accountability system will provide the incentive to motivate school leaders to determine what works best in their context Other questions remain: Should states allow school districts to have local assessments: can there be a mix of criteria for assessment: car assessments differ from year to year? What do we do in the two states that have provisions in their state constitutions that forbid statewide testing? What about Iowa and Nebraska, which have local control systems which each district creates or purchases its own assessment? There are other states that plan to do state as sessment in some grades and local assessment in others. The overarching question is how do we ensure that all the assessments are comparable? The federal policy is that the burden of proof is on the states to demonstrate that the ir assessments are adequate Implementing accountability The other big issue with assessment has to do with accountability. i know that there are some who believe that what we ought to be looking at is not where students stand at the end of the year but how much progress they have made since the preceding year. Because all students begin the chool year at a different level of achievement, the critical question is how far forward the school is able to move the student in a year. this has been dubbed the value-added" approach The intention of Congress is that we should pay special attention to the amount of progress that occurs over several years, because a single test is simply not sufficient to evaluate student achievement or a schools quality. I agree that we can learn much from the value-added approach, but it is not sufficient to ensure that all students meet a high standard. We should not be satisfied that children who are trailing their peers by several years in achievement simply improve some what year to year without ever catching up. We want all students to reach a rec zed standard States will have some flexibility in demonstrating that their students are meeting a high standard, but they have to demonstrate that all subpopulations are doing well. It will not be enough to raise the average score by raising the performance of the best students even higher That is not the spirit of No Child Left behind. The requirement that 100accountability system will provide the incentive to motivate school leaders to determine what works best in their context. Other questions remain: Should states allow school districts to have local assessments; can there be a mix of criteria for assessment; can assessments differ from year to year? What do we do in the two states that have provisions in their state constitutions that forbid statewide testing? What about Iowa and Nebraska, which have local control systems in which each district creates or purchases its own assessment? There are other states that plan to do state assessment in some grades and local assessment in others. The overarching question is how do we ensure that all the assessments are comparable? The federal policy is that the burden of proof is on the states to demonstrate that their assessments are adequate. Implementing accountability The other big issue with assessment has to do with accountability. I know that there are some who believe that what we ought to be looking at is not where students stand at the end of the year but how much progress they have made since the preceding year. Because all students begin the school year at a different level of achievement, the critical question is how far forward the school is able to move the student in a year. This has been dubbed the "value-added" approach. The intention of Congress is that we should pay special attention to the amount of progress that occurs over several years, because a single test is simply not sufficient to evaluate student achievement or a school's quality. I agree that we can learn much from the value-added approach, but it is not sufficient to ensure that all students meet a high standard. We should not be satisfied that children who are trailing their peers by several years in achievement simply improve somewhat year to year without ever catching up. We want all students to reach a recognized standard. States will have some flexibility in demonstrating that their students are meeting a high standard, but they have to demonstrate that all subpopulations are doing well. It will not be enough to raise the average score by raising the performance of the best students even higher. That is not the spirit of No Child Left Behind. The requirement that 100