The nation's governors have continued to lead the chorus of those who speak forcefully on the need for high national outcome standards defining what all students should know and do so that they can live, work, and compete successfully as adults. Working through the bipartisan National Education Goals Panel, they reaffirmed this position in November 1993, approving a statement of five general principles that should guide adoption of national standards.[25]

First, national outcome standards should be voluntary. No federal effort should be inaugurated to require states and communities to use them.

Next, standards should address core academic areas. They should not deal with nonacademic concerns such as students' values, beliefs, attitudes, and behaviors.

Third, national standards must be world class. While being uniquely American, the standards should be as rigorous as what other countries expect of their students.

Fourth, policymakers should use a broad-based, participatory, consensus-building process to develop outcome standards. This "bottom-up" process should include educators, parents, and community leaders.

Finally, standards must be useful and adaptable. States and communities must be able to design their own curriculum plans using the broad outlines suggested by the standards. And the number of standards should be limited to the most important knowledge, skills, and understandings we expect students to learn.

These standards clarify expectations about what we want all our children to learn. They are of two kinds. Content standards define what students should know and be able to do. Performance standards define what level of learning is good enough. Several national efforts now underway are doing useful work in developing these standards. But states and communities are where the action should be in defining these standards. It is at those levels that disputes are likely to be encountered and need to be resolved.

None of this should involve standardization or a national curriculum. There must not be any federal demand to read certain books, teach specific courses in a fixed order, or meet precise federal graduation requirements. These and other issues should be decided by states, communities and private citizens.

Also, good tests are needed to determine whether and how well students are learning what is taught. These exams, however, would not be more of the same standardized tests we have now, where many of our children live in a Lake Wobegon world where most are above average. These would be tests that teachers teach to and that have consequences for graduation, employment, and higher education. For those who work in the schools, compensation and advancement would be based on these results.

These tests examining the academic learning of students also need to permit individual student results to be compared across schoolrooms, schools, districts, the states, the nation, and internationally. And although tests are an important indicator of success, they are not the only ones. Other types of timely, reliable, and comparable information such as college entrance rates and placement in the workforce must be collected and made understandable to the public.

We should begin using the best testing tools we have rather than wait for the perfect tests. To assure fairness, such a system should be implemented in phases and be independent of those who govern, manage, and teach in the schools.

Almost every modern society the U.S. trades or competes with has woven these elements into its education system. We owe at least as much to our young people, especially those we label "at risk and disadvantaged." For them in particular, expectations in school are almost always low and the curriculum watered down. Demanding less of these children does not provide equity. It is served by demanding much of all and helping everyone meet these standards.

In summary, standards and tests comprise one element of a sensible two-fold education policy strategy we should pursue as a nation.