Can California Offer a New Model for Accountability? Or Are We Still Chasing Test Scores?
A couple of weeks ago, Linda Darling-Hammond and Randi Weingarten offered an interesting take on the path forward regarding Common Core and high stakes tests. They suggest that the real problem is not the standards, but the high stakes tests currently associated with them. Remove these destructive high stakes, they say, and the standards and tests will be transformed into something beneficial.
They offer California as a role model, and say this:
State leaders eliminated all of the old state tests while bringing in new and better Common Core assessments. These will be used, along with other measures, to inform instruction and professional development, not to punish children, schools, or teachers. State funds have meanwhile been invested in modernizing schools, improving instruction, adding technology, and expanding successful career academies.
When these measures identify struggling schools, intervention will come in the form of help from experts who are part of the California Collaborative -- a new entity that will provide teams of educators to diagnose what's working and what's not -- and will support ongoing improvement. Thus, the Common Core standards in California are an engine to drive better educational practice, not a hammer to threaten children, educators, and schools with failure.
When Darling-Hammond and Weingarten speak of the high stakes they oppose, they are focusing on the practice of basing school closures and educator evaluations on student test scores.
Here are my concerns about this approach.
Common Core standards are a bit like the growth charts a pediatrician might use when examining a growing child. They offer a detailed description of what every child should be able to know and do, from kindergarten through high school. If we accept the Common Core growth chart, this has huge consequences -- even in the absence of VAM-driven teacher evaluations.
The Common Core creates a template for learning, because these standards were written with measurement in mind. There is a reason test-writers were prominent members of the team that wrote the standards. And nobody is discussing eliminating the tests - just promising that the Smarter Balanced tests we are getting are somehow "better."
The results that we get on these tests are going to be used as a diagnosis, telling us what is right and wrong about instruction in our schools. On the good side, we are being promised that those who do poorly will not face school closure as a consequence. But there will nonetheless be a public process by which schools are identified, and intervention applied, with the goal of improving scores on Common Core-aligned tests.
Any doctor will tell you that growth charts provide only a rough guideline for human development. Depending on all sorts of factors, any given child may be above or below what is "normal" for their age. And academic development of children is even more variable, and dependent on all sorts of factors.
When we accept Common Core as our template for what students should be able to know and do, we are accepting a whole set of beliefs that drive our behavior as educators.
In the California scenario, we are supposed to accept the scores from the Smarter Balanced tests as an accurate diagnosis for what ails both individual students and entire schools. That assumes this test is valid, and the standards are a worthy and appropriate template. Superintendent Torlakson has even spoken of using the tests in place of the state high school exit exam, which would make it very high stakes indeed.
Weingarten and Darling-Hammond are correct in pointing out the terrible consequences of basing school closures and teacher evaluations on test scores. However, I remain concerned about the potential harm the Common Core standards can do to children and our education system.
Take a look at this video that is prominently placed on the Common Core web site.
Here is what the script says:
Like it or not, life is full of measuring sticks: How smart we are, how fast we are, how we can, you know, compete. But up until now, it's been pretty hard to tell how well kids are competing in school, and how well they're going to do when they get out of school. We like to think that our education system does that. But when it comes to learning what they really need to be successful after graduation, is a girl in your neighborhood being taught as much as her friend over in the next one? Is a graduating senior in, say, St. Louis, as prepared to get a job as a graduate in Shanghai? Well, it turns out the answer to both of these questions is "no." Because for years, states have been setting different standards for what students should know and be able to do at each grade level. That's making it too hard to know if our kids are really doing well enough overall and if they can really compete for a job some day.
The video concludes:
The world's getting more and more competitive every day. But now when our kids get to the top of their staircase, they can have way more options of where their life goes from there. Clear goals, confident, well-prepared students, that's the Common Core state standard.
So let's unpack the assumptions built into Common Core. First, "like it or not," we are told, our world is determined to measure everything. Bizarrely, we even have a picture of someone who looks like Albert Einstein measuring the circumference of his skull, as if this has any value. And these measurements are the basis for competition - and our students are in a race against one another, and against that kid in Shanghai, who may be better prepared for a job than our kids.
The way to make our students "confident" and "well-prepared" for this race is to set up their learning as a series of steps they must climb, and every student at a given grade must mount these steps in order, and at the same age.
This is a powerful framework for learning, and I think it is destructive. While Randi Weingarten has stated that she believes the K-2 Common Core standards are not appropriate, the rest of the sequence is taken as solid. We also must accept that standards and associated tests are accurate and useful indicators of what students can do. And we are asked to believe that economic opportunities are distributed on the basis of test performance. These are dangerous assumptions.
We have yet to see the results of the Smarter Balanced tests that Weingarten and Darling-Hammond suggest are better. Placing our trust in these instruments has huge systemic consequences. Even if we do not close schools and fire teachers based on these scores, they will be used to provide hugely consequential feedback to schools, teachers and students.
Here is my fear: I believe these tests will follow the same pattern as other Common Core-aligned tests, and yield results that show our English Learners and students living in poverty are in terrible shape. Why do I think that? First, because these tests have a great deal of language in them. Even the math problems require students to explain in words how they are solving problems. So we are likely to see schools with large numbers of ELs get terrible scores. So far every state that has given tests aligned with the Common Core has seen huge drops in proficiency levels.
The schools identified as low performing are likely to be the same ones identified that way under NCLB. They will get "intervention," meaning teams of expensive experts who will come to tell the teachers at these schools what they must do to raise their test scores. We will remain stuck in a data-driven paradigm, where test scores are treated as an accurate indicator not only of what ails us, but also to guide the steps we must take to improve our health.
Let's be clear. There is not one way to measure learning. As Rog Lucido pointed out here a while ago, student learning can only be described, not truly measured.
Every test we give is an attempt to capture a complex process in a small, definable cage. Every question intentionally excludes a host of contextual information that cannot be translated into simple data points. When expert teachers conduct formative assessments in their classrooms, they are capable of assimilating far more of this information, and thus, when their wisdom is honored by our schools, their students are likely to be better off. This is the system used in the best schools. Take a look at Sidwell Friends school, or Lakeside Academy in Seattle. These elite schools do not allow test score "data" to drive instruction. They provide teachers with the small class sizes that allow them to develop understanding of their students, and relationships based on what those students need in order to grow. The school system in Finland makes similar assumptions.
But the assumption in a test-score data-driven system is that teachers are not capable of making these assessments, and instead must be guided by standards and tests. Teacher and scholar Mercedes Schneider stated in recent correspondence,
From its inception CCSS was meant to drive the entire K-12 ed enterprise. CCSS was meant to drive curriculum, was meant to drive professional development, was meant to be closely tied to assessment, and was meant to foster the need for unprecedented data collection.
Even if you remove the harshest consequences, these elements will remain in place.
There is no special advantage conferred by everyone taking the same tests. The entire state of California has used a common set of standards and tests for more than a decade, and we continue to have great disparities in student performance from one district to another. These disparities are driven by the circumstances in which our students live, and the resources their schools have to work with. As Tom Loveless has pointed out, there is no evidence to support this idea that standards and tests such as these will yield improved learning.
In California, as Darling-Hammond and Weingarten point out, there is a commendable effort to shift resources, so the neediest students get the support they need. This is part of the answer. However, I feel compelled to point out that a similar promise accompanied No Child Left Behind. There was a deal struck - accept test-based accountability as a condition for additional resources. Here we have a similar pairing of resources with accountability, even if the sharpest teeth have been removed.
The promise of the Common Core is that we create confident students and help the under-privileged by measuring them on a set of difficult tests, but these tests will show that those who have always been behind are further behind than ever. I just don't see how this builds confidence. I think that in spite of the best efforts of teachers and leaders in our state, many of our students will do very poorly on these tests. And high-poverty schools will do worse than ever. We will then be obliged to use these scores as an accurate diagnosis of our problems, and in effect this will justify and reinforce inequities, rather than challenge them.
Weingarten and Darling-Hammond are correct that there has been less resistance to Common Core in California than elsewhere, largely because the rollout has been done more slowly, and thus far we have not even seen any test scores. State leaders acted wisely by canceling the tests aligned to the prior standards, so at least students only took one set of tests.
But we who are opposed to "high stakes tests" are opposed not only to the most vicious uses of those tests, to close schools and fire teachers. We are opposed to the practice of making standardized tests central to the educational process in our schools. And a supposedly "better" test does not fix that obsession. Teams of expert advisors instead of school closures does not fix that problem either.
We have another paradigm in mind. One that truly honors the expertise of classroom teachers not only of implementers of curriculum, but also as assessors of learning capable of diagnosing what children need, and finding ways to deliver it. A doctor might use a growth chart to provide some basic guidelines for how a child ought to be growing. But he does not use that chart to assess the child's health and make decisions about how that child ought to be treated. Educators do not always know best, but when they work with parents and members of their community to build their expertise, they have a far better chance of meeting their students' needs than clumsy systems of standards, tests and outside advisors.
I greatly admire Linda Darling-Hammond's contributions to the teaching profession. Her 1997 book, "The Right to Learn," inspired me to pursue National Board certification. In that book, she wrote some profound words that I think apply here.
It has taken nearly a century to discover that, as a form of organization, bureaucracy lacks the tools to manage complex work, handle the unpredictable, or meet distinctive client needs. By its very nature, bureaucratic management is incapable of providing appropriate education for students who do not fit the mold upon which prescriptions for practice are based. As inputs, processes, and measures of outcomes are increasingly standardized, the cracks through which students can fall grow larger rather than smaller because the likelihood that each accumulated prescription is suitable for a given child grows smaller with each successive limitation upon a teachers' ability to adapt instruction to students' needs.
Bureaucratic solutions to problems of practice will always fail because effective teaching is not routine, students are not passive, and questions of practice are not simple, predictable or standardized. Consequently, instructional decisions cannot be formulated on high then packaged and handed down to teachers. Nor can instructional problems be solved by inspectors who make occasional forays into the classroom to monitor performance and dispense advice without an intimate knowledge of the classroom context, the subject matter being taught, the goals of instruction, and the development of individual children.
If ever there was a bureaucratic solution, it is the Common Core, and the associated tests and treatment plans. Weingarten and Darling-Hammond have asked us to suspend our fears, and put our faith in a more enlightened bureaucracy in California. I do not think this is wise. In fact, I think it leaves us disarmed and our students vulnerable to what is likely to be a very destructive system.
I think the difference between my viewpoint and that of Weingarten and Darling-Hammond may stem in part from a difference in our positions. I see things from the point of view of a grassroots activist, and I see a national movement against the corporate standardization and privatization of education that is growing every month. There is evidence of this movement in campaigns actively supported by the AFT, such as Reclaiming the Promise, as well as initiatives such as United OptOut, the Network for Public Education, and the BadAss Teachers. As this movement gains strength, it has the capacity to create a new political reality in the future, which could result in very different policies. Based on our current strength, however, the bargain Weingarten and Darling-Hammond describe is probably the best that can be achieved. Because I believe we can change this situation through the actions we take, I think we ought to set our sights higher, and continue to push forcefully for better solutions.
Note: I have previously written about a wide range of concerns regarding Common Core, many of which go beyond the use of test scores to close schools and fire teachers.
- Classrooms of the Future: Student Centered or Device Centered?
- The Common Core Equity Swindle
- Common Core, China, and the Myth of Meritocracy
- Who Decides What is Taught in Our Schools?
- Common Core Standards: Ten Colossal Errors
- Process Matters in a Democracy: Common Core Fails the Test
What do you think? Is California capable of creating an enlightened accountability system based on the Common Core? Or is this setting us up for another pointless pursuit of higher test scores?
Continue the dialogue with Anthony on Twitter.