A Modest Proposal
This post is by Bob Lenz, co-founder and Chief Executive Officer of Envision Education
Recently, the College Board announced major revisions to the SAT, designed to more accurately assess college readiness. While this step acknowledges the importance of getting assessment right, the SAT test itself is still a deeply flawed approach to predicting college success. One test, no matter how many times it's taken, is only a proxy for college readiness; only actual student work can demonstrate real readiness for college-level work.
Even with the so-called bold revisions, the inherent problems with the SAT persist:
- No matter the test, people will game it; affluent kids will still have an advantage.
- Free SAT Prep is not the only consideration. Although the free SAT prep promised by the Khan Academy is clearly a positive thing, low-income kids without high-speed internet access will not have equal access. Furthermore, watching a video lesson and being coached 1:1 in person are not equal forms of access.
- Classroom instruction will still be dictated by the test. Is it really the best lever for preparing kids for college success? Wouldn't it be better for teachers to base classroom instruction on mastery of academic content and skills as well as on deeper leaning outcomes?
- The SAT is an expensive endeavor--hundreds of millions of dollars--with the goal of offering access to higher education and predicting college success. Recent studies show that grades are a better predictor of such success than the SAT. Retooling the SAT will only perpetuate this mismatch and waste valuable resources.
The solution? I propose that we drop the SAT completely and use those resources to create a national online deeper learning performance assessment system. In this system, a student would submit a portfolio of verified, original work that shows, in breadth and depth, that he or she is ready for college-level work. We can build this system to accommodate the essential requirements and ensure its validity.
- Currently, proctors of the SAT sign agreements that the test was administered fairly. We can do the same in an online assessment system; we can create legally binding methods for teachers to verify that submitted work is original.
- We can create and distribute a collection of performance assessment frameworks and rubrics for students and teachers all over the country to use to guide their work. The associated tasks would integrate deeper learning outcomes, while aligning with the typical tasks of a first year in college. Examples of work students could submit:
- Original research papers that use MLA citation
- Papers describing original scientific inquiry experiments
- Analysis of non-fiction and fiction texts
- Works of creative expression or analysis of art
- We can train and pay reviewers to assess student work, according to the same frameworks and rubrics.
- Students could include a personal statement of purpose that introduces their work and explains how it demonstrates their college readiness. This could also be something they submit with college applications.
- Similar to re-taking the SAT, students would have the opportunity to revise and resubmit their work for assessment, incorporating feedback or responding to weaknesses found by the reviewers.
- Students can include work sample summaries that explain their process, their learning, and the application of the learning to future tasks in college. This would allow them to show that they understand why their high school work matters to their future.
Such an assessment system would drive the kind of teaching and learning we want to see in our schools. If we want teachers to both convey content information and develop deeper learning skills, then we should be assessing students on those measures, and not on whether or not they know how to take a standardized test. And if we want students to "dive deep" into content and skills, then we should give them assessments that allow them to show what they have learned and what they can do. Actual student work, grounded in real learning experiences, will be the predictor of success.
Answering the Objections
- It's too expensive. A kid dropping out of college is far more expensive. With a nod to Frederick Douglass, it will be cheaper for us to build strong students than to fix broken adults.
- It's too challenging. Technological advances make this realistic and achievable. If we can create smart phones more powerful that the computers that put humans in space, then we can certainly create tools to verify, submit, and score original work.
- Teachers do not know how to teach towards a performance assessment. You can be sure that if the SAT became a performance assessment, parents would demand that teachers prepare their students accordingly, and that the Kaplans and Princeton Reviews would quickly adjust and offer learning experiences that would facilitate students completing rich performance tasks.
- It takes too much time to score the work. The College Board already has great systems in place to score hundreds of thousands of AP Test essays. They also have experience scoring Art Portfolios. They can use those same systems and experiences to take performance assessment to the next level.
We Already Know It Works
Examples of proven effective performance assessment are readily available:
- Stanford University recently released a study, entitled Student-Centered Schools: Closing the Opportunity Gap, documenting the successes of four urban high schools in California that are preparing students for success in college, career, and life. These non-selective schools serve populations that are predominately low-income students of color. In just one measure of college success from that study, 85 percent of San Francisco's City Arts & Technology High School's 2009 graduates who enrolled in college persisted for at least four years, compared to a national average of 65 percent. This school, which is part of Envision Education's network, uses exactly the kind of performance assessment tools I am suggesting we adopt as a nation.
- The Deeper Learning Student Assessment Initiative (DLSAI) is a collaborative partnership of organizations with a common mission to support and improve student assessment practice. Envision Learning Partners (ELP) provides leadership to the DLSAI, which includes ELP, Asia Society, ConnectEd, New Tech Network and the Stanford Center for Assessment and Learning (SCALE). The DLSAI is creating tools and resources that aid in the effective implementation of performance assessment. Their work is all about finding the tools and technology that make deeper learning student assessment systems efficient and effective for networks, districts, schools, and teachers.
In addition to the DLSAI members linked above, there are multiple examples across the country of school networks and systems that are implementing deeper learning student assessments. Here are some strong practitioners we can learn from in order to make the shift:
- New York Performance Standards Consortium
- The New England Network for Personalization and Performance
- Center for Collaborative Education
Let's learn from the work of these innovative educators. Revising the SAT is nothing more than tinkering with a flawed tool. It's time to start over. We have the ability to create a system that is both engaging and challenging, and that will accurately assess whether students are ready for college. What's more, we have proof that deeper learning performance assessment works with all kinds of students, including low-income students and students of color.
One question remains: do David Coleman and the College Board have the courage to start over, or will this be another expensive rearranging of the deck chairs?