Cross-posted from Education Week
The signature difference between the Elementary and Secondary Education Act of 1965 and its direct descendant, the No Child Left Behind Act, is the way these two pieces of legislation treated the issue of accountability.
The designers of both pieces of legislation focused on social justice for poor and minority children, but the frames were very different. The reformers of 1965 saw the problem as the students and the reformers of 2001 saw the problem as their teachers. What a difference four decades makes!
Back in 1965, educators and legislators viewed the challenge as a need to compensate for the deficits that poor children and minority children bring to school. The answer was compensatory education. Not to compensate for poor teaching but to compensate for the deficits of the students. They did not assume that teachers knew everything they needed to know to address the problems of these children and so they included in the legislation some sound measures to build the capacity of teachers to meet their students’ needs.
But, 36 years later, the cost of elementary and secondary education had more than doubled in inflation-corrected dollars, the performance of America’s fourth graders on the NAEP reading test had barely budged and one nation after another had overtaken the United States in international comparisons of student performance. It looked as though the professional educators had been perfectly willing to take the money but not to act as if they felt any responsibility to produce results with it. Legislators on both sides of the aisle were angry. No Child Left Behind is the result of that anger. It was written by people who had come to see professional educators as the problem, and who were determined to hold them accountable for their poor performance as the price for getting more federal funds.
NCLB tackled this problem in three ways. First, states were required to set the same standards for poor and minority students as for majority students and were required to report their performance against those standards publicly. Second, the states were required to set rates of progress toward those standards that would affect every school; for the first time in history, federal legislation contained provisions that could lead to firing teachers if the achievement of poor and minority students did not progress on the schedule to which the states committed. Third, parents were guaranteed a right to move their children from low-performing schools to higher-performing schools, on the theory that competition from stronger schools would improve the performance of lower-performing schools.
But, as we all now know, most states low-balled their standards, to reduce the likelihood that the state would have to take unpopular measures to punish low-performing schools. Second, the states almost invariably chose the least-aggressive punishments for the schools that failed to perform. Third, parents failed to take advantage of their right to move their children to higher-performing schools and on those occasions when they tried to do so, they often failed, because the few higher-performing schools available were full. And fourth, good teachers who had choices to make started to avoid schools serving low-income and minority students, because that is where their own jobs were most at risk.
Since No Child Left Behind was passed, reformers have not changed their view that teachers are the problem, but they have shifted their focus from the school to the individual teacher. The aim now is to sort out the effective from the ineffective teachers and get rid of the latter. The chosen instrument for doing this is analysis of student test scores, tying the student scores to their teachers. But it turns out that, when analysts look at the data for individual teachers, it is very unstable, so that a teacher who looks like a real winner one year looks only so-so the next, when judged by the performance of her students on NCLB tests. In any case, to make such a system work, the schools would have to have comparable standardized student testing data available for every student at every grade level for every subject taught by every teacher. That would entail a vast expansion of a testing regime that most people already believe to be too burdensome.
Now, at the tenth anniversary of NCLB, it seems that we have tried every available approach to accountability and nothing has worked. But if we look to the experience of the top-performing countries, we can see that there are viable alternatives, alternatives that could easily be adapted for use in the United States. In future blog posts, I will go into more detail about what other countries are doing on this front; in the short space that follows, I will give you a quick tour of some of the possibilities.
One of the great successes of NCLB was its requirement that the states publish data on school performance in a uniform format and that that data be disaggregated so that schools could not hide the poor performance of poor and minority students behind the superior performance of majority students. If you want to see a really good example of what can be accomplished along these lines, take a look at http://www.myschool.edu.au, the new system of online school profiles in Australia. If you would like to see what a state report on its own performance in education could look like, in this case from the Netherlands, check out “Key Figures 2006-2010,” Netherlands Ministry of Education, Culture and Science.
But gathering performance data and sharing it publicly is only part of the accountability picture, and it is not the tough part. The world’s top-performing countries hold both their schools and their teachers accountable, but not the way the United States has been doing it.
Most of the top-performing countries have some form of standardized tests, but usually at not more than three grade levels through the whole elementary and secondary education grade sequence. These tests are typically primarily used to help determine a student’s path from one stage of the system to the next. Sometimes these tests are used for public accountability purposes as well. Some countries use sampling procedures to administer tests that are not taken by all students and are used only to report on the performance of the system, like NAEP. However, none of these countries use standardized tests of student performance as the sole measure to determine whether a school is sanctioned for poor performance and none use them as the primary means of evaluating individual teacher performance, and only one that I know of uses them as one element in the evaluation of teacher performance.
Publicizing student scores on standardized tests by school appears to be most effective as an accountability measure in those countries in which test scores are an important determinant of student’s admission to secondary schools or to university once a student has completed secondary school. In those countries, elementary and middle schools are ranked by success in getting their students into the more desirable secondary schools and the secondary schools are ranked on their ability to get their graduates into the more desirable higher education institutions – both academic and vocational. It is this link between school performance as measured by test performance and the opportunities provided by different levels of school performance that makes the system work. For descriptions of such systems, see the chapters on Japan and Singapore in the OECD publication, Strong Performers and Successful Reformers in Education: Lessons from PISA for the United States.
In some countries, the primary instrument of school accountability is the creation of an inspectorate, an organization of experts usually employed by the state to visit schools on a regular schedule and report publicly on what they find, against a set of predetermined criteria. In some countries, all schools are visited on a regular schedule in any case. In most countries with such a system, low-performing schools are visited more often until their performance improves sufficiently. In some countries, the inspectors are not allowed to prescribe remedies for a school’s perceived failings, for fear that doing so will compromise their objectivity. In others they are encouraged to do so. In some cases, the only tool the inspectors have available for improving a bad school is disclosure of their findings. In others, they have the power to close the school altogether. For a superb report on two countries with inspectorates, see a paper by Helen Ladd: Education Inspectorate Systems in New Zealand and the Netherlands, published in Education and Finance Policy, 5, 378-392.
Publicizing a school’s shortcomings doesn’t do much good if the faculty of that school is not capable of fixing them. In Strong Performers and Successful Reformers in Education: Lessons from PISA for the United States, Chair Professor of Education at the University of Hong Kong Kai-ming Cheng offers an excellent description of the education system in Shanghai. In it you will find a portrayal of the strategies used by the Shanghai Education Commission to improve the performance of their low-performing schools.
In later blog posts, we will look more closely at the way other countries collect and publish data on their performance and their schools’ performance, the way different countries use the idea of inspectorates as a keystone of their accountability systems, the incentives that schools in these countries have to respond to the public reports on their performance, and the strategies that other countries use to improve the performance of low-performing schools. But this should get you started.