Tuesday, February 05, 2008

Responding to Garrison Keillor: Phonics, Reading First, and a failure to communicate

ELL Advocates: Blog of the Institute for Language and Education Policy

Responding to Garrison Keillor: Phonics, Reading First, and a failure to communicate
February 4, 2008


Garrison Keillor has caused quite a stir with a recent essay in which he defends Reading First and the teaching of phonics and accuses liberals of opposing Reading First because it is the product of a Republication administration. His points, briefly, boil down to these:
  • We have a reading problem in this country.
  • Research shows that phonics works.
  • Therefore, we should all support phonics and Reading First.
  • Keillor’s evidence that a reading problem exists is made up of two familiar arguments: The first is the high percentage of students in who score “below proficient” on reading tests and the second is the fact that there are significant gaps between different groups of children on these tests. In each case, Keillor is referring the scores on the NAEP examination (National Assessment of Educational Progress), a test given to samples of students in grades four, eight and twelve.

    The “proficient” level

    The first argument is bogus. Gerald Bracey (2006) has pointed out that the NAEP performance levels, such as “basic” and “proficient,” are not based on anything solid. In particular, the “proficient” level is set much too high: Other countries that consistently rank near the top of the world in reading would not do well on our NAEP: Bracey notes that, for example, only one-third of Swedish children would be considered “proficient” on the NAEP, nearly the identical percentage of US fourth graders (31% in 2005). Thus, scoring “below proficient” does not necessary mean poor reading ability.

    The gap

    As Mr. Keillor notes, children from low-income families score much lower than children from high-income families on tests of reading. This is the crucial “gap” we need to discuss. When researchers include the effect of poverty in their analyses, minority status is no longer a significant predictor on tests of language arts and math (see e.g. Darling-Hammond, 2007).

    The question then becomes why children from high-income families read better. The answer has nothing to do with phonics. A major reason poverty has such an influence on reading ability is that children from high-income families have far more access to books. They attend schools with better classroom and school libraries, live in communities with better public libraries and more book stores, and have far more books in their homes. Researchers Susan Neuman and Donna Celano concluded that children from high-income families are “deluged” with books, while children of poverty have a hard time finding anything to read.

    This is important because research consistently tells us that those with more access to books read more, and those who read more read better (McQuillan, 1998; Krashen, 2004).

    The first step in eliminating the gap is to improve access to books, and the easiest way to do this is to improve libraries, something Reading First has shown little interest in doing, despite consistent research showing that library quality is related to reading achievement (McQuillan, 1998; Krashen, 2004).


    Mr. Keillor thinks that research shows that phonics works. There are two very different views of phonics. “Basic” phonics refers to teaching those sound-spelling rules that children can actually learn and apply to texts. Every professional in education understands the value of basic phonics. Reading First insists on an extremist view of phonics called “intensive systematic phonics,” which means teaching all the major rules of phonics in a strict order, to all children. These rules can be extremely complex and often have numerous exceptions.

    Intensive systematic phonics has been shown to be effective only for performance on tests in which children read lists of words in isolation. It has only a microscopic influence on tests in which children have to understand what they read, tests of reading comprehension given after grade one (Garan, 2002). Rather, study after study has shown that performance on tests of reading comprehension is heavily influenced by access to books and the amount of self-selected free voluntary reading children do.

    Track record of Reading First

    Reading First has failed to improve academic achievement in the nation as a whole and on tests given by individual states. Most of the increase in national (NAEP) reading scores that President Bush and Education Secretary Spellings keep talking about happened before NCLB/Reading First went into effect On an international test of reading (PIRLS), our fourth graders scored 542 in 2001 and 540 in 2006. No improvement, despite huge amounts of money and added instruction in reading. (For details, please see my postings on this topic at http://www.districtadministration.com/pulse/.)

    Mr. Keillor also appears to be unaware of the seamy side of Reading First, the scandals, the accusations of preference shown for certain publishers connected to the Bush family in awarding grants.

    We have, however, learned something from this experience.

    When such a brilliant, insightful and usually reasonable person such as Garrison Keillor has such a distorted point of view, we have to ask why. Most likely, what Mr. Keillor knows about Reading First and phonics is what he has read in the newspapers. This tells us how thoroughly we have failed to communicate with the media and with the public. The counterarguments I used in this response are well-known to many educators, but are nearly completely unknown to the general public. Those who are not part of the world of education should have easy and ready access to more than one point of view.

    Note: Keillor’s essay, in two slightly different versions and with different titles, has appeared in many places, including the following: Salon (Jan 29), Arizona Daily Star (Jan 29), Chicago Tribune (Jan. 30), Green Valley News (Jan. 31); Salt Lake Tribune (Jan. 31), Fort Worth Star Telegram (Jan. 31), Baltimore Sun (Jan. 31), Yakton Daily Press (Feb. 1), San Antonio Express (Feb. 1), Kansas City Star (Feb. 1), Kitsap Sun (Feb. 2).


    Bracey, G. 2006. Reading Educational Research: How to Avoid Getting Statistically Snookered. Portsmouth, NH: Heinemann.

    Darling-Hammond, L. 2007. The flat earth and education: How America’s commitment to equity will determine our future. Educational Researcher 36 (6): 318-334.

    Garan, E. 2002. Resisting Reading Mandates. Portsmouth, NH: Heinemann.

    Krashen, S. 2004. The Power of Reading. Portsmouth, NH: Heinemann; Westport, CT: Libraries Unlimited.

    McQuillan, J. 1998. The Literacy Crisis: False Claims and Real Solutions. Portsmouth, NH: Heinemann.
    Neuman, S., and Celano, D. 2001. Access to print in low-income and middle-income communities. Reading Research Quarterly 36 (1): 8-26.

    Stephen Krashen

    1 comment:

    Anonymous said...

    It has been noted that the "proficient" level for NAEP is about the same as a typical classroom performance level for students whose report card grades fall in the B+ to A grade range.

    For more about this claim and others about NAEP scores see