Experts from around the world gathered at TC in March to debate whether standardized tests are used in fair and valid ways .
Cheating on U.S. standardized tests appears to be on the rise, even as the testing industry introduces tighter security measures each year and the penalties for cheaters grow ever more severe.
So, what to do?
To Eva L. Baker, an assessment expert at the UCLA Graduate School of Education and Information Studies, the answer is simple: publish the questions — and the answers – in advance. Goodbye to safeguards that don’t work, Baker says – and hello to collaboration and peer learning among students.
Baker was among 28 invited presenters and stakeholders in late March at “Educational Assessment, Accountability and Equity: Conversations on Validity Around the World,” a conference co-sponsored by the Assessment and Evaluation Research Initiative (AERI) at Teachers College and the Educational Testing Service (ETS). Proposed and co-organized by Madhabi Chatterji, Associate Professor of Measurement-Evaluation and Education, and Director of AERI, the conference drew over 250 assessment experts, researchers and educators from around the world, including representatives of the prestigious International Association for the Evaluation of Educational Achievement (IEA), the World Bank, and UNESCO.
Standardized assessments are being used to evaluate and compare students, teachers, principals, schools and entire nations, with far-reaching fiscal and policy consequences. Yet the validity of test scores “frequently and repeatedly breaks down,” Chatterji told the assembly, because of a “widening gap” between assessment professionals who work in the rarified world of test development and psychometrics research, on the one hand, and non-technical end- users such as policy-makers, educators, the public and the media. “The validity of assessment information depends not only on how tests are designed and formally validated,” Chatterji said, but on how the results are “eventually put to use in everyday practice, policy, research, or other – sometimes politically charged – contexts.” <Read more.>