The Public Policy Institute

Separate And Unequal

Section 4

Teaching All Kids to Read

As the data demonstrate, Commissioner Mills was offering a notable understatement when he said last year that the new 90 percent standard raises "significant implementation issues" for the system.

"We have produced long lists of 'failing schools' in the past only to be distracted by controversy about the accuracy of data and the adequacy of funding," Mills says.

"Many educators are concerned about a change of standards but they are determined to increase standards and performance for their students. The task is to raise the expected level of performance for schools in a manner that inspires productive action and does not provoke excuses or self-protective behavior."(9)

Watch the winners

Nearly all the children who attend high-minority public schools in New York's poor inner-city neighborhoods not only should be taught to read -- they can be taught to read. State test data point to enough high-performing schools, in both the public and non-public sectors, to prove it can be done.

The list of high-minority public schools at the end of this report includes several dozen schools that consistently produce above-average reading scores among students from similarly disadvantaged backgrounds. Perhaps more time and attention should be devoted by the Education Department and by other schools to figuring out what the high-scoring schools are doing right -- and to how it can be replicated elsewhere.

For example, the Family Academy in Harlem produced a dramatic jump in its students' reading scores in 1996 after revamping its curriculum to create a "finely tuned balance" between two district approaches to reading -- skill-based phonics and whole language. The Manhattan Institute's privately funded Center for Educational Innovation (CEI) sponsored a series of weekend reading retreats for city school teachers, aimed at replicating the Family Academy's success. The result was consistent improvement in scores among the mostly high-minority schools involved in the program, CEI reports.

But the system today provides no meaningful incentives for success. In terms of pupil performance, there were absolutely no strings attached to the record school aid increase in the 1997-98 New York State budget. As in the past, schools that consistently fail to teach their pupils to read will get added funding on exactly the same basis as similarly situated schools that do a much better job. And tens of thousands of kids will continue to move through the public schools of New York every year without learning how to read.

Research-based methods

Nor does the system seem to have learned how to transfer the techniques that are working in some schools, and put them to use in the schools that aren't performing. There is, however, growing attention being paid in education circles to an idea that seems fairly obvious, but has long been given short shrift:

Undertake objective and scientific research to figure out which teaching techniques work, and which ones don't -- and then use the ones that work.

Perhaps the most comprehensive research of this kind has been undertaken by the National Institute of Child Health and Human Development (NICHD), an arm of the National Institutes of Health in Washington. Over the past 15 years NICHD has studied the efforts of some 10,000 children nationwide to learn to read -- a project the results of which have only recently been widely distributed among education policy-makers.

Dr. G. Reid Lyon, chief of the child development and behavior branch of NICHD, summarized the findings in testimony for Congress in July:

"We have learned that for 85 to 90 percent of poor readers, prevention and early intervention programs that combine instruction in phoneme awareness, phonics, spelling, reading fluency, and reading comprehension strategies provided by well-trained teachers can increase reading skills to average reading levels."(10)

Lyon's research is at the heart of a growing national controversy over methods of teaching reading. Teachers' colleges, consultants and many state departments of education (including, at times, New York's) have for years been promoting the so-called "whole language" approach to the teaching of reading. This lays aside the traditional method of teaching pupils that speech is made of small sounds ("phonemes") and that letters represent sounds ("phonics"). Instead, in whole language (also sometimes referred to as "look-say" or "the sight method"), students are to concentrate on whole words, figuring out their meaning as much from context as from the letters in the individual words themselves.

The initial research on "whole language" was said to have demonstrated that pupils found it less tedious than phonics-based learning. But now there is a growing suspicion that many students who are alleged to have learned reading from "whole language" programs also got unofficial instruction in phonics -- at home, from their parents, often beginning before they started their formal schooling. In other words, it is possible that they learned reading in part because their parents taught them phonics, even if the school didn't. If that is the case, then it would follow that students without this kind of family backup will do less well in learning to read in a "whole language" environment.

There is anecdotal evidence suggesting that parochial schools, which do better in teaching reading to high-minority student bodies, adhere more often to the more traditional phonics-based methods; but there is no comprehensive information on their methods. From the data now available, in fact, it is not possible to track the impact that "whole language" and the rejection of phonics might be having in high-minority schools. The state Education Department says it has not recommended a particular method of teaching reading to the schools, and it does not track what method is used in individual schools. Reviewing an early draft of this report, Education Department officials said "there is no proof that the difference in performance between public and non-public schools can be attributed to differences in instruction and program." They agreed, however, that "programs of instruction should undergo rigorous evaluation and those that do not improve student learning should be discarded." The department, they added, "is already working on doing this." And they said they believe there should be "a balanced approach between skill-based phonics and whole language reading instruction."(11)

The research conducted by the NICHD has concentrated on pupil-by-pupil performance (as opposed to school-by-school), and Lyon says the findings are quite clear:

"In order for a beginning reader to learn how to connect or translate printed symbols (letters and letter patterns) into sound, the would-be reader must understand that written spellings systematically represent the phonemes of spoken words. Difficulties in decoding and word recognition are at the core of most reading difficulties."(12)

Why not put more time on the task?

Incredibly enough, another reason for the sub-standard reading performance of many New York schools may be that they don't spend all that much time teaching kids how to read.

A survey conducted in the spring of 1997 by New York State United Teachers (NYSUT) found that students in kindergarten through fourth grade in New York spend an average of 68 minutes a day in formal reading instruction. The amount of time dedicated to learning how to read was greatest in first grade, where the average was 80 minutes, according to the survey. NYSUT, however, has recommended that instructional time should be a minimum of 90 minutes a day in early grades, especially in first grade.

With an emphasis on early intervention to correct reading problems before they become severe, NYSUT's comprehensive "Reading for Life" plan also calls for standard readiness tests for all students entering kindergarten and "safety-net" tests to identify reading problems in first through third grades.(13)

Testing still a key

It will also be important to ensure that there is valid testing of results, on a basis that makes it possible to determine the progress -- or lack of progress -- being made by individual schools and by the overall system.

New York City, where the problem of failing schools is concentrated, has been inconsistent in its local reporting of reading test results over the years.

Like the rest of the state, the city is covered by the Regents' requirement to administer the DRP reading test in both third and sixth grades. In addition to this mandate, however, the city is required by the 1969 school decentralization law to rank all schools "in order of the percentage of students reading at or above grade level"(14) -- a relativistic standard based on statistical norms, as opposed to the fixed standard used for statewide measurement purposes.

New York City's ranking of schools according to "grade-level" reading ability has discouraged useful comparisons between the city and other urban areas of the state that share similar social and economic challenges on a smaller scale -- but that often produce better results.

The DRP was used as the citywide reading test in all grades from 1985 through 1995. When the city switched to a different test in 1996, the percentage of students reading at or above grade level plunged by six points. City school officials blamed the difference on the new test, which they said was "more difficult."(15) Implying that there actually had been little real change in reading performance, the city's official report of reading test results said the 1996 average score for grades 3 through 8 on the new test was equivalent to a decline of just 0.5 percent scoring at or above grade level on the DRP.

While the city acknowledged that it was still administering the state-mandated DRP test in Grades 3 and 6, its June 1996 reading report did not disclose those results. The Public Policy Institute's analysis of the results, however, showed more significant declines of two and four percentage points, respectively, in the number of pupils scoring above the state minimums on the third and sixth grade.

The Education Department is now reformulating the testing program for the early grades. Much will depend on whether the new assessments are rigorous and consistent, and produce results that can be compared year-to-year and school-to-school.

Another step forward would be the adoption of an intensive remedial reading program for kids who are identified as falling behind -- such as Governor Pataki's proposal for a six-week summer program funded largely by state aid.

Bottom-up reform

If the public schools do not begin producing better results, it is clear that parents, particularly in the poorly served minority communities, would like the chance to take matters into their own hands.

Politicians and educational policymakers have stubbornly resisted the idea of expanded school choice, including choice of non-public schools. But at some point, New York will have to confront the question of whether the state has a right to force minority kids to stay in schools that don't perform and don't improve.

Two privately funded scholarship programs have provided plenty of evidence that parents whose children are stuck in low-performing schools would welcome choice. In 1996, Albany's nearly all-black Giffen Elementary School posted the worst reading scores in upstate New York. When philanthropist Virginia Gilder offered to pay up to $2,000 in private or parochial school tuition for any Giffen student, many families initially jumped at the chance to pull their children out of the school. Even before the 1996-97 school year started, the choice program had led to a wholesale restructuring of Giffen and its programs -- a fact that, in turn, led a number of families who had initially planned to move their children to private school to stay with Giffen.

In New York City, meanwhile, 16,000 parents and guardians entered a lottery that distributed 1,300 privately funded scholarships in private and parochial schools, starting in the 1997-98 school year. Eighty-five percent of the scholarships have been reserved for students attending the city's worst public schools, as ranked by performance data.

The idea of school choice is deeply controversial; the education establishment insists that it will only undermine the public schools. Yet it is hard to imagine that anything could undermine the public schools more than the system's own failures -- particularly in teaching minority kids to read -- have already done. For all their eagerness to fight off choice, educators will eventually have to confront the fact that it is the system's manifest failures that lie behind the pressure for this kind of change. It will be increasingly hard to deny the concerns of parents who want to find a way for their children to get out of rotten schools. And it will become increasingly hard to justify a policy of denying kids a way out of schools that don't perform and don't improve.

Ultimately, the duty to teach minority kids to read is a fundamental challenge that the public schools must meet. Or their failure will cost New York dearly.

9. Commissioner's Report to the Board of Regents, March 1996.

10. G. Reid Lyon, testimony before the House Committee on Education and the Workforce, July 10, 1997.

11. Roseanne DeFabio and Martha Musser, memorandum dated Sept. 19, 1997.

12. Op. cit.

13. NYSUT news release, May 1, 1997.

14. State Education Law Section 2590(j)5(a).

15. "New York Elementary Students' Math Scores Rise in All 32 Districts," New York Times, June 21, 1996, p. B1.

Contents  Preface Introduction  Section 1  Section 2  Section 3   Appendix I

The Public Policy Institute of New York State, Inc.
111 Washington Avenue, Suite 400 |  Albany, New York 12210-2289  |  518-465-7511