The City Sentinel

Data-driven school reform failures, and hopes for future progress

Staff Report Story by on November 19, 2011 . Click on author name to view all articles by this author. You can follow any responses to this entry through the RSS 2.0. Both comments and pings are currently closed.


By John Thompson


It has been a bad year for data-driven school reform. Last year when the documentary “Waiting for Superman “debuted, it was claimed that accountability, driven by standardized metrics, was overcoming poverty to increase student performance. Virtually all of such claims have been discredited. Even true believers in test-driven accountability are publishing results that call their experiment into question. (They also stress, accurately, that there have been a range of increases in Math scores, while acknowledging failure to improve Reading scores.)


The most comprehensive study into standardized testing’s failure was issued by the National Academies of Science. The blue ribbon panel, citing twenty years of data, showed that these reforms have not increased student performance, even though they cost tens of billions of dollars and reduced the graduation rate by 2%.
Last month, an equally august body, The Consortium of Chicago School Reform, determined, “the publicly reported statistics used to hold schools and districts accountable for making academic progress are not accurate measures of progress.” It explained why twenty years of reforms in Chicago produced minimal gains, despite the district’s annual claims to the contrary. (Oklahoma is another poster child for producing a “bubble” in test scores, reported under the NCLB – No Child Left Behind – rules, as our results on more reliable tests have been flat.)


We have also learned that the too-good-to-be-true results of schools in Atlanta, Baltimore, D.C., Philadelphia, and New York City were the result of cheating, as well as other ways of fabricating data. Increases in graduation rates, for instance, were padded by “credit recovery” gimmicks or the massive re-scoring of tests. Recently, another New York City high school was caught in a scheme to pass students who failed classes.


The Charlotte Mecklenburg School system recently won the Broad Prize for market-driven reforms, but its hugely expensive multi-year experiment in school choice failed to improve student performance. The study’s co-author, Tom Kane, heads the Gates Foundation’s effort to evaluate teachers using test score growth, but his study questioned whether it makes sense to hold individual schools accountable for not overcoming poverty. That, of course, was the foundation of the data-driven reform movement.


Houston also has stumbled when in applying “No Excuses” methods and data-driven instruction to neighborhood schools. At the cost of more than $2,000 per student, Houston produced increases in Math but not in Reading. Online tutoring and double doses of Math and Reading classes were ineffective. Although Houston has not been forthcoming about their attrition rate, its program started last year with 7,385 students. They were 86.6% economically disadvantaged. The students who stuck it out to the spring testing were 61% economically disadvantaged. Apollo schools reopened this fall with only 6,156 students.


It is to the credit of these reformers that they concede that old-fashioned policies, which they had dismissed as the “status quo,” were more effective.


Even though test-driven, market-driven reform failed again, increased investments in tutoring (one adult for two students), counseling, and establishing safe and orderly school environments were shown to increase the high school graduation and the college-going rates.


If nothing else, research by the best of the accountability hawks are providing a service in documenting how their preferred policies have failed, and how teachers and education scholars had long been prescient about what actually works. We are still clueless in regard to scaling up practices for overcoming intense concentrations of generational poverty. It has become clear, however, that traditional reforms that have long been advocated by teachers are more effective than newfangled data-driven experiments


Editor’s Note: John Thompson holds a doctorate from Rutgers University, and is an award-winning Oklahoma historian. Formerly an award-winning teacher in the city public schools, and a member of the KIDS (Keep Improving District Schools) Committee that laid the basis for MAPS for Kids, Thompson is now an education writer based in the MidTown area.

Comments are closed

Click For Western Concepts
Log in | Designed by Gabfire themes