Despite the substantial investment in developing high standardsand "minds-on," "hands-on" curricula, there is limited empirical evidence that large-scale science assessment programs have adequately tested what students know and can do. The Third International Mathematics and Science Study (TIMSS) and the National Assessment of Educational Progress (NAEP) rely heavily on multiple choice test items and brief open-ended questions. Such formats have been criticized for focusing on recall, requiring limited levels of inferencing, and for testing decontextualized, discrete skills that are not integrated in real-world task. Thus, these formats may be poorly suited for measuring elaborated conceptual understandings and key components of scientific inquiry.

• Develop a more concise picture of the National Science Education Standards measured by the items and task formats used on the nation's most influential reference examinations
• Study how item format contributes to variation in student performance
• Interpret unexpected or discrepant results among students' performances on NAEP and TIMSS
• Determine what additional types of assessmentsystems and designs might portray student progress in science more completely than current reference examinations