A core belief of education reformers is that, after a child's parents, teacher quality is the most important factor for school success.
Since taking control from the Board of Education in 2003, Mayor Michael R. Bloomberg has fought mightily to let principals pick the teachers they want, believing that he could then hold them accountable for the results: student test scores.
The problem is that New York City students’ performance on the National Assessment of Educational Progress has not changed much since 2003. More than our state exams, NAEP is a credible check of student learning.
Last month the former deputy schools chancellor, Eric Nadelstern, wrote in an essay that improved graduation rates at recently opened small high schools “vindicate an approach that has proven effective.”
But graduation rates paint a partial picture. Finishing high school is not always the same as being ready for college. In 2008, Bill Gates, whose foundation backed small schools in New York and across the nation with $2 billion, concluded that while they raised graduation rates, their graduates were no likelier than any city student to be prepared to go on to college.
Expanding on this point, last summer the state education department documented that just 21 percent of New York City high school graduates are ready to do college-level work. This is several years after the city embarked on its small school movement. Last fall the city itself acknowledged the yawning gap between high school graduation rates and true college readiness.
But it isn’t just that our students' progress is limited. Education spending has increased from $12.7 to $23.9 billion since Mayor Bloomberg took over the city schools. Per-pupil spending has almost doubled. Marginal improvement at twice the price hardly seems the kind of outcome taxpayers expected from a successful businessman.
Testing is critical to understanding what is really happening at our schools. Our state exams yield scant insight on student performance, little accountability about schools, and in my opinion have generally weakened instruction. Like most parents, I understand intuitively that teacher quality matters. But by the time the mayor and the United Federation of Teachers hash out "how much," my first grader will be 21.
The Board of Regents has acknowledged some of the problems with their tests. But how their current “fine tuning” plans, including lengthening tests to three or four hours, will yield richer insight is unclear to me.
Implicit in the federal requirement that every student be tested annually was the belief that parents and policy makers want the same information. But the test that tells a mayor if education spending is effective probably isn’t the same test that tells a parent how well his or her child is doing.
A well-designed audit could judge the effectiveness of a school system. Multiple and focused exams that include written and oral assessments, conducted several times each year for statistically valid samples of students, would yield more insight at a lower cost. Administrators couldn’t game the system, as the students and subjects would be selected randomly.
Over the course of a year the tests could cover many kids, teachers and topics, without unduly burdening any individual school or student. Rotating the testing contract among various universities would ensure no one got too cozy with any one administration, and keep the evaluators honest.
Another benefit of this approach is that schools would have to teach to all the standards, instead of just the ones on the test.
Who knows, the Department of Education might even be obliged to develop a curriculum so students learn something deeper than "finding the main idea" or "determining the author’s purpose."
Either way, an audit program to replace our minimally informative state exams could inform taxpayers whether their investment in education is yielding results, or just press releases.