Sound advice, and it didn’t cost anything.


Dear President Obama:

I am writing to you as a fellow magna cum laude Harvard Law grad (’73) and fellow Democrat.

Please reconsider two of your Administration’s Race-to-the-Top education policies – 1) high-stakes testing, and 2) charter schools.  Under superficial analysis, these policies appear sensible; under in-depth analysis, they are destructive.

In developing school reform policies, the first step is to identify the problems.  In the US, most suburban schools are doing OK.  The problems exist in the low-SES schools, particularly in the inner-cities.  Veteran teachers and principals in the low-SES schools – writing in first-person books, on education blogs, and newspaper websites – routinely cite two main problems – students reading far below grade level and classroom misconduct – and two additional problems – chronic absenteeism and chronic tardiness.  In the low-SES schools, these problems are endemic; in the suburban schools, these problems exist, but they are occasional rather than endemic.

Rational school reform should, therefore, focus on these problems that are endemic in the poorly-performing low-SES schools but exist only occasionally in the well-performing suburban schools.  Such rational school reform would directly address the problems that plague the low-SES schools.  We should be testing and implementing reforms in the low-SES schools regarding reading skills, student behavior, and attendance.  Unfortunately, Race-to-the-Top largely ignores these problems.

High-stakes testing (and the resulting teacher discharge) address a different problem – that is, the problem posed by bad teachers.  However, bad teachers are not responsible for the poor performance of the low-SES schools.  The suburban schools are doing OK and there is no reason to believe that teacher quality in the suburban schools is uniformly stronger than in the low-SES schools (particularly given the flood of hyper-talented Teach for America grads into the low-SES schools during the past 15 years).  If bad teachers was a major problem, we’d see the low-performing schools distributed roughly evenly between the suburbs and the low-SES areas.

Moreover, assuming arguendo that identifying/discharging bad teachers would significantly improve low-SES schools, high-stakes testing is an awful way to identify/discharge bad teachers.  First, high-stakes testing is unreliable – it yields too many false positives and too many false negatives.  There are many non-teacher-controlled variables that impact a teacher’s student test scores under even the most sophisticated value-added model – i.e., the number of “difficult” students who disrupt the class and require extra attention, the class size and the teacher’s total pupil load, the number of different preparation the teacher has, whether the teacher has taught the course/grade before, whether the teacher is teaching within his/her core expertise, the amount of classroom support (aides), and the amount of central office support (regarding discipline, etc.).  Second, high-stakes testing has major adverse side effects.  It encourages teaching to the test, encourages narrowing the curriculum, encourages cheating, discourages teacher-teacher cooperation, and, most importantly, discourages teachers from teaching “difficult”/high-risk students.

And, there are more productive ways to identify/discharge bad teachers.  For example, Montgomery County, MD (a large mostly suburban school system near Washington, DC) has successfully used a peer-review approach (called “PAR”) for over 10 years.  PAR has resulted in the discharge or resignation-in-lieu-of-review of over 500 teachers; the teachers union supports PAR; the teachers view the system as fair; few of the discharges have been challenged; and there is no high-stakes testing with its adverse side effects.

Charter schools, like high-stakes testing/teacher discharge, fail to address the problems that plague the low-SES schools.  Instead of solving the problems, charters simply remove some of the children – the children of the parents who are functional enough to investigate charters, complete the application process, and provide the daily transportation usually required for charter students – from low-SES neighborhood schools.  By enrolling only the children of the more functional parents, the charters largely avoid – rather than solve – the low-SES school problems of students reading far below grade level, classroom misconduct, absenteeism, and tardiness.  And, by siphoning off the children of the more functional parents, the charters increase the concentration in the neighborhood schools of children of the less functional parents – thereby exacerbating the problems that plague the low-SES schools.

The charter approach – protecting the children of functional parents by sacrificing the children of the dysfunctional parents – will ultimately impose huge social costs on society as the children of the dysfunctional parents become drop-outs, teenage mothers, barely-literate unskilled workers, welfare recipients, and criminals.  A far better approach would be to abandon the charters and instead implement school reforms directly targeting the problems that plague the low-SES schools.

Please stop the high-stakes-testing/teacher-discharge and charter reforms that are damaging, rather than improving, our low-SES schools.  Please encourage the low-SES schools to implement reforms directly addressing the problems that plague the low-SES schools.

Sincerely,