OnCUE

Do We Still Need Standardized Tests? A Bold New Approach

Test zen renci— Special Guest Editorial to the CUE Blog —

Over the past several years, there has been an ever-increasing uproar about the over-testing of our children, especially the big standardized tests imposed by the No Child Left Behind Law. These tests stressed out teachers and students while undermining our children’s opportunities to learn for several weeks to frantic periods of test preparation. On top of those issues, there have been seemingly endless attempts to perfect the standardized test itself, at the cost of hundreds of millions of dollars and dividing educators and communities. And then there are the vast costs for purchasing, administering, and scoring the tests, all of which falls on hard-pressed districts. In other words, the government created a monster.

The problem was sufficiently troubling that educators hoped Congress would abandon the requirement for such tests when it rewrote the No Child Left Behind law that mandated them in the first place. It did not. Civil rights advocates, conservatives, and assorted other groups formed an alliance during the drafting of the new bill that worked to preserve the requirement that children be tested in third through eighth grades and once in high school. However, the new law, the Every Student Succeeds Act, did recognize that there might be a better way to generate data for accountability purposes and offered states the chance to experiment with new ideas.

There are indeed better ideas out there.

One of the most positive developments in education technology in recent years has been the ever-growing adoption of digital instructional programs. That growth is going to continue because they help schools deliver differentiated, personalized instruction, which educators believe is one significant key to improving student achievement.

We believe that when students use such programs, they are creating more than enough data to satisfy whatever accountability purposes the state may have.

Although most students continue to attend classes when using these programs, their primary source of content and skills instruction comes from working on their computers. Before they can move from one unit to the next, they are required to take a small quiz (sometimes called a gateway exam) to see if they have mastered the unit’s material. The primary use of those results is to tell a teacher in real time who is struggling or who is on track, so that the teacher may provide help or guidance as necessary.

Under our proposal, that same data is used for a second purpose: The ability to track a student’s acquisition of knowledge through gateway exams allows teachers to measure a student’s progress (e.g., she showed mastery of the material on sixteen gateway exams this year), and based on the last gateway exam a student passed before a date (such as the end of third grade), the data identifies the student’s current level of mastery of content and skills. The use of all that data gives an even better picture of student achievement and, more importantly to many teachers, the student’s growth, than the current standardized tests, and, since it is embedded in daily learning, takes no time away from learning or other class activities.

One major question would be whether the gateway tests are valid measures of student learning. We suggest that the state guarantee this by drafting rigorous standards (consistent with the Common Core, at least here in California) reflected on two smooth curves, one for literacy and one for numeracy, starting in kindergarten and ending in 12th grade. We characterize the standards as “smooth curves” in recognition that not every ten year old, for example, is going to be achieving at a fourth grade math level. Rather, they are moving at their own pace up the curve from kindergarten to senior year.

To keep anyone, including the state, from backsliding on the rigor of the standards, the 12th grade standard would be set for what it would take for a student to earn a passing “3” on an Advanced Placement exam, or equivalent. (Just to be clear, in California students generally are required to achieve no better than mastery on 10th grade material in order to graduate, and nothing here changes the fact that students have twelve years to master ten years of work.) The specifics of each curve would be the standards required on each of the gateway exams.

Not only would that ensure rigor, but also it would integrate state standards and district grading. That means we would have one set of data that satisfies accountability for all systems – parents, schools, the state, and the federal government.

Cute little schoolchildren using desktop PC at desk in computer labWe recognize that our idea works best when every child has full-time access to a computer. We are not there yet in California, but the federal waiver provision allows a state to test out an idea with only some of its districts participating and scale up over time.

There is strong support among leading educators that the proposal will work, that it will generate all the data we need, and that it will save our kids from test stress and give them back the weeks we currently steal from them for test prep.

It is time to end those big standardized tests. We really can do so much better.


Dr. Barbara Nemko has served as Napa County Superintendent of Schools since 1997. Before coming to Napa she was at the University of California, Davis for 12 years as project director and principal investigator for 11 California Department of Education research projects. During her early years in education she taught elementary and junior high school in New York City. Barbara has been a frequent speaker at many local, state and national conferences. She is passionate about helping teachers and administrators to implement digital resources to help close the achievement gap, and has spearheaded a very successful program that put iPads in the hands of preschool English learners to provide reading readiness skills. Barbara currently serves as the chair of CUE’s Nominating Committee.

Harold Kwalwasser is the former General Counsel of the Los Angeles Unified School District.  In the winter of 2011-12, Rowman and Littlefield published his book “Renewal, Remaking America’s Schools for the 21st Century,” which describes and analyzes how 40 school districts, charters, private and parochial schools are making great education happen.

About author View all posts

Barbara Nemko, PhD and Harold Kwalwasser

5 CommentsLeave a comment

  • Except that instruction through a computer is a terrible learning experience for most kids. Computers can’t respond like humans can, the activities are almost always isolated in a time when students need interactions since they are learning how to learn from each other and adaptive systems and differentiated instruction have very scarce (legit) evidence that they make any kind of difference in learning despite conceptually seeming like a good idea.

    We might just have to focus on heping find and train great teachers.

  • This is not fair to fine arts and physical education teachers- who cannot teach physical and motor skills through a computer screen. We have standards and benchmarks to reach too! Also, what happened to the visual-aural-kinesthetic approach? How are we going to help those who cannot handle sitting down in front of a computer screen?

    Also, in rural areas where students have computers, the cost to keep internet that isn’t dial up is outrageous. If you go over your gigabyte usage you could owe hundreds of dollars. That’s negative for those families.

  • Taking the CST once a year wasn’t the end of the world, as many have claimed. What was overkill was the additional testing districts added to give students “practice” . The fact careers were put at stake as the additional tests were used as “hammers” on teachers and site admin caused these professionals to focus totally on the “power standards” that gave you higher returns. This led to gaps in the curriculum that widened significantly as students progressed through grades. I believe the SBAC tests provide a better idea of how students are learning but that is handicapped by lack of experience on the machines. Just eliminating testing has caused a huge divergence of pedagogy and materials, which will now take time to overcome. We need to have measurements that bring accountability and improvement, not punition.

  • What kinds of data do we need? We really should be working to get away from the idea that data only means numeric information.

  • We don’t “still” need standardized tests because we never did need standardized tests. All they tell us is that poor kids are poor. There are cheaper ways of figuring that out.

    Moving kids along the same track at their own pace is not real differentiation, and putting content delivery into the hands of a distant developer is not teaching. Passing a series of quizzes is not education. Pushing the right buttons is not learning.

    This is, in short, a terrible idea unless you are hoping to get a giant data mining contract. It can only hurt the students.