This is a new piece I wrote as the foreword to a book by J. M. Beach — Can We Measure What Matters Most? Why Educational Accountability Metrics Lower Student Learning and Demoralize Teachers — which will be published by Rowman and Littlefield.
For me, this was a chance to provide a brief summary of my thoughts about the problems that accountability metrics pose for schooling. See what you think.
In this book, J.M. Beach provides a devastatingly effective analysis of the accountability metrics that have wrought so much havoc in the American system of schooling. The accountability movement started at the state level in the U.S. in the 1970s under the label of curriculum standards, ramped up as a national issue in 1983 with the publication of the report A Nation at Risk, and made its way into federal law in 2002 with the passage of the No Child Left Behind act. This sweeping reform process set out a radical agenda for schools, which dictated that the prime criterion for success for students, teachers, and school systems was scores on tests that measured student understanding of core curriculum subjects.
The accountability pandemic is now a global phenomenon, pushed by governments around the world and enforced by the tests written by OECD’s Programme for International Student Assessment (PISA). These scores are averaged out for each country and the countries are then ranked by score, creating an international measure of educational fame and shame. Heaven help the government in power when the latest scores show the country has fallen in the rankings. PISA rules.
As Beach shows, the problems with this system are legion. But it’s worth noting that the urge for accountability is not unreasonable. Education should be accountable. It’s a public institution that needs to be effective at meeting the goals society sets for it, and such determinations can’t just be left to the preferences of teachers or parents or students or administrators. In addition, it’s not ok that many students don’t succeed in school and that their social origins are key determinants of their success or failure. Schooling whose outcomes simply reproduce its inputs is not good schooling. These equity concerns are visible in the names of the two key US laws governing accountability – No Child Left Behind and its 2013 successor, Every Student Succeeds Act.
The problems with accountability lie in the way it is implemented. The accountability movement in the US and in the world of school reform has relied on a method that defines school success through a small number of metrics – scores in tests that measure comprehension of the formal curriculum.
One problem is that even if these measures capture core aims of schooling – which, as I’ll suggest, they don’t – they would still become distorted by the effort to achieve them. These outcomes – test scores – become the primary target for everyone’s educational efforts (students, teachers, administrators, policymakers). This is a glaring case of what has come to be known as Goodheart’s law, named for the British economist who developed it: “Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes.”
In regular language this means that when you have a valid measure of some social phenomenon, it loses its validity when it because a target of policy, because it motivates actors to attain this metric by any means possible. Instead of teaching the curriculum, you teach to the test, so you end up raising scores instead of improving learning.
A more basic problem is that this accountability system, when applied to education, radically narrows the aims for schooling to a few outcomes that are readily measurable but not very important. It turns our attention to one-shot scores on tests of a tiny number of academic subjects – language, math, and science – at the expense of the larger goals that explain why we have invested so much time and money into erecting school systems. We want schools to serve a political goal, creating competent citizens who can function in a democracy. We want them to serve an economic goal, creating productive workers who can fuel economic growth. And we want them to serve a social goal, allowing individuals to prove their merit in the pursuit of social opportunity. It’s not obvious how test scores capture or promote any of these goals.
Accountability metrics misread the nature of the educational system and in the process end up creating dysfunctions that are no less damaging for being inadvertent. Start with the fact that schooling is a complex social process that requires the cooperation of a huge number of actors in a large number of institutional settings, including 50 million students and 3 million teachers in 100,000 schools.
It’s a system that depends on the motivated compliance of the key actors, teachers and students. Teachers can’t make students learn. They need to find way to motivate students to learn what they’re teaching, and this isn’t easy since students didn’t arrive in the classroom begging to be taught. Instead, they’re compelled to attend by truancy laws and by the credentialing requirements for entry into the workforce. These are not problems that face other professionals, such as lawyers and doctors, who work with clients and patients who ask for their help. Students don’t contract with teachers to teach them reading and algebra. Teaching is forced on them, and it only succeeds if they choose to play along and learn the curriculum.
As a result, teaching and learning in schools is a messy and inefficient process that doesn’t necessarily work as planned. Grounded on the indirect process of motivating the student, it doesn’t operate on the kind of mechanical rules that apply to factory production. There is no one right way to teach effectively. Instead, teaching involves an ongoing relationship with individual students. It requires teachers to figure out how to draw on their own personal and professional resources in order to get through to students who come in with different levels of skill, knowledge, and motivation.
The whole massive and complex educational system comes down to the student-teacher relationship in 3 million classrooms across the country. A top-down mandate to raise test scores only makes it harder on teachers to do their jobs. Instead of intense pressure from above, they need the support and resources and autonomy they need in order to make the pedagogical relationship work, classroom by classroom and student by student. Beach recognizes this core issue and makes the case for why accountability metrics are the problem, not the solution.
Palo Alto, CA