Channeling innovation in the anti-testing fervor


Dec 4, 2015

You don’t have to look far to find controversy regarding standardized testing.

Federally mandated testing has been one of the major points of discussion in Congress’s efforts this year to reauthorize the Elementary and Secondary Education Act (ESEA). Last spring, the press was full of headlines about families opting out of state testing in protest to current testing policies.

Opponents of standardized testing raise a number of legitimate issues that education leaders and policymakers ought to address. But getting rid of standardized tests would be a major mistake, especially considering the important role these test play in shedding light on achievement gaps between advantaged and disadvantaged students. Instead of retreating from the idea of testing, we need to learn from the flaws of our current assessment system and take advantage of new ideas and technologies. This way, we can ultimately improve how assessments support progress for our students and our schools. Below are a few ideas for how we might do this.

Use adaptive assessment
One of the problems with traditional standardized assessments is that, in order to keep the tests from becoming unrealistically lengthy, most paper-based assessments do not measure the understanding of students who are far ahead of far behind the average level of achievement. Fortunately, as states increasingly move to computer-based assessments, they are also able to take advantage of adaptive assessment technologies. These assessments can adjust the test items as the student takes the test and tailor the assessment items to each students’ current level of understanding. Such adaptive assessments provide a much more accurate measure of the strengths and learning needs of both advanced and struggling students.

Embed assessment into the learning experience
One of the biggest critiques of current tests is that they take time and focus away from learning. But testing does not have to be a distraction from learning. With blended learning, assessments can be integrated into online practice activities and checks for understanding, making them part of the learning and feedback process, rather than an onerous activity that is separate from learning. Additionally, some of the latest innovations in computer-aided assessment are increasingly able to gather deeper insights about students’ learning through interactive simulations. For example, a recent TechCrunch article explains that as students play SimCityEDU, “the system is capturing their actions—such as the sequence of what they do or requests for help—and interprets patterns in data to assess how well the player understands important concepts. This helps teachers better evaluate how a student solves problems, rather than just the final product of their work.”

Measure growth, not just proficiency
Another one of the flaws of current tests is that they often fail to recognize teachers’ and students’ hard work. If a student starts the school year two years behind and then works hard with his teacher to finish the school year only two months behind, this is a major accomplishment. But unfortunately, tests that only report on students’ performance at the end of the year relative to state-determined benchmarks do not recognize the hard work of such students and teachers. To address this flaw, states should instead administer assessments at multiple times throughout the school year in order to measure students’ learning growth over the course of a year. Additionally, assessments administered during the school year would give teachers important feedback as they teach, rather than after the school year is over, in order to help them improve their practices.

Use multiple measures
One of the critiques that resonates with many people—especially those who have gone on to succeed in life despite the fact that they didn’t perform well on standardized tests—is that traditional standardized tests don’t measure all the things that matter in education. Unfortunately, when we rely heavily on traditional assessments for evaluating students and holding school systems accountable, we inevitably neglect or ignore other important aspects of education such as deeper learning, creativity, social skills, and non-cognitive skills. The solution: First, states need to adopt better assessments that can capture more of what matters. For instance, one reason why some testing critics have been in favor of new Common Core-aligned tests is because they are designed to be better measures of deeper learning. Second, states need to use other measures to evaluate students and schools. California is making major strides in this regard.

A new day for assessment
Fortunately, Congress’s draft bill for reauthorizing the ESEA offers great promise for improving current assessment policies and practices. The current version of the bill allows new leeway for states to measure individual student growth, use multiple measures of student learning, move away from end-of-year assessments, use data from multiple points in time to calculate summative scores and growth, and use adaptive assessments that measure the student’s actual achievement level. Additionally, the bill establishes an Innovative Assessment Pilot program that will allow a handful of states to be forerunners in developing innovative approaches to assessment.

Hopefully, as education leaders and state policymakers continue their efforts to improve our education system, they will have the wisdom take advantage of the opportunities afforded in the ESEA reauthorization bill. This means they will need to see testing critiques not as justifications to back away from assessments and transparency, but instead as sources of impetus for progress and innovation in assessment.

This post also appeared on EdScoop

Thomas Arnett

Thomas’ research focuses on the changing roles of teachers in blended-learning environments and other innovative educational models. He also examines how teacher education and professional development are shifting to support the evolving needs of teachers and school systems.

  • I wish I was confident that States will “…have the wisdom take advantage of the opportunities afforded in the ESEA reauthorization bill” and had robust enough staffing in their State Department’s assessment division (assuming they have such a thing) to avail themselves of the very good ideas included in this post… but the practical reality is that SOME states will be basing assessments on bogus scientific knowledge (e.g. excluding climate change and evolution) and MOST states will continue to rely on inexpensive bell-curve tests instead of the adaptive ones recommended in this post. Oh… and some states will continue to use these tests to implement some form of value added measure to identify the “bad teachers” who are causing 50% of teh students to score below average on a standardized test.

  • Harvey Armbrust

    To answer each point made by Mr. Arnett, I will focus briefly on each paragraph.

    The tests [California Assessment of Student Performance and Progress(CAASPP), Smarter Balanced Assessment Consortium(SBAC)] we use are adaptive. Because of the technology, there is another learning curve for every student. Students find computer based tests more challenging than paper pencil.

    There is no way around it. Testing does take away from learning. It is a fact of educational life. Again, because of the computer based tests, students need to learn how to read a test passage on the computer and scroll up and down the find the questions and answers. It can be very complex for a 3rd grader. Struggling students don’t have a chance.

    I fully agree with your point about progress over proficiency. It sounds good but administrators at the district level only see test scores. They don’t know the individual students behind them.

    I agree with the comments about multiple measures, however common core has made the tests more complex. The disadvantaged as you say (children living in dysfunctional homes) are really struggling with these tests.

    As a teacher on the front line, we find ourselves often in the middle of the assessment confusion with little influence.

  • Holly Hart

    I wonder why we would need summative standardized assessments at all if we have quality embedded adaptive assessments that provide data on individual student growth. We could certainly combine that data into school, district and state summaries without compromising individual student privacy. Any type of summative standardized assessment limits the ability to breakdown grade distinctions that currently hold back our brightest and leave behind struggling students.