Data Opinion

The Great Accountability Hoax

By Diane Ravitch — June 15, 2010 5 min read
  • Save to favorites
  • Print

Dear Deborah,

The evidence continues to accumulate that our “accountability” policies are a great fraud and hoax, but our elected officials and policymakers remain completely oblivious to the harm caused by the policies they mandate.

Over the past several years, efforts to “hold teachers accountable” and “hold schools accountable” have produced perverse consequences. Instead of better education, we are getting cheating scandals, teaching to bad tests, a narrowed curriculum, lowered standards, and gaming of the system. Even if it produces higher test scores (of dubious validity), high-stakes accountability does not produce better education.

In their eagerness to show “results,” states are dumbing down their standards. The New York state education department dropped cut scores on the state tests from 2006 (the year that annual testing in grades 3-8 was introduced) to 2009. In 2006, a student in 7th grade could achieve “proficiency” by getting 59.6 percent of the points correct on the state math test; by 2009, a student in the same grade needed only 44 percent of the available points. Back in the pre-accountability days, a score of 60 percent would have been a D, not a mark of proficiency, and a score of 44 percent would have been a failing grade. According to a report by The Civic Committee of the Commercial Club of Chicago, the gains registered in the elementary schools of Chicago during Arne Duncan’s tenure were almost entirely the result of changes to the scoring of the tests, rather than evidence of any genuine improvement in student learning.

When gains are manufactured in these ways, children are cheated. Children who need extra help don’t get it, but adults trade high-fives for their “success” in raising scores and enjoy the adulation of the media.

When New York state’s education department was criticized for dropping the cut scores on its tests, officials responded by insisting that the department dropped the cut scores because the tests were actually harder than in previous years. This was utter nonsense because the passing rates soared as the cut scores fell, which would not have been the case if the tests were “harder.” So, although it never acknowledged its past chicanery, the state education department claimed that the tests would really, really, truly be hard this year and that standards would once again be high.

However, some whistle-blowing teachers tipped off the New York Post that the scoring rubrics for this year’s test recommended giving half-credit for wrong answers and even for no answer at all. Here are examples from the 4th-grade scoring guide, as reported in the Post:

  • “A kid who answers that a 2-foot-long skateboard is 48 inches long gets half-credit for adding 24 and 24 instead of the correct 12 plus 12.
  • A miscalculation that 28 divided by 14 equals 4 instead of 2 is “partially correct” if the student uses the right method to verify the wrong answer.
  • Setting up a division problem to find one-fifth of $400, but not solving the problem—and leaving the answer blank—gets half-credit.
  • A kid who subtracts 57 cents from three quarters for the right change and comes up with 15 cents instead of 18 cents still gets half-credit.
  • A student who figures the numbers of books in 35 boxes of 10 gets half-credit despite messed-up multiplication that yields the wrong answer, 150 instead of 350.”

One hopes that these students never become pharmacists or engineers or enter any other line of work where accuracy matters.

The scandal of high-stakes testing is not limited to New York and Illinois. Last week, Trip Gabriel of The New York Times reported in a Page One story about the ubiquity of cheating scandals across the nation. My guess is that he revealed only the tip of the iceberg. I was in Baltimore on May 27, when The Baltimore Sun wrote about a major cheating scandal at an elementary school that had been widely recognized for its excellent test scores. In 2003, only one-third of the students in the school passed the state reading test, but within four years, almost all did. This was a “miracle” school; it won a federal Blue Ribbon for its remarkable gains. But it turned out that the school’s success was phony: Someone had erased and corrected many student answers.

The more that test scores are used to measure teacher effectiveness and to determine the fate of schools, the more we will see such desperate efforts by teachers and principals to save their jobs and their schools.

Yet even as more cheating scandals are documented, even as the perfidy of state testing agencies is documented, our federal policymakers plunge forward, blithely imposing unproven policies as well as “remedies” that have been tested and found wanting. Latest example: The June 9 issue of Education Week has a front-page story with this headline: “Merit-Pay Model Pushed by Duncan Shows No Achievement Edge.” The inside jump headline reads “Student Progress No Better in Chicago Schools Using TAP.” (TAP is the Teacher Advancement Program, which gives extra compensation to teachers for higher “performance.”) In the same issue, on Page 24, is a story about the $437 million for the federal Teacher Incentive Fund program, which will dispense dollars to do what failed in Chicago. Secretary Duncan hopes to expand funding for this program to $900 million next year. Mr. Duncan says of the program, “There’s no secret that historically there’s been some apprehension about doing this kind of stuff. You have to expose yourself a bit and put things on the line, but where folks are willing to do that and do it together, we see the benefits for students. It’s remarkable.”

Merit pay has been tried and found ineffective again and again since the 1920s, but repeated failure never discourages its advocates, who are certain that if the incentives were larger, or if some other element was adjusted, it would surely work. We hear that about every failed experiment. If only we had done it differently....

More emphasis on test scores. More money for teachers if the scores go up. More punishment for teachers and schools if the scores don’t go up. More cheating. More gaming the system. More concentration on basic skills (they count) and more indifference to the arts, history, science, foreign languages, etc. (they don’t count).

Are we in an era of National Stupidity or National Insanity? Or is this what happens when educators imagine they are thinking like corporate executives? If it is the latter, I recommend that they read the writings of W. Edwards Deming, the management guru, who steadfastly opposed merit pay because it destroys collaboration and teamwork, undermines long-range planning, and incentivizes the wrong behavior. If it is the former, well, we will just have to ride out this terrible era and hope that wiser heads someday prevail.


The opinions expressed in Bridging Differences are strictly those of the author(s) and do not reflect the opinions or endorsement of Editorial Projects in Education, or any of its publications.