The sound of bubbles bursting: Student gains on state test vanished into thin air

BY Diane Ravitch

Sunday, August 1st 2010, 4:00 AM

Every year for the past four years, the New York State Education Department has announced dramatic test score gains. And every year, it turns out they were misrepresenting reality. This year, New Yorkers got an accurate accounting of student performance, and it was sobering.

Since 2006, scores have gone through the roof. Teachers and principals quietly told reporters that the tests were getting easier to pass, but no one listened. A few critics and testing experts warned that outsized annual gains were not credible, but no one listened.

At the same time that the state was announcing phenomenal annual gains, national tests administered by the federal government - exams considered the gold standard - told a different story. On those tests, the state's scores in reading were flat from 2000 to 2009. Math scores were up in fourth grade, but not in eighth grade, where they were flat from 2005 to 2009.

New York Commissioner of Education David Steiner made a bold move. He decided to end the inflation - and administer some shock therapy. The sharp contrast between mostly flat scores on national tests and dramatic annual claims by the state made it necessary for him to act, and he did.

Now we know the painful truth. Last year, 86.4% of the state's students in grades three to eight were deemed proficient in mathematics; today it is 61%. Last year, 77.4% of students in the same grades were deemed proficient in reading; today it is 53.2%.

When the scores were released, there was a sound of bursting bubbles across the state. What once were miracles turned into mirages.

Since 2005, Mayor Bloomberg and Chancellor Joel Klein have trumpeted historic gains. But after the state's adjustment, the pass rate on the state reading test among city students fell from an impressive 68.8% to an unimpressive 42.4%, and from an astonishing 81.8% to a disappointing 54% in mathematics. Overnight, the city's historic gains disappeared.

Now, look at the achievement gap between the performance of white students and that of minorities. Last year, black students were 22 points behind white students in passing the state English exam. This year - after the state corrected its scoring - the gap increased to 30.4 points.

In math, the gap grew even more. Black students were 17 points behind whites last year. Now they've fallen 30 points behind.

Charter school advocates saw their bubble burst as well. The pass rates in the state's charter schools, overall, dropped even faster than those in regular public schools. In third grade math, it plunged from 96.1% to 61.6%, and in eighth grade, from 84.5% to 50.4%. On the 2010 reading tests, the scores of charter students in New York City were nearly identical to those of district schools: 43% compared to 42%.

In math, 63% of the city's charter students passed, compared to 54% in public schools, which was an advantage but nothing like the miraculous results previously claimed by charter promoters.

Among other bubbles that popped were the city's school report cards, which based 85% of their grades on the state's test scores, mostly on gains on the test now proven to be vastly overstated. Some schools were given an A for "progress" on dumbed-down tests, and others were closed because they didn't make the grade. But the measure was a deeply flawed instrument.

The hundreds of millions of dollars that the city has spent on test preparation turned out to be a bad investment. Students were learning test-taking skills, not truly learning reading or mathematics.

As a result of the fiasco, we now know that the bonuses of more than $30 million handed out last year to teachers in schools that made "gains" on the state tests were a waste of precious money.

Why does test score inflation matter? Aside from the fact that the state misled the public, the inflated scores caused tens of thousands of students to be denied needed remediation. The inflated scores also help to explain why 75% of the city's high school graduates require remediation when they enroll in community colleges at the City University.

Now we know that achievement in the city and state did not grow by historic proportions, as officials claimed.

The way to avoid similar messes in the future is to use test scores for information and diagnosis, not for rewards and punishments.

Two questions remain: Will Bloomberg and Klein accept this new reality or will they continue to deny the plain facts and refuse to be held accountable? And will the state education department find and fire the bureaucrats and private contractors responsible for this scandal? Unfortunately, the prospects for genuine accountability by the city and state are not promising.

Ravitch is Research Professor of Education at New York University.