Letters 98
EPP August 1998 Comments on State Class Size Reduction Regulations

August 31, 1998

Mr. James Kadamus
Deputy Commissioner
New York State Education Department
Room 875 Education Bldg Annex
Albany, NY 12234

Re: Proposed Regulations on Class Size Reduction:
Proposed amendment to Section 144.11 Regulations of the Commissioner of Education pursuant to Chapter 436 of the Laws of 1997

Dear Deputy Commissioner Kadamus:

As you may know, the Educational Priorities Panel is a very strong supporter of initiatives to reduce class sizes in the early grades. We were disappointed that, at the conclusion of the legislative negotiations surrounding the August 1997 budget agreement, funding for class size reduction grants 1) were made available only starting in the 1999-2000 school year and 2) would not enable all public schools in the New York City to reduce their class sizes to an average of 20 students from kindergarten through third grade. Of even greater disappointment to us was the failure to secure voter approval of the School Construction Bond Act in 1997 and the Governor’s recent veto of $500 million in additional funding for school facilities. Both houses of the legislature intended to provide school districts with the resources to create additional classroom spaces, but these resources are now not available for the 1999-2000 school year.

On the positive side, however, the lag in implementation has provided crucial planning time so that New York State can avoid the mistakes of California, where hastily drafted state directives have not encouraged a cost effective use of resources or the best strategies for increasing student achievement. EPP is impressed with the thoughtfulness of the proposed regulations for New York State’s initiative. Given the length of this letter, we will dispense with an itemization of numerous agreements with the proposed language and will restrict our comments to the sections where we have serious concerns:

§144.11 (b) (5) (c): Definition of District Plan

While EPP is a strong advocate of class size reductions in the early grades, we do not believe that these reductions are a "magic bullet" that will automatically result in high quality instruction. Smaller class sizes are a "value added" benefit. They will make a good teacher a better teacher and an inadequate teacher less inadequate. Similarly, class size reductions will improve student achievement levels in both high-performing and low-performing schools, but class size reductions will not transform a low-achieving school into a high-achieving school. For example, four years after Tennessee began reducing class sizes in 16 of its poorest districts (Project Challenge), the average student test scores in second grade moved these districts up on a list of 138 school districts by order of achievement from 97th place to 78th place in reading and from 90th place to 56th place in math. In short, these schools moved from low performance to average performance, not high performance. Class size reductions are a means of "leveling up" student academic achievement in High Needs school districts.

Research on the effects of class size is still ongoing, but the soundness of the research designs as well as detailed data collection of the Tennessee Student/Teacher Achievement Ratio (STAR) project and the Milwaukee Student Achievement Guarantee in Education (SAGE) project strongly indicate that:

• The greatest gains in student academic achievement occur in kindergarten and first grade and these gains are then sustained in second and third.

• The academic achievement gap between white and black students and between lower income and higher income students closes significantly for students who have been in small class sizes from kindergarten through first grade. Once the academic achievement gap occurs for students in large class sizes, their entry into smaller class sizes in second grade and third grade does not close the gap.

We agree with the conclusions of Charles Achilles, Steven Bingham, and Alex Molnar that class size reductions are a preventive strategy, not a remediation strategy. Smaller class sizes in kindergarten and first grade produce robust gains in student learning and statistically significant decreases in rates of special education referrals and grade retention. These students’ higher achievement levels are sustained in the higher grades. But students who have been in large classes in kindergarten and first grade do not make significant learning progress once their cohort enters smaller classes in second and third grade.

A cautionary tale is the haphazard way in which class size reductions have been implemented in California. In the first year, the Los Angeles Unified School District spent its funds to reduce class sizes in first and second grades rather than kindergarten and first grades and also allowed "team teaching" despite no observable benefit from this strategy in Tennessee and Nevada. In contrast, the San Francisco Unified School District, in addition to first and second grade class size reductions, reduced all of its kindergarten classes, while in the rest of the state only 16% of kindergarten classes were reduced to 20 students. True to the experience of Tennessee and Milwaukee, San Francisco first graders made a net gain of 7.0 Normal Curve Equivalents in their reading scores on the Comprehensive Tests of Basic Skills and a net gain of 5.8 NCE in their math scores on the CTBS. NCE net gains for second graders, who had not been in smaller kindergarten and first grade classes, were modest and in the range of 2.5 for both reading and math.

What particularly concerns EPP about the absence of a "best practices" requirement in the District Plan is that the NYS Education Department’s accountability system for elementary school performance currently begins at the third grade and next year will begin at the fourth grade. Principal after principal told us in our interviews for Getting Off the List that their strategy for preventing placement on the SURR list and for getting off the SURR list was to assign their best teachers to the third grade and, in some cases, reducing the class sizes for this grade alone. We also note that superintendent contracts for independent school districts and for community school districts are often only of a two or three year duration, thereby also encouraging investment in third grade classes because this is the grade that that is closest to the benchmark tests that will put their schools "on the radar screen" for measurements of academic performance. EPP strongly urges the State Education Department to create an inducement for school districts to begin implementation of class size reductions in kindergarten and first grade in order to counter all the built-in incentives for implementing them in the third grade.

In most cases, EPP fully supports flexibility at the school district level and an absence of micromanagement from SED. But as advocates of cost effectiveness and, and most importantly, as advocates for children’s learning, we believe that the best use of taxpayers’ money for the greatest benefit to children requires that the Regents and the State Education Department provide some clear and unambiguous guidance to local school districts as to where best to begin implementation of small class sizes. There may indeed be sound reasons why a few school districts might need to implement reductions beginning in other grades beyond kindergarten and first grade, so waivers should be allowed. EPP has done the math, however, and based on the experience in Tennessee, Milwaukee, and San Francisco, school districts that begin class size reductions in the third grade and then implement reductions in the lower grades in subsequent years will only begin to see significant student academic gains in the fourth grade test scores after eight years. In contrast, those districts that begin reductions in class size for kindergarten students and then reduce class sizes as this cohort moves to higher grades should see higher student achievement on fourth grade tests in five years.

EPP is especially anxious that the student cohort that has participated in the first year of the universal pre-kindergarten program enter into smaller kindergarten classes. We are disappointed by the absence of an attempt to strengthen both LADDER initiatives through better directed planning. In addition, we note that New York State is now "under the gun" in fully complying with the Least Restrictive Environment and Neutral Funding requirements of the 1997 reauthorization of IDEA. Class size reductions, if implemented first in the earliest grades, can be an important support for school district efforts to prevent inappropriate special education referrals and to integrate more children with disabilities into the general education classroom.

§144.11 (b) (3): Definition of Schools with the Greatest Need

We want to commend your department for the draft language for this definition. By the end of the 2001-2002 school year, all other eligible schools districts will be able to apply for grants to reduce early class sizes in all their schools, so a school ranking of "greatest need" is of particular importance to New York City where class size reductions will not be universal. EPP fully supports school district options of giving priority to:

The fourth option of giving priority to schools with class sizes that are more than 33 percent larger than the school district-wide average class size, however, may unintentionally have the effect in New York City of denying resources to children in some lower achieving schools. Attached to these comments are a graph EPP has prepared for all community school districts in New York City showing that within 26 out of 32 districts, the three schools where children had the highest average scores on the May 1997 CTB-R English Language Arts test had higher building utilization rates than the three schools where children had the lowest average scores. This is consistent with EPP’s previous study of schools that were removed from the SURR list and our current study of 16 high-achieving schools in low-income neighborhoods, where we have made the finding that school enrollment and building utilization can be influenced by the reputation of the school among parents.

EPP does not want to second guess the intentions of the drafters of the legislation, but it is entirely possible that legislators assumed that community school districts that had higher average class sizes were also those that were overcrowded or had lower performing schools. Quite frankly, EPP held a similar assumption until we looked at each community school district and compared the frequency of class sizes from kindergarten to third grade with average district building utilization rates and average student test scores. Even though classes from first grade to third grade are supposed to have an average of 25 students, this average is exceeded in many districts. The relationship of the three factors (class size, district utilization, and student achievement) is complex and variable across districts. The socioeconomic characteristics of schools and community school districts are another factor that needs to be analyzed. As you know, school performance is correlated to the socioeconomic profile of students served. Poverty status may also play a role in district policies on class size. In EPP’s comparison of community school districts with similar building utilization rates, we find that those serving low-income children tend to have more students in larger class sizes than those serving more middle income communities. For example, Community School Districts 19 and 31 both have a 100% building utilization rate and over 400 early grade classes, yet 5% of District 19’s classes have over 30 students while District 31 has virtually no class sizes above 30 students. 54% of the early grade classes in District 19 range from 25 to 30 students while 57% of classes in District 31 range from 23 students to 26 students. By prioritizing schools with higher than average class sizes on the basis of academic performance, the counterproductive practices of community school districts serving low-income children could be eradicated.

§144.11 (d): Staff Qualifications

EPP supports having the best-qualified teachers in schools, which at a minimum means that they should be licensed or certified to teach children. We do not want to see the class size reduction initiative result in an increase in the numbers of unlicensed teachers working in public schools, which has been the California experience. EPP is also confident that there is currently such a surplus of teachers seeking jobs in elementary schools in New York City that teacher recruitment will not be a problem for these grades. Ironically, the requirement that all teachers providing instruction in the early grades shall possess a teaching license or certificate will result in less capable teachers unless New York City public schools are allowed to choose among licensed and certified teachers. It is urgent that the Regents analyze the city’s teacher assignment system and ask the legislature to make reforms in the Education Law in the next legislative session.

Requirements for teacher licenses and certificates are, at best, only a modest guarantee that some proportion of the least prepared teachers will be disqualified through minimal-competency, "paper-and-pencil" tests. For reasons that we do not fully understand, in the process of eliminating the cumbersome New York City Board of Examiners testing system, the legislature created a streamlined New York City teacher licensing system administered by the Office of Recruitment, Professional Assessment and Licensing that still required an additional bureaucratic hurdle for those seeking to be teachers in the New York City public school system. Possibly because of fear of patronage abuses under the previous school governance system of New York City, central assignment of licensed and certified teachers was retained, which in effect prohibits school principals and school planning teams from directly hiring teachers unless they choose to be an "Options" school. The same problems that EPP documented in our 1987 report, Teacher for the Apple, continue to plague aspiring teachers and school administrators years after the Board of Examiners was eliminated. In many respects, by eliminating assignment of teachers by rank order of their test results (because neither the state examinations or the city ORPAL oral interviews can be validated) and instead requiring assignment of teachers by random order, principals became even more concerned about the capabilities of licensed teachers assigned to them. By 1995, when we issued our Swimming Upstream report, EPP made the belated discovery that some high achieving principals purposely do not declare vacancies and instead hire unlicensed teachers so as to avoid having less capable, but fully licensed teachers assigned to them.

Over the last three years, EPP has been impressed by the New York City Chancellor’s efforts to allow more hiring and assignment flexibility at the community school and district level. Several community school districts have found ways to allow school teams to make final selections of licensed or certified teachers through in-depth interviews and lesson demonstrations. We would like to see the Regents and the State Education Department take the time to understand the causes for the higher numbers of unlicensed teachers in New York City rather than merely focusing on compliance. If the Regents approve these regulations, we earnestly urge them to look at the problem of teacher assignment for New York City under the Education Law and to take action before the 1999-2000 school year.

In closing, we again want to commend the State Education Department for the crafting of these proposed regulations. From the perspective of the Educational Priorities Panel, class size reduction grants represent a significant investment in closing the achievement gap between students in high needs school districts and students in the rest of the state, which reflects one of the major goals of the Regents. It is a source of "leveling up" funding that goes directly to the classroom for the benefit of children’s learning. In a state where little progress has been made in ending stark inequalities among districts in per-pupil funding, class size reduction in the early grades provides a unique opportunity to begin providing the basic instructional supports for inner city students that are common place in more affluent school districts. We hope that the Regents and the State Education Department will do their best to ensure the successful implementation of this initiative.

Sincerely,
Marilyn Braveman, Chairperson
Noreen Connell, Executive Director

CC:
Members of the Board of Regents
Commissioner Mills
Chancellor Crew
Deputy Chancellor Rizzo
Deputy Chancellor Spence