
Within the current debates on education policy, there is widespread agreement among teachers, policymakers and testing experts that the current multiple-choice-dominated K-12 tests used in the United States are in need of improvement. U.S. Secretary of Education Arne Duncan’s recent announcement that he will set aside $350 million of the $4.35 billion in discretionary aid to create Race to the Top Fund to improve assessments has these stakeholders debating not if they can improve assessments, but how they can improve assessments.
Testing experts such as Randy E. Bennett of the Educational Testing Service say that the money could help revise tests that would better measure students’ critical thinking skills and improve teacher and student engagement in the assessment process. However, such a revision may force federal officials to rethink the current parameters around assessment and accountability in the current No Child Left Behind (NCLB)law.
Multiple choice tests have largely become analogous with the NCLB law. One reason why multiple-choice tests are prevalent is that they can efficiently determine whether a student can assemble discrete pieces of knowledge across a subject. The results are typically highly reliable, which is desirable for the high-stakes nature of NCLB tests.
However, multiple-choice tests are not ideal for identifying whether students can take multiple pieces of domain-specific knowledge and analyze, integrate and apply them in unfamiliar contexts. International student assessments such as the Program for International Student Assessment (PISA) include such items. Assessments like the PISA measure abilities that are common in applied fields such as medicine where examinees are required to diagnose and treat patients. A drawback of performance-based assessments is that they are scored by humans, which raise the costs of examination.
I anticipate much debate on how to use the funds provided via the Race to the Top Fund. Questions that may be part of this debate:
• How broadly the material will be examined?
• How to balance critical thinking skills to basic knowledge assessment?
• What will be the length of the assessment?
Also likely to be debated are ideas such as the extent of technology-based testing and whether revisions to these assessments will need revisions to application of NCLB.
The Improve Group has conducted multiple assessments in the areas of education and we have found that a comprehensive approach to assessments is generally best when measuring student outcomes. However, a debate about improving current testing methods provides helpful opportunities for both discussion and analysis.
I would appreciate hearing any ideas you may have regarding revisions to current educational assessment methods.