FOR RELEASE Thursday, August 2, 2001

Contact: Lawrence Feinberg (202) 357-6938

The NAEP 2000 Mathematics Assessment, whose results are being reported today, is an important landmark for the National Assessment of Educational Progress. It is the fourth test in a ten-year period assessing the same math content and standards. It is also the first time we can use NAEP to look back over a decade to see not just how well the country has progressed, but also to find out what has happened in particular states.

As Commissioner Phillips has said, it was in 1990, at the start of the past decade of ferment and reform, that NAEP began testing representative samples of students in the states, and the state-level assessments started with eighth-grade mathematics. The state samples in fourth grade math began in 1992.

Also, this marks the tenth anniversary of the achievement levels, set by the National Assessment Governing Board. These provide a uniform national benchmark of basic, proficient, and advanced performance for each grade and subject tested by National Assessment exams. The first achievement levels set by the Governing Board were for the 1990 assessment in mathematics.

Generally, as you already know, the news in math achievement has been positive over the past decade, except for a troubling drop over the past four years at grade 12. Even with that decline, 12th graders are ahead of where they were in 1990. But clearly the nation has made much greater progress in teaching math to young students than it has in teaching math in the high schools. And unfortunately, this pattern has shown up repeatedly on international comparisons with U.S. students doing relatively well in math in the early grades but falling toward the bottom by the end of high school.

For grades 4 and 8, however, the new NAEP results are good, though clearly we have a long way to go until all students reach the Proficient level. Over the decade the gains in these grades have been steady. They have occurred for all groups of students-whites, blacks, Hispanics, and Asians, in both public and private schools, and in almost all of the states for which we have NAEP data over the decade.

One effect of this general rise in average math achievement is that most of the achievement gaps—by race and region and gender—have stayed about the same over the past decade, although it should be noted that the gender gap has fluctuated and is really quite small.

However, when we look at the results by achievement levels, the picture is more complex. The greatest gains have been in the proportion of students reaching the Basic achievement level; there have been smaller gains at Proficient and Advanced. This has been true nationwide, and in most of the participating states.

Let me speak a little about the achievement levels. These are performance standards for NAEP. They are set by the Governing Board based on recommendations by panels of experts—mostly classroom teachers—from all across the country. The process for setting the levels is complex, and has been managed for the Board by American College Testing (ACT), which has enormous experience in the field.

The central level for each grade is Proficient, which the Board has defined in general terms as "solid academic performance" that shows "competency over challenging subject matter." This last is a phrase and a concept that the Board took from the National Education Goals, which were formulated in 1990 jointly by our first President Bush and the nation's governors, whose association was led then by Gov. Bill Clinton, of Arkansas. The Basic achievement level shows partial mastery over skills that are fundamental for Proficient performance and Advanced signifies "superior performance."

For each achievement level on this assessment there is a general description of the mathematics knowledge and skills that students should know and be able to do at a particular grade in order to meet the policy definition. There also is a cut-score on the test for each level and sample items that illustrate what each level means.

Obviously, setting achievement standards is a matter of judgment and judgment is always open to debate. Our Board has tried to make sure that its judgments are well informed and that the process we use is creditable. Proficient is a strong but reasonable standard that we would like all American students to reach. It is quite different, and higher, than average "grade-level" now, and it will take some time to get there.

The Basic achievement level is higher than minimum competency though less we would hope for. But it does mean that students are on their way, and over the past decade many more have crossed that mark.

To get a better idea of what is happening, let's look at some tables on the trends in performance at the different achievement levels-first nationwide and then in the two states where I have taught during the past ten years. Please see the exhibits at the end of this statement.

Back in 1990, I was living in Anderson, Indiana, and taught at Anderson Community Schools Corporation. Since August 1994, I have been teaching at Hornedo Middle School in El Paso, Texas. So I will show you the detailed data for Indiana and Texas to illustrate what has happened.

These three tables show that in the national assessment and in both states there have been substantial increases in the percentage of students scoring at or above the Basic achievement level. The proportions at Proficient and Advanced have also increased, but in nearly all cases by not nearly as much.

With only a few exceptions, the rate of progress for minorities has been greater than that for whites, but the gaps are still wide, and in the national samples the percentage of whites meeting each standard has increased more than the percentage of either blacks or Hispanics.

In both Indiana and Texas, however, black and Hispanic students had a larger increase in the percentage at or above Basic than did the white students. This means that the racial gap at the Basic level is closing even though at the Proficient level and for average scores that usually isn't so.

In some cases, the gains and the differences are quite striking. For example, among fourth graders in Texas from 1992 to 2000, the percentage reaching Basic climbed from 29 to 60 percent for blacks and from 43 to 68 percent for Hispanics. For whites the proportion reaching Basic also went up—from 72 to 89 percent, but that increase of 17 percentage points was much less than the 31 point increase for blacks and the 25 point increase for Hispanics.

In the proportion reaching Proficient, both blacks and Hispanics had a much greater rate of increase than whites, but in terms of the percentage reaching the standard, whites improved more—by 18 percentage points, compared to 9 points for blacks and 7 points for Hispanics. So by this measure the gap has grown. The pattern is similar in Indiana.

What is going on, I think, is that both states—and especially Texas—have focused on teaching and testing for basic skills among the students who are behind. While the Texas standards and testing program, the TAAS, certainly hasn't been perfect, it has given teachers a very clear idea of what their students should learn in each grade and it has forced us to pay attention to the lowest-performing groups. What NAEP gives us is an independent measure, not just the state's own test, to show that there has been substantial improvement. Of course, NAEP also shows that Texas has a substantial way to go to equalize educational achievement. But at the Basic level, which is important, we have made important gains.

Now I would like to discuss calculators, a topic of considerable interest to teachers and of controversy among the public. According to the background information NAEP collects from students, there is a substantial difference in the relationship between calculator use and math achievement at grade 4, compared to grades 8 and 12. Generally, at grade 4 using calculators more is linked to lower achievement. At grades 8 and 12, the students who use calculators more tend to do better, and the positive relationship is especially strong at grade 12.

As you know, the National Assessment is a survey; it does not follow the same children from grade to grade. Thus, it can show the correlation of one factor or another to student achievement. But it cannot be used to prove cause and effect.

But I think there may be several reasons for these relationships. An important part of fourth grade is for students to gain fluency in adding, subtracting, multiplying, and dividing whole numbers and also to develop the strategies for estimating answers and judging the reasonableness of their results. It is important for them to learn the algorithms to compute in whole numbers, and to record their computational thinking with pencil and paper. This fosters an understanding of the base-ten number system that a calculator cannot provide.

Interestingly, there was a significant drop from 1996 to 2000 in the proportion of fourth-graders using a calculator every day for classwork or homework. The proportion using a calculator once a month stayed about the same, and this group of occasional calculator-users performed as well or better than students who never used one.

In 8th and 12th grades, on the other hand, the use of calculators every day was associated with higher scores. Clearly, calculators are necessary for more advanced math, and the use of graphing calculators is expanding, even in eighth grade. I know in my own classroom that using graphing calculators lets students link tables, graphs and equations, and helps them test conjectures. It also is a great way to get them involved and excited about some fundamental concepts of math.

Certainly, technology should not be a replacement for teaching basic facts or developing basic understandings and intuitions. But it can foster these understandings and help build on them, and it has now become essential for higher-level mathematics.

The NAEP 2000 Mathematics Report Card has an enormous amount of information, but I think its central message is clear. Over the past decade, we have made substantial progress, but we still have a great deal to do.