Four Lessons From New York's Test Results

Ω

by Gordon MacInnes, a fellow at The Century Foundation



New York State Chancellor Meryl Tisch and Commissioner of Education Joseph Steiner deserve the "Whistle Blower of the Year" award for laying bare the deception and softness of state standards and testing. In releasing the 2010 state test results on July 28, they expose the majority of states that have been lowering proficiency standards as a part of the No Child Left Behind game. Their effort to align New York's tests to what students need to be college-ready sets an example for every state and, one hopes, federal education officials.

One would be hard-pressed to argue with the assumptions behind New York's new policies. First, the goal of a public school education should be college readiness. With half of "high-need" students not finishing high school and one quarter of first-year college students requiring remedial course work, this goal is obviously not being met. Second, the state's curricular standards should be aligned with what high school graduates need to succeed in college. Third, assessments of those standards should provide a reliable indicator if a student is on the path to college readiness, i.e., if deemed proficient on 8th grade math that there is a strong chance that he or she - with continued effort - will pass the Regents examination and gain admission to college.

In raising the bar, New York ends the fantasy about swift and dramatic gains in student achievement. Not surprisingly, the consequences are most severe for low-income, black and Latino students:

- The percentage of economically disadvantaged students in grades 3-8 scoring "proficient" or advanced proficient declined from 66.9 percent in 2009 to 39.1 percent in 2010;

- The percentage of English Learners found below basic more than quadrupled from 9.2 percent in 2009 to 39.4 percent in 2010;

- The percentage of black students scoring proficient or better tumbled from 64.3 percent to 34.4 percent; and,

- The fall was deepest for students in charter schools, the percentage of whom scoring proficient or better fell from 76.1 to 43.0 percent.

The sensible honesty of the New York leadership deserves national attention, particularly from advocates of "transformational reform."

Lesson One: Claims of rapid, consistent, and sustainable gains for concentrations of students from poor families should be received with skepticism.

Mayor Bloomberg's 2009 campaign trumpeted the sharp gains made by New York City students under mayoral control of the schools. In fact, New York City students have made modest - not dramatic - progress as measured by scale scores and by comparison with other big city districts in New York. However, most of the progress of NYC students is explained by the State's poorly designed tests that changed little year-to-year and by too-low proficiency standards. The 2010 results also call into question the credibility of a widely touted evaluation of NYC charter schools by Caroline Hoxby of Harvard that projected closure by high-need students with the achievement in high-performing suburban districts.

Lesson Two: state administration of standardized tests is not to be trusted when there are material consequences to low performance.

In the culture of public education, the Tisch-Steiner initiative represents a rare act of courage. As previously documented, most states have played games with the requirements for "adequate progress" under No Child Left Behind. New York's leaders pointed out the gap in the stagnant and modest level of achievement by New York students on the National Assessment of Educational Progress (NAEP) contrasted with the rapid, substantial improvements on New York's own tests. Most states ignore these gaps, happily pretending that their students are highly proficient and ready to meet the demands of a globalized economy. For example, states with the highest proportions of proficient students on state tests like TN, GA, and MS are found in the bottom quintile of performance on NAEP.

Lesson Three: because of Lesson Two, policymakers should avoid using the results on state-administered tests for judging schools, principals, teachers, or districts.

Using New York City as a case study, a large percentage of its schools avoided being tagged as "in need of improvement" under NCLB between 2006 and 2009 because the percentage of "proficient" students rose steadily. Those same schools are now jeopardized by the sharp reversal in the 2010 results. Commissioner Steiner pledges to seek a waiver from the US Department of Education so that schools will not be penalized. If rejected, there will be a wholesale surge in "failed" schools.

Given the great diversity among states in the quality of their academic standards and the reliability of their tests, it is not sensible that federal policy requires all states to use standardized test results in evaluating teachers and principals. Yet, the USDE mandated that each state certify that there is no statutory or regulatory bar to matching test results to individual teachers as one of only four absolute requirements to even apply for Race to the Top grants. Of course, test results can be in the mix of criteria for judging teacher performance, but they cannot dominate the evaluation. Secretary Arne Duncan, while recognizing on the one hand the weakness in the standards and assessments practices of so many states, pushes on the other hand for wedding test results to individual teachers for improved "accountability."

Lesson Four: the victims of game-playing on state assessments are students and their parents. The negative consequences are material and life-long.

Here is how Chancellor Tisch put the issue:

We are doing a great disservice when we say that a child is proficient when that child is not. Nowhere is this more true than among our students who are most in need. There, the failure to drill down and develop accurate assessments creates a burden that falls disproportionately on English Language Learners, students with disabilities, African-American and Hispanic young people and students in economically disadvantaged districts - who turn out to be much further behind than anyone recognized.

Judging poor performance "proficient" produces false hopes of college readiness. For example, SUNY and CUNY campuses use a score of 80 on the Regents math examination to qualify for admission and avoid remediation. Yet, 8th grade students deemed barely proficient on the 2009 state math test have less than a one in three chance of scoring 80 on the Regents test. Moreover, forty-four percent of those admitted to community colleges face remedial courses that take time, tuition, and defer graduation. Many of these students were declared proficient on New York tests.

There is another word for this official deception:

cruelty.