School evaluation that’s data driven—over a cliff

Posted on June 10, 2015

Last week, a very distinguished panel convened by the National Research Council published an Evaluation of the Public Schools of the District of Columbia. The report is 341 pages long and cost millions of dollars to produce. What’s most impressive about this Evaluation is how very far removed from reality it is.

The experts who contributed to the analysis relied principally on data sets that covered the city’s DC-CAS standardized tests, the NAEP nationwide standardized tests, and the local teacher evaluation model called IMPACT. They also considered other data such as graduation rates, attendance, dismissal, and teacher retention. The third of three major recommendations from this Evaluation cannot be denied: the school system needs to address the so-called “achievement gap,” which—as noted elsewhere—has been greatly exacerbated since “school reform” came to the District in 2007.

What are recommendations one and two? The first is to create “a comprehensive data warehouse.” The second is to pay for ongoing independent evaluation of this data. Really.

The purpose of this column is not to criticize the National Research Council’s Evaluation. To be fair, the Council did what it was asked to do. The problem is not this particular study; it is how this Evaluation perfectly illustrates the way our nation’s education debate has become a shipwreck, adrift in a sea of numbers. In search of objective “metrics,” education experts have lost sight of the purpose of public education: to provide each and every child the opportunity to achieve their fullest potential in life.

To compile a real evaluation of the District’s public schools, one needs to interview teachers who have served long enough to remember what the school system was like before the mayoral takeover of 2007. (Unfortunately, the overwhelming majority of experienced teachers have been driven out of D.C.’s high poverty schools, and those few who are left insist on anonymity because truthful testimony would surely get them fired.) But here is what they would tell you:

  • Students in high-poverty schools are forced to waste tremendous amounts of instructional time preparing for, practicing, or taking standardized tests. Most students took about 20 days of tests and pre-tests this year. All the test-taking skills taught and employed during these days are useless in real life.
  • Teachers in high-poverty schools are required to teach to the standardized tests; in fact, their jobs depend on it. Homework is often in the style of test questions downloaded from a central test-oriented website. This mode of instruction and form of homework would astonish those of you who attended middle-class or upper-middle-class schools.
  • Course offerings to high-poverty students have been narrowed to focus their studies on reading and math. It is common for these schools to provide little or no band, chorus, art, or foreign languages. Some schools have required students to take double periods of reading and math. Some require kids to take classes specifically to improve test-taking or study skills. These children’s knowledge of geography, history, social studies, and science is often abysmal—because these subjects are hardly taught. (Why? They’re not on the standardized tests.)
  • When untested subjects are offered, they are frequently overcrowded. A social studies teacher might have 40 students enrolled in a class. (Again, why? It’s not on the test.)
  • Because teachers who have a lot of experience in high-poverty schools know that this system is damaging the students and can’t bear to participate, most left long ago. The teachers in these schools now are overwhelmingly young, inexperienced, and far less capable of controlling children with behavior problems. The principals tend to be inexperienced and ineffective as well.
  • Numerous experimental programs have been tried in high-poverty schools, and to the best of my knowledge, not a single one has proven successful. One program that paid cash to students for grades, test scores, and behavior was a particularly catastrophic failure. “Blended learning” (i.e., computer instruction) for math crashed and burned. And while the school system continues to lengthen the school day for high-poverty students, crowding out afterschool enrichment programs and cutting elementary school recess to 15 minutes, it simply has not worked—it’s just an exercise in “the beatings will continue until morale improves.”
  • And then there are the charter schools, which now enroll about 45 percent of public school students. Using intimidating enrollment procedures and “counseling out” low-scorers and behavior problems during the school year, charters have systematically cherry-picked from neighborhood schools the great majority of easier-too-educate students. One illustration of this: of the 31 schools with the highest percentages of at-risk children (as defined by the school system) three are charters and 28 are regular neighborhood public schools (see that here).

(Thanks to a half-dozen current and recently-retired DCPS teachers for the information above.)

How can any objective group evaluate a school system and miss all these crucial facts? By focusing on “data”! When dealing with the behavior of human beings, only the tiniest slice of useful information can be quantified. In real life, almost everything about humans is too complex to be reduced to averages. The testing/data mania in education is not only failing to measure how well schools are serving their students, it is acting as a smokescreen that covers up the real problems in public education.

America needs public schools that focus on offering every child his or her best opportunity to learn. To accomplish that, we have to recognize there are no standardized children; every single child has different strengths and weaknesses. That’s why all our schools—and especially high-poverty neighborhood schools—must offer a complete curriculum provided by professional teachers who have the training to give the individualized attention every child needs. We’re not doing that in the District of Columbia—we’re doing the opposite! And D.C. is not unusual. By making “data” the be-all and end-all of evaluation, we are failing our nation’s children and damaging our nation’s future.