Friday, May 22, 2015

Translating Chapter 7 of the Gordon Commission report

Throughout the readings, the concept that kept jumping out at me was that "effective formative assessment causes large improvements in learning" 
Black and Wiliam (1998)


So much talk today is about summative assessment, high-stakes tests that compare our schools to those neighboring towns and schools halfway across the world. As I was reading, my mind wandered back to my youthful impression of the Iowa tests in Elementary school. I considered myself smart, and I wanted to be placed with the smart kids, so I took the tests seriously. At the time, I did not understand that these were formative tests, instructing teachers about skills gaps and missing concepts. The test was designed to form instruction.

I chose to read Chapter 7 of the Gordon Commisson report because it focused on the best that summative testing using technology can be. There are 13 different components that summative tests like the new SBAC look at. I will focus on three.

1. Provide meaningful information: Data collected from the results of the tests must be "trustworthy and actionable." The results must be true and have practical value. Information from high stakes tests is used in studies like the PISA report, which compares countries from around the world with one another. It has become like a competition. It is important we have the correct data to make the comparisons. 2. Satisfy multiple purposes 3. Use modern conceptions of competency as a design basis 4. Align test and task designs, scoring and interpretation with those modern conceptions 5. Adopt modern methods for designing and interpreting complex assessments 6. Account for context: The answers to the tests are facts. The reasons why the student may have given the answers are subject to interpretation. Data about each school - demographics, ethnicity, free lunch recipients, etc. - can cause a child to answer differently than peers in other contexts. This must be taken into consideration during testing. 7. Design for fairness and accessibility 8. Design for positive impact 9. Design for engagement: There is much discussion that students must be engaged in order to learn. Similarly, when students are engaged in their assessments, they do better. Using gaming principals can help make tests seem more relevant. 10. Incorporate information from multiple sources 11. Respect privacy 12. Gather and share validity evidence 13. Use technology to achieve substantive goals

No Child Left Behind became synonymous with "frequent testing" and teaching to the test. Common Core and SBAC are trying to remedy that with more rigorous standards and adaptive testing. Testing using technology at this time still uses easy-to-score-questions. As test designers become more familiar with what works, they will be better able to use the above best practices.

Resources:

Bennett, R. (2013). Preparing for the Future: What Educational Assessment Must Do. Gordon Commission on the future of assessment in education. 123-141.

Kahl, S. (2015) Technology and the future of assessment: Pitfalls and Potential. Measured Progress.


No comments:

Post a Comment