Education Letter to the Editor

Is Skilled Decoding Needed in Learning How to Read?

January 05, 2009 4 min read

To the Editor:

According to the recent Reading First final impact study, children participating in the program did better than their peers in comparison schools on a test of decoding in grade 1, but did no better on reading-comprehension tests in grades 1, 2, and 3, despite considerable extra instructional time (“Federal Path for Reading Questioned,” Dec. 3, 2008).

This is not the first study to show that children following an intensive decoding-based curriculum do well on tests of decoding but not on measures of reading comprehension when compared with regular students. Similar results were shown in the report of the National Reading Panel, the foundation for Reading First. As Elaine Garan, a professor of literacy and early education at California State University-Fresno, has pointed out, the panel found that systematic phonics was superior only on tests of decoding, and that its impact was small or insignificant on reading-comprehension tests after grade 1.

The same pattern appears in research on direct instruction, which assumes that students must know how to sound out words before they can read with understanding. Children taught using direct instruction did much better on the Wide Range Achievement Test (decoding) than on the Metropolitan Achievement Test (reading comprehension and vocabulary), even when followed to grade 9.

Such results suggest that a high level of proficiency in decoding is not necessary in order to learn to read. This is also supported by studies showing that children in classes emphasizing interesting, comprehensible reading, and with less decoding instruction, perform as well as or better than children in decoding-emphasis classes on decoding tests, and typically score higher on tests of reading comprehension. There are also cases of children who have learned to read on their own with little or no decoding instruction.

This is consistent with the views of reading researchers Frank Smith and Kenneth Goodman, who hypothesize that we learn to read by understanding what is written. Our ability to decode complex words is the result of reading, not the cause.

Stephen Krashen

Professor Emeritus

Rossier School of Education

University of Southern California

Los Angeles, Calif.

To the Editor:

Irony of ironies: After $6 billion of federal investment in the Reading First program, America’s children have improved at sounding out scrambled-up nonsense words and identifying words on word lists, but they are no better at comprehending the stories they read. Reading experts are now trying to figure out why.

The reason should be crystal clear. The requirements of Reading First reflect the singularly focused beliefs of G. Reid Lyon, once called George W. Bush’s “reading czar.” By his own account, Mr. Lyon failed to teach children to read when he was a 3rd grade teacher (so stated in an April 2001 U.S. Department of Education community newsletter), and blamed an apparent lack of phonics and decoding instruction in the early grades.

Subsequently, from his position as the chief of the child-development and -behavior branch of the National Institute of Child Health and Human Development, he supported neuroimaging research by Yale pediatrician Sally E. Shaywitz that focused solely on identifying words on word lists, not reading comprehension. Dr. Shaywitz’s “phonological processing” research later was used to trump other well-defined studies that pointed to the value of balanced reading instruction.

Mr. Lyon’s carefully orchestrated efforts to focus early reading instruction on phonics and decoding did not create competent readers. His efforts yielded competent decoders. Any person who can read the following must admit that “decoding” and “reading” cannot be the same: Exllnt raeedrs cnstrct maennig form lteter clues anywhr tehy appr in txet. Tehy aks thmeslvse, “Waht deos the txet say?” not “Waht is ech lteter and how is ech wrod spleled?”

Compliant teachers did exactly what Reading First required, and children consequently became efficient decoders, not effective comprehenders.

Rhonda Stone

Shelton, Wash.

To the Editor:

How sad that Reading First’s creators and supporters haven’t a clue about why it has failed to improve children’s reading comprehension. Why can’t they see that their narrow conception of reading as a hierarchical set of discrete skills stands in the way of children’s learning to read? The current programs approved for Reading First schools are like tennis rackets with only a couple of strings. No wonder most kids can’t hit the ball.

Not only does the array of knowledge and skills needed by even beginning readers far exceed Reading First’s narrow emphasis on phonemic awareness, phonics, and fluency, but all the tools of reading also have to be used at once and in concert. G. Reid Lyon and his colleagues say young children can’t do that? Nonsense! They learn to walk and talk and use computers, don’t they? And nobody teaches them.

If Reading First is part of the reauthorized No Child Left Behind legislation, it should be redesigned to include a little phonics, taught at points of need; well-written, interesting stories; writing, right from the first day; poems, recited aloud; dramatization of stories; and, most important, daily teacher reading of good books.

Joanne Yatvin

Portland, Ore.

The writer, a former president of the National Council of Teachers of English, was a member of the National Reading Panel.

A version of this article appeared in the January 07, 2009 edition of Education Week as Is Skilled Decoding Needed in Learning How to Read?