From the implementation of the Common Core, to the recent debate surrounding teacher tenure, nearly every issue in public education today can be seen as a facet of a single, fundamental policy question: how should we use standardized assessments and the student achievement data these tests produce? With Secretary Arne Duncan’s recent announcement in support of DC’s decision to delay the use of value-added by one year, the issue is coming into increasingly sharp focus.
On one side, there are those who argue that tests are not valid measures of teaching or learning. They say that the tests’ results are solely reflections of systemic inequities, and cannot be used to assess individual students, teachers, or schools.
On the other, there are some who argue that standardized assessments provide the only measure of teacher effectiveness and student achievement. This camp claims that tests, as currently administered, are the only fair and “objective” measure of progress and should determine all or most high-stakes decisions.
Given this dichotomy, I choose ‘none of the above.’ We can acknowledge the shortcomings of our existing system of assessments without dismissing the potential of standardized tests to inform teachers and students and improve our accountability models.
This past spring, I joined a team of 14 teachers from across New York City that put the typical rhetoric aside and paired our collective experience with the existing body of research about standardized assessment. Together, we created a series of recommendations outlined in a new policy paper, “None of the Above: A New Approach to Testing and Assessment,” focused on four key areas for schools, districts, and state policymakers.
Design: Improving the Validity of Assessments. Research has shown repeatedly that standardized tests can provide a meaningful measure of student achievement and that score increases on these tests correlate with positive future outcomes for students. That being said, poorly designed tests will give us poor measures of teaching and learning. In order to ensure well-designed assessments, creators must strive to measure higher-order thinking skills. We also think this will largely address concerns about excessive “teaching to the test” – research suggests that for well-designed assessments, excessive test prep is actually counterproductive while a rigorous curriculum that prioritizes critical thinking is likelier to lead to better student results.
Furthermore, tests should be created with the input of classroom teachers throughout the design process, from start to finish, to actively solicit our feedback on alignment, bias, and logistics. One of the key ways we can monitor test quality is by releasing the vast majority of state test items for public review, a goal which the New York State Education Department moved closer to recently.
Culture: Creating a Positive Learning Environment. Despite their potential to inform teachers and students, in some schools testing has contributed to a negative classroom culture that harms student investment. Teachers must lead the culture change in our classrooms, but we cannot do it alone. We need to audit the number of tests administered throughout a school year and eliminate unnecessary and redundant exams. We were also glad to see the recent expansion of the New York Performance Standards Consortium, which use performance assessments as a means of breaking the monotony of traditional multiple-choice tests without losing the critical data that assessments provide to inform teaching and evaluate performance. Some encouraging evidence suggests that this system works well, but more research should be conducted to ensure that these assessments are compatible with data-driven improvement and accountability.
Teaching: Using Data to Improve Instruction. One of the most significant benefits of assessments is the data they provide that can inform teachers’ practice. Unfortunately, this is a benefit that sometimes goes unrealized. Our schools and districts must invest in high-quality professional development opportunities and data systems that help educators effectively use achievement data to make instructional decisions. Given the specialized nature of this kind of quantitative analysis, we also recommend that schools identify a ‘data specialist’ in the faculty who can support his or her colleagues in using data in real-time. Both of these priorities fit with the New York City teachers’ contract’s emphasis on professional development time and new teacher leadership pathways.
Accountability: Informing High-Stakes Decisions. Perhaps most controversial of all is the use of student achievement growth to make high-stakes decisions. Studies have repeatedly shown that students benefit when schools and educators are held accountable for student outcomes, as part of a multi-measure system. However, we must be careful not to penalize those of us working with the highest-needs student populations, and we recommend using a two-step value-added model in order to ensure that there are no incentives against teaching at-risk students, while identifying and rewarding those teachers that are most successful with such students.
It’s also crucial that any decision-making system that considers student growth and achievement uses them as one of several measures. Assessment data provides a snapshot of information, but not a complete picture. Fortunately, New York City’s teacher evaluation does just that, and so does the new framework for making student promotion and retention decisions. However, one area where New York State has fallen short here is in regards to high-school exit exams: currently, students must pass five state Regents exams (pdf) in order to graduate. Both anecdotal and rigorous empirical evidence suggest that this is a misguided practice that harms our most disadvantaged students. No single test should serve as the sole barrier between an otherwise qualified student and a high school diploma.
These represent just a handful of the recommendations outlined by our policy team. Taken together, we believe we have spelled out an approach to standardized testing grounded in the fact that assessments can gather critical information about our students’ growth and our own teaching practice, while acknowledging that this potential will be lost if we ignore the need for improvements to our current system.
Change will not come easily, and it will require open and honest dialogue between teachers and all stakeholders. We welcome the challenge to get this right for our kids.
— Trevor Baisden
Trevor Baisden is a founding fifth grade ELA and history teacher at a public charter school. He is a member of Educators for Excellence – New York, where he was recently part of a Teacher Policy Team on testing and assessment.