Last week, the New Jersey Board of Education released information about student performance on the statewide Early Warning Test (EWT) administered earlier this year to all eighth grade students in the state. The EWT is intended to determine which students may have difficulties, three years later, on the eleventh grade High School Proficiency Test (HSPT) that they must pass in order to earn a high school diploma. The "early warning" provides schools, teachers, and parents with the challenge and the opportunity to help students who are having difficulties to overcome them during the ninth and tenth grades.

The statewide results on the mathematics component of the EWT were disappointing. Only about 30% of the students in the state, and fewer than 7% of the students in the thirty special needs districts, performed at a level designated "clearly competent." What do these results mean? Are they evidence that our current eighth graders are less capable mathematically than those of a generation ago? In this article, we will try to present a useful perspective from which to view the results, and to suggest some interpretations.

First, it is important to note that the EWT is a different test than the kind of test that has been used in the past. It is also, intentionally, a more difficult test than the commercial standardized tests used in most public schools, and undoubtedly more challenging than tests in use twenty years ago.

Standardized tests are, by and large, tests for the presence of skills; the EWT tests for application of those skills. Standardized tests involve, by and large, multiple choice one-step problems; the EWT contains many multiple-step problems, of which a number are so-called open-ended problems for which a variety of correct answers are possible. In short, success on the EWT requires students to think, not simply repeat remembered facts and memorized procedures.

The distribution of mathematical content included in the EWT is also different from that of commercial standardized tests, which still spend a considerable portion of testing time on computational skills. The EWT includes problems involving beginning concepts of algebra and geometry, as well as a major emphasis on manipulating and interpreting mathematical data.

In evaluating the recent performance on the mathematics component of the EWT, we must therefore take into account the facts that it is a harder test than the commercial standardized tests that previous groups of eighth graders have taken, that it measures problem-solving and reasoning in addition to procedural skills, and that aspects of more advanced mathematics have only just begun to be included in the test. This combination of problem-solving and skills, applied to a greater mix of mathematical topics, is precisely the kind of mathematical competence our children will need to succeed in the increasingly technological twenty-first century.

Thus we should respond to the results of the EWT not by deploring student performance and bemoaning deterioration of standards, but rather by applauding the increased expectations of the EWT and encouraging students to work hard, persevere, and meet the increased challenge.

The New Jersey Mathematics Coalition supports and applauds the Department of Education for creating an assessment program which accurately measures the skills that our children need, and for reporting the results honestly, without creating an illusion that more students have these understandings and skills than actually do. The clear challenge to the Department now, and to the educational community at large, is to find a way to provide access to that knowledge to all students in the state.

An important context for this discussion should be the efforts currently underway to set high standards for all of our students and to provide schools and teachers with the support they need to ensure that all students meet those standards. The new EWT is really just one local manifestation of a strong and vibrant national movement to improve mathematics education; the EWT is part of the solution, not part of the problem. The new national and state standards that define what constitutes an appropriate K-12 mathematics education direct districts and teachers to incorporate more discussions of data, algebra, and geometry in the earlier grades, to decrease the traditional emphasis on out-of-context computational skills, and to use reasoning and problem-solving as the focuses of the school curriculum.

Like it or not, tests drive instruction. Unfortunately, the message currently being sent to the teachers and students of New Jersey is mixed. In addition to the forward-looking and progressive EWT and HSPT, the present system also mandates that commercially available standardized tests be used every year between third and tenth grade to monitor student progress. This requirement encourages districts to focus on low-level, primarily computational, skills. In many classrooms, in the early spring, all other instruction stops and teachers and students devote their attention to drilling for the test. The teachers and children learn that the standardized tests are what is important, and all of our talk about the importance of problem-solving, of reasoning, of connecting mathematics with the real world, of communicating about mathematics, becomes just talk.

In order to support the reforms necessary in mathematics education, we need to adopt a coherent assessment program that tests what we value and abandon requirements that work against our goals. The new EWT, by its focus on problem-solving and by emphasizing content areas that have been traditionally neglected in the early grades, will provide strong impetus to districts to re-examine their 5-7 grade curricula and ultimately, if not sooner, align them with national and state standards in mathematics education.

However, the kinds of mathematical reasoning and problem-solving skills that students need to succeed on the EWT must be nurtured beginning in the primary grades. We therefore support the creation of a 4th grade assessment program that reflects what we value -- one that would provide the same impetus to improve mathematics education at the K-3 grade levels that the EWT now provides to improve mathematics education at the 5-7 grade levels.

With state standards and a coherent statewide assessment program in place, we can expect that in a few years we will be able to point to tremendously improved EWT-Math scores for our eighth graders -- scores that reflect the attainment of and facility with mathematical knowledge that we truly value.

Joseph G. Rosenstein is Director and Warren Crown is Associate Director of the New Jersey Mathematics Coalition. Both are on the faculty of Rutgers University, New Brunswick.

These two problems from past EWTs show the kind of thinking and mathematical skills that 8th graders must demonstrate on the test.

1. The graph below represents Erin's walk from home to school. Which of the following is most likely to have happened between time x and time y?

A. Erin stopped several times at different traffic signals.

B. Erin ran to catch up with a friend.

C. Erin went back home to get a book.

D. Erin stopped and talked with a friend.

2. Using a straight edge, draw on the grid [both are provided in the test materials] two non-congruent rectangles with perimeter 20 units each. Indicate the length and width of each of your rectangles by writing the appropriate number of units next to sides of each them.