Maintaining a precise, invariant unit in state, national and international educational assessment

学校成绩测试对州和国家政府来说是一项高风险的活动，而不仅仅是对学生来说。重要的教育政策决定取决于对不同时期、不同州和不同国家的考试结果的比较。这些决定基于对国家测试的有效性和准确性的假设。特别是，目前的测量系统假设评估量表有共同的单位。从经验上看，很明显，许多因素会损害这一假设，使其很少被证明是合理的。这项研究将通过建立解决这一问题所需的理论和技术来服务于国家利益。

School achievement testing is a high stakes activity for state and national governments, not just for school students. Significant educational policy decisions turn on comparisons of test results over time, and among states and nations. These decisions rest on assumptions about the validity and precision of national testing. In particular, current measurement systems assume that assessment scales have common units. Empirically, it is clear that many factors can compromise this assumption, making it rarely justified. This study will serve the national interest by building the theory and technology necessary to solve this problem.