International Benchmarking of Student Achievement Eric Hanushek There has been considerable discussion about the advantages of benchmarking the performance of American students in various states and localities to international tests. In simplest terms, this is something we should support because it would provide new and important information to both states and localities. This new information would also provide added impetus to the imperative to improve our schools. The U.S. is in the throes of developing new standards and new tests of student performance, actions that reflect a general dissatisfaction with the level of student achievement. Much of this movement is hooked to a focus on better preparing students for college and work – a focus partly emanating from the consensus opinion that we must improve our human capital if we are to be internationally competitive. But the available (and prospective) information on student performance is extraordinarily hard to interpret. The states can get some information about student performance through the NAEP tests, and local districts can find out how they are doing relative to other districts in the state through state accountability measures. And, while states or localities that perform relatively low on their specific tests know that they could improve, higher performing districts or higher performing states often develop a complacency that has serious side effects. This general complacency about performance is quite unwarranted when information from international competitors is added. Students from Massachusetts, our best performing state, place 17th in international league tables for advanced math skills. And, by Jay Greene’s calculations, even the best districts tend to lag the international competition. Yet, none of this has made it far into the public consciousness or into the policy makers focus. Consider now giving a sample of students in a district or in a state the PISA examination – the international examination given throughout the developed world. The primary advantage of this exercise is comparing student performance to “what is possible.” As such, it personalizes and makes more understandable the educational challenges we face. And, it likely helps in developing a broader constituency for improvement. Most educational standard setting, performance assessment, and judgments about appropriate levels of achievement today are based on history and custom with a little bit of “professional dreaming.” The process generally lacks any context of what our international competitors are doing. Why do local benchmarking? The overall performance challenges to U.S. students can of course be seen from the national statistics for PISA or TIMSS, the regular international assessments. But, states and localities, with their own tests and outcomes, can simultaneously retain their own views of performance that is not anchored in the reality of other locales, particularly the international competition. Ultimately, it is easy to discount or ignore any deficits. The local testing makes it clear what can be done and what needs to be done in order to compete with other nations. It is an unambiguous statement of the level of performance. The state and local assessments can also provide localities with some idea of how improving their labor force might affect future economic growth. The national implications of improved schooling are both clear and large. While local areas must worry more about the migration of skilled labor to other areas, the long run impact of improved schools cannot be ignored. And, as such better information can foster and reinforce pressure for improvement from local and state groups that see the need to reform the schools. From a national perspective it is likewise very important that states and districts understand this challenge of international competition – because of the primacy of states and districts in education policy and implementation. Improvement will come from the actions of the states, and these actions are not readily dictated from the national level. What benchmarking does not do? It would be wrong, however, to use the international tests to design a local curriculum. Test design is often linked to a starting point of clear standards, and curricula define how learning and instruction can achieve the goals imbedded in the standards. The opposite is not the case. One would not want to take the international assessments – which are attempts to develop tests that are not dependent on specific national curricula – and then try to reverse engineer standards or curricula. The presumption of the international assessments is that achieving high performance on the local and national standards will also produce high scores on PISA or TIMSS. The international tests give a direct measure of what high performance means. What’s at stake for the nation? While approaching the level of a cliché, it is simply the case that competition is not just with others in the same city or same state with the internationalization of the world economy. International trade moves the location of economic activity to the place that has a comparative advantage in production. Historically, the U.S. has dominated the world in high-skill, high-value-added production, in large part because its workers have had more human capital when compared to workers in other countries. This dominance in worker skills appears to be ending, as evidenced by performance on PISA exams that is below the average for developed countries. This skill gap forces us to rely on other national advantages if we are to continue as the world economic leader. We do have the best economic institutions – free and open labor and capital markets, secure property rights, and limited governmental intrusion. We also have strong colleges and universities, and we have been open to accepting immigrants with high skills. But, each of these advantages is fading and is unlikely to carry us in the future, as other countries are moving to address these issues themselves. A future in which the U.S. is no longer the high-skill country and in which other countries increasingly are the innovators does not mean that U.S. workers will be unemployed or that they will do only menial jobs. It does imply slower economic growth in the U.S. and a resultant stagnation in incomes. It also implies a noticeable change in the locus of world economic activities and leadership.