School Reform: Rise of the Machines
With his appointment of Chris Cerf as Commissioner of Education, Chris Christie is rebuilding New Jersey public education using sweeping, data-driven methods that have been tested (and sometimes bitterly contested) in New York City and Washington, DC.
Sign up for the Smarter Faster newsletter
A weekly newsletter featuring the biggest ideas from the smartest people
What’s the Big Idea?
What does the future of education look like? Picture 100 students sitting at terminals. Five teachers circulate, monitoring student progress. Onscreen, a virtual tutor with human-like charisma assesses what little Billy knows about, say, the Pythagorean Theorem. It detects a gap. It accesses a database containing Billy’s entire educational history – test scores, skills, deficiencies, everything. Because Billy learns best through games, a geometry game appears onscreen. It’s designed incrementally so that Billy always wins, and always learns the material.
This utopian or dystopian vision is still a long way off, however. In the meantime, reform-minded school systems are tinkering with the great-grandparents of these machines – databases for collecting and crunching information about the educational progress of every child, in every classroom, everywhere. Armed with more precise data, states, cities, and schools will be better able to determine who’s learning what, who isn’t, and what to do about it.
The stakes are undeniably high. Failure in school is a statistically significant predictor of problems in later life, including long-term unemployment, drug abuse, and incarceration. High school graduation rates in US public schools are dismally low and skewed steeply by household income. In other words, the public school system is a factory for producing social inequality.
This utopian or dystopian vision is still a long way off, however. In the meantime, reform-minded school systems are tinkering with the great-grandparents of these machines – databases for collecting and crunching information about the educational progress of every child, in every classroom, everywhere. Armed with more precise data, states, cities, and schools will be better able to determine who’s learning what, who isn’t, and what to do about it.
The stakes are undeniably high. Failure in school is a statistically significant predictor of problems in later life, including long-term unemployment, drug abuse, and incarceration. High school graduation rates in US public schools are dismally low and skewed steeply by household income. In other words, the public school system is a factory for producing social inequality.
New Jersey School Reform: Not About Making Everybody Happy
Leading reform efforts in the State of New Jersey is Chris Cerf, recently appointed Acting Commissioner of Education by Governor Chris Christie. A former New York City Deputy Chancellor under zealous school reformer Mayor Michael Bloomberg, Cerf is pursuing system-wide change on a template that mirrors Michelle Rhee’s recent efforts in Washington, DC, and City Hall’s in New York. As in those cities, changes in the New Jersey Department of Education will be swift, acute, and sometimes unpopular. Popularity is not the point, Cerf argues. The point is “maximizing the number of children who get a quality public school education.”
While New Jersey has one of the highest overall high school graduation rates in the country, it is 47th out of 50 states in terms of socioeconomic educational disparity.
Averaged Freshman Graduation Rates* in New Jersey, 2007-2008
Pushback from Teachers’ Unions
Teacher quality and student data tracking are the twin fronts in this war, and it is a war. Data-driven reformers have identified teacher quality as the single most important predictor of student success in school. They are going after teacher-tenure laws that make it difficult to fire veteran teachers, regardless of ability, and pushing for higher teacher salaries and merit-based pay –raises tied to student performance. The focus on data collection means more testing and evaluation of students, teachers, and schools.
Not surprisingly, these efforts have met with fierce resistance from teachers’ unions. Critics argue that teachers are being blamed for system-wide deficiencies and forced to “teach to the test.” While working in New York, Cerf fought widely publicized ideological battles with Randi Weingarten, president of the American Federation of Teachers. In spite of their disagreements, Weingarten credits Cerf with “remarkable flexibility in working with teachers, not against them,” and believes that students’ well-being is his first priority.
Cerf calls the argument against testing and teacher accountability a straw man. “Every single instance where tests are used is based on progress, not performance.” Students within a school are tracked relative to peers of similar ability, not compared automatically with the top end of the spectrum. Over time, the data reveal patterns: Boys’ reading scores in Mr. X’s class get consistently worse while girls’ scores get better. Ms. Y’s special needs students remain stagnant while Ms. Z’s excel.
Cerf calls the argument against testing and teacher accountability a straw man. “Every single instance where tests are used is based on progress, not performance.” Students within a school are tracked relative to peers of similar ability, not compared automatically with the top end of the spectrum. Over time, the data reveal patterns: Boys’ reading scores in Mr. X’s class get consistently worse while girls’ scores get better. Ms. Y’s special needs students remain stagnant while Ms. Z’s excel.
NJ Smart – A Linked Community of Educators
Central to Acting Commissioner Cerf’s data-gathering efforts is a system called NJ Smart. By the fall of 2012, it will be able to track student data at the level of the classroom.”It’s going to be a queriable database,” says Cerf, that will “enable teachers form all across the state to identify pockets of success elsewhere with children very much like they’re serving and then to access lesson plans and suggestions different teaching methods.”
That’s only the beginning. The information can be used in many ways––to redistribute students to classes that can better meet their needs, to provide professional development to teachers, to replace a principal whose hiring practices seem arbitrary. Most teachers, says Cerf, are not at the far ends of the spectrum––brilliant or incompetent. Most are not likely to be anointed or summarily fired as a result of the data. Furthermore, Cerf points out, student testing is only 50% or less of the information used to evaluate teacher effectiveness. Classroom observations by staff developers and principals add nuance to the picture. Lastly, while teacher effectiveness is an active focus of these reforms in New Jersey and elsewhere, so are overall school quality, administrative efficiency, and curriculum development.
The swift, comprehensive nature of these reforms pretty much guarantees that there will be casualties at first. Individual schools and teachers will be unfairly judged or penalized based on imperfect data-collection systems or policies that don’t fully take local realities into account. This is deeply unfortunate, but not necessarily permanent. Reforms can be reformed. Misguided policies can be rewritten in light of new information. Bad tests can be evaluated and improved. What data-driven reform offers is a more precise picture of what’s happening in schools. It can’t generate individualized instruction like the robot-teachers of the future, but it can help teachers, administrators, and school systems to do so.
What’s the Significance?
Perhaps Cerf himself puts it best:
You look at the unemployment rates. You look at the rates of contact with the judiciary system, incarceration rates. You look at all the leading indicators of failure that stem directly from the failure of our system to adequately prepare kids for life. Then come back to me and say, well, we actually ought to be moving more slowly here and we ought to be being more careful. We ought to be more thoughtful about bruising feelings.
I just fundamentally reject that. I think that we are presiding over a deeply shameful reality where the fundamental problems of equality are not being resolved. And if you were born poor or black or Latino in the urban core of America, the possibility that you will graduate from high school prepared to be successful is tragically low. And I think we should, with every fiber of our beings, resist accepting that and fight against that with all we have.
That’s only the beginning. The information can be used in many ways––to redistribute students to classes that can better meet their needs, to provide professional development to teachers, to replace a principal whose hiring practices seem arbitrary. Most teachers, says Cerf, are not at the far ends of the spectrum––brilliant or incompetent. Most are not likely to be anointed or summarily fired as a result of the data. Furthermore, Cerf points out, student testing is only 50% or less of the information used to evaluate teacher effectiveness. Classroom observations by staff developers and principals add nuance to the picture. Lastly, while teacher effectiveness is an active focus of these reforms in New Jersey and elsewhere, so are overall school quality, administrative efficiency, and curriculum development.
The swift, comprehensive nature of these reforms pretty much guarantees that there will be casualties at first. Individual schools and teachers will be unfairly judged or penalized based on imperfect data-collection systems or policies that don’t fully take local realities into account. This is deeply unfortunate, but not necessarily permanent. Reforms can be reformed. Misguided policies can be rewritten in light of new information. Bad tests can be evaluated and improved. What data-driven reform offers is a more precise picture of what’s happening in schools. It can’t generate individualized instruction like the robot-teachers of the future, but it can help teachers, administrators, and school systems to do so.
What’s the Significance?
Perhaps Cerf himself puts it best:
You look at the unemployment rates. You look at the rates of contact with the judiciary system, incarceration rates. You look at all the leading indicators of failure that stem directly from the failure of our system to adequately prepare kids for life. Then come back to me and say, well, we actually ought to be moving more slowly here and we ought to be being more careful. We ought to be more thoughtful about bruising feelings.
I just fundamentally reject that. I think that we are presiding over a deeply shameful reality where the fundamental problems of equality are not being resolved. And if you were born poor or black or Latino in the urban core of America, the possibility that you will graduate from high school prepared to be successful is tragically low. And I think we should, with every fiber of our beings, resist accepting that and fight against that with all we have.
*High school graduation rates for students who entered 9th grade in NJ public schools. Commissioner Cerf’s office points out that until NJ Smart, graduation rates were reported by school districts, not overseen at the state level. Other states that have made the switch have seen a marked decrease in official graduation rates.
Sign up for the Smarter Faster newsletter
A weekly newsletter featuring the biggest ideas from the smartest people
