researchers—argue that a person’s reading comprehension ability can be predicted, with a high degree of accuracy, based on measures of decoding skills and language comprehension. Researchers who hold to the simple view say, “Tell me a person’s decoding ability, as ascertained by a word-reading task, and tell me that person’s language comprehension ability, as ascertained by a listening comprehension task, and I can make a very accurate prediction of that person’s reading comprehension ability.” If the person is a rapid and fluent decoder and also able to understand a wide range of oral language —for instance, a news items on the radio, books on tape, etc.—then it’s a safe bet the person will also do well on tests of reading comprehension.

The simple view can account for reading failures as well as reading success. Simple view researchers have shown that, when a student has deficits in reading comprehension, that student generally has deficits in either language comprehension or decoding skills, or both.

An interesting thing about the simple view of reading is that it can be expressed as an equation:

R = D x C

In this equation, each of the letters stands for a specific skill:

R is a measure of reading comprehension ability
D is a measure of decoding skills
C is a measure of language comprehension ability

Each of these skills can be quantified as a numerical value between 0 and 1, where a zero stands for no ability whatsoever and a 1 stands for perfect, not-to-be improved upon ability. Obviously most people have a skill level that falls somewhere between these two extremes.

Since any number multiplied by zero is zero, a person with no decoding skills (D = 0), will also have no reading comprehension ability (C x 0 = 0).  But the same is true of language comprehension. If someone has no ability to understand a language (C = 0), it does not do any good to know how to sound it out (D x 0 = 0). I am reminded of a time in younger days when a friend and I sat up all night drinking red wine with some people from the former Soviet Republic of Georgia. They were determined to teach my friend and me the Georgian alphabet and the sounds the letters represented. After several hours of drill, I could roughly sound out Georgian words —but, of course, I had no clue what they meant. My companions had taught me rudimentary decoding skills (D>0) but I knew no Georgian words and no Georgian syntax, so I had no ability to understand the words (C = 0). When I woke up in the morning I found that— in addition to having a slight headache— I still had no ability to read Georgian (R = 0). I also had a nasty hangover.

Another silly but helpful example (which I borrow from Gough) can help make the “simple view” concrete. This example comes from English literature and involves the English poet John Milton, the author Paradise Lost and other well-known poems. Milton went blind late in life. Since Braille had not yet been invented, this meant he could not read for himself. Milton found a way to keep learning from books: he had friends and relatives read books aloud for him. But he couldn’t always find a scholar who had both the ability and the free time to read to him in Hebrew, Greek, Latin, and other ancient languages. The solution? Milton taught his daughters to decode these languages so they could read books in those languages aloud to him. But Milton did not teach his daughters the actual languages—the thousands of words and tens of thousands of meanings. That would have been a difficult, time-consuming task. He only taught them the phonemic rules they would need to turn letters into sounds. Thus, his daughters had solid decoding skills for these languages (D>0), but they would have scored a zero on any measure of language comprehension (C=0). It was with Milton’s daughters as it was with me that night in Georgia: they could turn symbols into sounds, but they had no idea what the sounds meant. Milton, on the other hand, on account of his blindness, had no functional decoding skills (D=0). However, by virtue of his great learning, he had a very high value for C. He could understand the words his daughters read aloud. Between Milton and his daughters, you might say, there was reading comprehension (R), but the younger generation brought the D and the patriarch brought the C.

These are two rather odd examples meant to show that D and C can be separated. Of course, the goal of reading teachers everywhere is not to separate D and C, but to build both D and C, so that children are able to understand what they read. It seems to me that the simple view has clear and important implications for reading instruction, implications which have not been fully understood in the K–8 educational community. In particular I have found the simple view to be useful in terms of combining what we have learned from recent research findings regarding decoding with what E. D. Hirsch has been saying about reading for many years.

The simple view points up the importance of teaching decoding skills—and we are beginning to get some clear evidence as to how those can most effectively be taught. Teachers should be familiar with the findings of the National Reading Panel regarding the importance of phonemic awareness, phonics, and fluency. But the simple view does not equate reading with decoding. It reminds us of the importance of language comprehension ability, which is precisely the point that E. D. Hirsch has been trying to make for the past two decades. No matter how good your decoding skills may be, you will not be a good reader if you have an impoverished vocabulary and insufficient knowledge of the sorts of things writers take for granted. This was the point of Cultural Literacy; it is also the point of Hirsch’s latest book, The Knowledge Deficit. It is the reason why the Core Knowledge Foundation exists and the reason why the Core Knowledge Sequence lists all those highly specific things to be learned in each grade.

Schools looking to improve reading achievement need to take a long-term view. They need to combine effective decoding instruction with a specific, carefully sequenced curriculum designed to build background knowledge over time. (It might be the Core Knowledge curriculum, or it might be something else of comparable scope and rigor.) If the two-part view of reading laid out in the simple view is correct, only a two-pronged program of instruction of this sort is likely to succeed.

The simple view also lends support to another view advanced by Hirsch. For many years, Hirsch has argued that disappointing reading scores in fourth grade and beyond are largely attributable to deficiencies in background knowledge. Recent research by scholars investigating the simple view seems to support this interpretation. A team at the University of Kansas looked at measurements of reading comprehension (R), decoding/word recognition (D), and listening comprehension (C) for the same 570 students in second, fourth, and eighth grade. They found that D x C accurately predicted R in each grade, but they found that one of these values became more important over time. Can you guess which one?

The measure of decoding (D) was extremely important in the second-grade results. 27 percent of the variance in reading comprehension in second grade could be explained by decoding skills (D) alone; by contrast, only 9 percent of the variance could be explained by listening comprehension (C) alone. By fourth grade, however, the measure of listening comprehension had begun to account for more variance: the unique contribution of C rose to 21 percent while the equivalent number for D fell sharply. By eighth grade, fully 36 percent of the variance in reading comprehension scores could be explained with reference to the children’s listening comprehension ability. The unique contribution of D sank even further. In other words, while reading comprehension depended on D and C at every stage, as the simple view would predict, C got more important as time went by. Once the intricacies of decoding are mastered (and in English this takes some time), reading comprehension depends more and more heavily on language comprehension. And language comprehension depends on knowledge.

In future issues of this newsletter I hope to use the simple view as a framework within which to explain some of what we at the Core Knowledge Foundation are trying to do in the innovative new reading program we are developing.

Next article