The great K-12 assessment drama playing out in America was served as the main course--will American lock in gains of Common Core Standards by developing inexpensive common end of year/course exams, or will a new generation of comprehensive assessment systems be developed? Dinner participants included a Common Core leader, an architect of No Child Left Behind, a state commissioner, and a foundation executive.
The answer appears to be a bit of both. As part of Race to the Top, $350 million is at stake. About half the states are focused on less expensive common tests; the other half will venture into the realm of possibility. Lagging state budgets are likely to push more states toward cheap/common rather than new/improved.
These decisions are a big deal for America. Testing regimes developed and adopted in the coming months are likely to drive how schools operate for the next decade--frameworks as important as NCLB was for the last decade.
EdWeek's Catherine Gewertz has been tracking this closely with a full run down here outlining competing camps. In one corner, "Florida-Achieve" wants to lock in the gains of broad support for high standards with less expensive reliable tests. Folks in the "Smarter Balanced" corner want to develop a comprehensive system that includes performance assessment (i.e., writing and projects). Late entrants include a career tech consortium, and updated end-of-course exams from College Board and ACT.
Civil rights, gap-closing advocates, and conservatives (yes, an interesting alliance) line up with Florida-Achieve given their interest in comparable disaggregated achievement data. They worry about the dreamy liberals in the Smarter Balanced camp and suspect they don't really buy high standards and strong accountability.
Like observing open heart surgery from the operating room balcony, I'm watching this unfold having known most of the folks involved this drama for a decade. There are folks in both camps that appreciates the flood of achievement data to come from the digital learning revolution--adaptive quizzes, games, simulations, and virtual learning environments all queued up in personal learning playlists. These new data streams create the opportunity for a new way of thinking about assessment--less high stakes end of year tests and more medium stakes end-of-unit tests with individual kids moving on when they are ready.
We have the opportunity to build forward leaning assessment frameworks that anticipate and encourage the digital learning revolution or we can build more paper tests and encourage production of more #2 pencils. As states think about joining together to build tests for the next decade, the following guidelines would meet most of the interests of both groups:
1. Make better student learning experiences the first priority. Provide incentives for content and assessment providers to provide real-time feedback to students and teachers. Stop buying textbooks and purchase digital content with embedded assessment.
2. Make value-added engines the second priority. Build systems that help us get a lot smarter about what improves student learning. Using all available data, build smarter instructional systems, performance-based employment systems, and growth-oriented school accountability systems.
3. Make kids show what they know. For example, all secondary students should demonstrate expertise in science fairs every year and we should be a lot better about assessing project work in rigorous and consistent ways.
4. Build assessment systems for the future not the past. The age-cohort model is giving way to personalized learning. Our tests should enable not block individual progress models. That means more on-demand online assessment and an end to the stop-everything-you're-doing-in-May-two-day-marathon-bubble-sheet-tests.
5. Leave room for innovation. We'll have an order of magnitude more data in three years. We are close to being able to send an iTouch home with kids and get more achievement data from three hours online that we get in a year today. If your state assessment plan doesn't anticipate data from learning games, hit reset.
6. It's time for online assessment. We've networked every school in America and have a computer for every three kids. Device costs are plummeting. We can build better, cheaper, faster assessment systems and can solve security and access issues.
The Hewlett Foundation recently approved Deeper Learning, a long-term initiative to promote better instruction and assessment. Like Hewlett, I'm a fan of the College and Work Readiness Assessment, an online task that requires students to analyze data and construct an argument.
We can meet the demand for engaging instruction and consistent and comparable achievement data. We can maintain high standards and build learning systems that encourage more thinking and less regurgitating. Let's not blow this.