EDUCATORS ARE still mulling over the results of national standardized test scores released last month, which showed an unusually clear national trend: While there have been some slight improvements in elementary and middle school math scores, the rate of gain is slowing. Meanwhile, reading scores are stagnant. These results contrast sharply with the scores of many states' own standardized tests, which purport to show clear gains. Already, some have pointed to this gap as evidence that the No Child Left Behind law, the president's plan to make states set standards and show annual academic improvement, isn't working.
In a narrow sense, the critics are right: The gap is indeed evidence that many states are still using tests that are too easy, and they have not faced up to the genuinely difficult challenge of improving their schools. But the gap does not negate the value of using standards and high-stakes testing to improve student performance. On the contrary, many of the states that have shown the most improvement are precisely the ones that have been using statewide standards for the longest period. One is Massachusetts, which has long used testing to measure achievement and is now at the top of the country in both reading and math. Another is Virginia, which, thanks to its Standards of Learning tests, has also made gains on national tests over the past five years. States that have not been using standards for quite so long have not done so well: In Maryland, for example, some scores have dropped since last year.
The gap has also set off a discussion of what, if anything, can be done at the national level to help states raise their students' achievement levels. Some are advocating the setting of national standards, a proposition that sounds nice in theory but seems politically impossible in practice. For the first time, there is a national discussion of teacher quality, too: what it means, how to define it, how to improve it. The administration has launched teacher mentoring and teacher "e-learning" programs. Several states have reexamined their teacher training requirements. Others are arguing that states need to rethink their pay scales: The Education Trust recently surveyed teachers within a single California school district and discovered that the most experienced teachers, and therefore the best paid teachers, were all in wealthier schools. Inequality persists not only between districts, in other words, but within them.
The scores should also cause educators to think about the deeper causes of low student achievement. Teach for America, the charity that sends high-achieving college graduates to teach in low-income schools, recently published a survey of its alumni, who overwhelmingly believe that schools underrate children, fail to challenge them and resist imposing higher standards because they simply don't believe the students will meet them. Higher expectations, Teach for America argues, can actually lead to higher test scores.
Standardized math and reading tests are, by themselves, not sufficient to improve American education. But without a recognition that higher standards are needed, improvement isn't even possible.