I am pleased to be able to speak with you today, but I'm afraid the news we have to present is disappointing. After a considerable amount of effort over the past two decades, the reading achievement of students across the country shows very little change.

Yes, there was a slight uptick at grade 8 in 2009, but the average score last year was just about the same as it was back in 1998. And since the current series of NAEP Reading Assessments was first given in 1992, the percentage of eighth graders who are rated Proficient in reading has risen only slightly, from 29 to 32 percent.

At grade 4 there was no change in average scores from 2007, and this is especially disappointing because the NAEP math results, released in December, also showed no change from two years earlier. It's too soon to tell why progress in the early elementary grades has stalled. On the NAEP Assessment, as on any large-scale academic exam, it's important to look at trends over an extended period of time to understand what really is happening. By that measure, there has been a slight improvement in fourth grade reading over the past decade—from 29 percent Proficient in 1998 to 33 percent Proficient last year. But over the previous six years there was no change at all.

What the NAEP Assessment shows us over almost two decades is that in reading there have been only slight gains and no sustained trend of improvement.

The situation in mathematics is quite different, even with the flat fourth grade results last year.  Since 1990, when the NAEP Math Assessment was first given, the average score for fourth graders has risen by 27 points. That is more than half of the 50-point difference between fourth and eighth graders in the first year of testing—an enormous gain. Another way of looking at this is by the percentage of fourth graders rated Proficient in math, which rose from 13 percent in 1990 to 39 percent in 2007 and 2009. In eighth grade math, the percentage of students reaching Proficient rose steadily over two decades from 15 percent to 34 percent. In my own state of West Virginia, there also are major differences in the pattern of achievement changes between reading and math.

Why are there such sharp differences in these two major subjects, which, after all, are taught to the same students in the same schools? I don't think these achievement trends can be explained by the demographics of the students in the schools or by how the schools are governed—by local school boards or states or as independent charters, or even by the accountability programs that have proliferated over the past two decades. For changes like these, which have been substantial, affect all of the students and teachers in a school. You might expect them to have about the same impact on achievement in all subjects. But that's not how achievement has trended, according to NAEP. The difference in achievement gains is striking between reading and math.

For better or worse, NAEP itself cannot tell us why this happened. As Deputy Commissioner Stuart Kerachsky has explained, the National Assessment tests representative samples of students at different points in time. It doesn't follow the same students in different settings over a number of years, and it is not designed to prove cause and effect.

I think the best explanation for the differences is in the nature of the subjects of reading and math themselves and in how the skills that NAEP tests are developed.

As I think everyone would agree, mathematics is a subject taught primarily in school and primarily in mathematics classes, while reading comprehension, which is what NAEP tests, is a skill developed not just in a reading or English class, but through the reading and writing students do in all subjects throughout the curriculum. Also, the reading and writing that students do outside of school has a powerful influence, while in math, understandably, students do very little math other than class assignments. This makes reading achievement, much more than math, a shared responsibility both within the school and, more broadly, in society. Value must be placed on sustained reading and sustained analysis, not just on e-mail snippets and sound bites.

In mathematics many states have developed a cohesive and sequential curriculum, based in many cases on national curriculum standards, developed by the National Council of Teachers of Mathematics and adopted in the major mathematics textbooks. In reading there has been no similar agreement and no similar cohesion or emphasis. Even worse, according to research assembled for the Common Core State Standards Initiative, in which I have been involved, the difficulty and complexity of school textbooks in many subjects has trended downward for several decades.

The only area of reading where there has recently been an emphasis and some agreement is in early reading instruction. This also is the only aspect of reading that is almost entirely a school subject. And it is only here, at fourth grade, where students are supposed to have mastered basic skills, that NAEP shows substantial gains during the past decade. Since the year 2000, the percentage of students reaching the Basic achievement level in fourth grade reading has climbed from 59 percent to 67 percent. The improvement in basic reading skills is also reflected in the 15-point gain in average scores that both black and Hispanic students have made in fourth grade reading since the year 2000. The average score for white students remains considerably higher, but it has risen considerably less. 

It is true that the proportion of students identified as limited English, most of whom are Hispanic, has grown over the past 20 years, but nearly all of the increase occurred before 2003. Among non-Hispanic whites there has been almost no change in reading achievement at both the fourth and eighth grades since 2002 and only a 5 point rise in the decade before that—far less than the gains these students made in math.

Of course, the schools have no control over immigration and very little immediate impact on social and economic trends outside the classroom, but they do control their standards and curriculum. Here there are some promising developments. The Common Core State Standards that I mentioned earlier are being developed by the National Governors Association and the Council of Chief State School Officers. I am president-elect of the CCSSO and will be testifying before Congress tomorrow about the proposed standards.

The draft we have sent out for public comment does not have standards for reading per se but rather for what we call English Language Arts and Literacy in History/Social Studies and Science.

The basic idea behind this mouthful is that reading and writing cannot be taught well as skills in isolation, but must be developed as part of a full, demanding curriculum. The common core standards draw explicitly on the framework for the 2009 NAEP Reading Assessment, particularly on NAEP's increased emphasis on informational text. Like NAEP, the common core standards also would require students to integrate and analyze what they read, not just pick out discrete facts, and to support their judgments with evidence and examples. These are the complex, analytical skills that NAEP tells us many students have difficulty with now.

We hope the common core standards will be adopted and implemented—voluntarily by the states—as part of a transformation of the content and testing in our schools. In time, we believe, this richer curriculum should produce gains on the NAEP Assessment of reading comprehension, and NAEP, in turn, could serve as an external monitor of what the common core standards achieve.

As you know, I have great respect for NAEP and the integrity of its data. However, I remain concerned, as I have for a number of years, about the continuing differences in exclusion and accommodation rates in the different states for students with disabilities and English language learners. According to the new Reading Report Card, West Virginia tests almost all the students selected for its NAEP reading samples. Our exclusion rate is just 2 percent of the sample in both the fourth and eighth grades, compared to 9, 11, and 12 percent in some other jurisdictions. These discrepancies do not seem fair. Inevitably, they raise questions and concerns. 

At its meeting earlier this month, the National Assessment Governing Board adopted a new policy to reduce exclusions and make the inclusion rate on NAEP Assessments much more uniform among the states. I was a member of the Board's ad hoc committee that worked on the new policy for almost two years. I hope we will begin to implement the policy promptly and show that NAEP is dealing with this concern.

Thank you very much.