Smaller=better?

Bell, Julie Davies

A wave of popular support has buoyed up the idea of reducing school class sizes across the country. But does it work? Maybe so, maybe not.

Reducing the size of elementary school classes is one of the hottest education reforms to hit state legislative agendas. Nationwide, schools average 23 students per class - too large, some say, to provide the kind of individualized attention students really need to succeed. Using research as a guide, supporters are suggesting that states should focus on creating smaller classes. Now states are figuring out what's involved in getting from here to there. It's a story of huge political potential and huge price tags.

It's hard to find many public policy proposals more popular than reducing class size. A March 1997 Wall Street Journal poll found that 70 percent of adults believe that reducing class size would result in big improvements for public schools. A 1997 Education Week survey found that 83 percent of teachers and 60 percent of principals believed classes should not exceed 17 students. Parents say their children are happier and learn more in smaller classes. Teachers report they have fewer discipline problems, are able to give students more individual help and can cover material faster.

Such strong public opinion numbers rarely surface. Naturally, policymakers everywhere are paying attention. It seems simply to make intuitive sense - children can get a better education in a smaller class, where the teacher can get to know them better and give them more personalized attention. But as legislators get into the issue, they are finding they must proceed with caution. Reducing the size of classes may be the most expensive education initiative they have ever considered. And the research is not at all clear on whether smaller classes translate into improved learning. So, how far should legislators go with something that is politically popular, very, expensive and has uncertain outcomes?

"Lower class size is the argument for the reformation of the public school system," says Michigan Senator Joe Conroy, a passionate supporter of smaller classes. Since 1994, Michigan has funded a pilot program in Conroy's home of Flint. It has cost the state approximately $6 million. Conroy says the results are significant: "Forty-three percent more fourth graders are passing the state reading test, and 18 percent more are passing the state math test." Conroy visits one school building a week in his district. He says he has observed a "sea change" in the way classes are conducted, including greater decorum and teacher enthusiasm. "The train is on the track for more to come," says Conroy. This year he helped to develop the next step for Michigan - a $20 million program for statewide implementation of class size reduction in the neediest districts.

Michigan is among 20 states that have passed or are considering passing bills mandating smaller classes. These programs vary in several ways. Most focus on the early grades - primarily kindergarten through third - but some, like Tennessee's, target all grades. States may prescribe classroom averages that schools and districts must maintain or may impose a maximum number of students per class. Some states, such as Arkansas and Virginia, prescribe an average or maximum student-teacher ratio rather than an average or maximum class size. This allows schools and districts additional flexibility because they are able to use teacher aides and still meet state requirements.

20 STUDENTS MAXIMUM

Typically states are setting averages in grades K-3 at around 20 students. Nevada has the lowest mandated size, requiring no more than 15 students per class. Several states approach class size reduction by offering fiscal incentives. In Michigan, the state provides 75 percent of the funds for poor districts that reduce classes to an average of 17 students with a maximum of 19. Oklahoma began its effort with a grant program, but found implementation cumbersome. The state has now set a five-year schedule to reduce classes statewide to a maximum of 20 students per class in grades K-6.

President Clinton has also jumped on board, adding momentum to the already popular movement. In his 1998 State of the Union address, he announced a national effort to reduce classes to an average of 18 students in grades one, two and three. To help states accomplish this goal, he vowed support for hiring 100,000 new teachers who have passed state competency tests, and a construction tax to help build and modernize schools. Estimates are that this plan would cost about $12 billion over seven years and would be an additional incentive for states to reduce class sizes.

If Clinton's plan becomes law, it will help states with the significant cost involved in smaller classes. States considering the move must make a serious budgetary commitment. In states facing increasing school enrollment, the costs will be even greater. In California it is costing an extra $800 or so per pupil per school year. This doesn't even include the cost of new classrooms, estimated to be $1.1 billion. The Michigan class size reduction program for FY 1999 provides nearly $20 million for grants to eligible schools. Tennessee invested about $600 million between 1991 and 1996 to implement its program. In Philadelphia, Superintendent David Hornbeck unveiled plans to reduce class size in kindergarten through third grade from an average of 27 students to 20 students by the year 2002. Estimates are that the program would require a thousand new teachers at $50 million a year and 35 new schools at $470 million.

NO MAGIC NUMBER

Policymakers who look to research to help answer questions about the effectiveness of smaller classes will find some general recommendations, but little definitive evidence. Most states are heeding two general conclusions: They are most effective in the early grades, and they may be the most valuable for at-risk students.

How small do classes have to be before results are observed? Most research indicates that classes must drop to no more than 15 to 17 students. Otherwise, there is no evidence that smaller class size results in better student achievement. But few states have tried such small classes.

"The research hasn't yet answered the questions legislators ask," says Mike Kirst, Stanford University professor and co-director of the Policy Analysis Center for Education (PACE), which provides much of California's education policy research. "There's no consensus on the magic number. There is certainly no research to justify reducing all grades. If we know anything, it is that positive effects are observed when classes around 23 to 25 students are reduced to 15 to 17. Positive effects are observed in grades K-2 - not even third grade - and positive effects are observed in disadvantaged kids."

Most legislators pushing change in class size know these statistics, but argue that any reduction is a positive move. "In a state like California, a reduction from 30 to 20 students is huge. It's the best we can do," says Assemblywoman Kerry Mazzoni, who chairs the Education Committee. The parents, teachers, students and policymakers in California, in fact, are extremely happy over that state's efforts. The research may say that California's new classes are not small enough to make a difference in student achievement, but the classes are definitely small enough to make a difference in public opinion.

Nowhere has class size reduction been such a hit - or as expensive - as in California. The California class size reduction program, passed in 1996, cost nearly $1 billion - the largest infusion of money into the state's K-12 education system since the tax-limiting Proposition 13 passed in 1978. About three-fourths of the money was used to reduce class sizes and the remainder went to extra classroom space. The program is voluntary for school districts, but those participating must reduce class size to 20 or fewer in grades K-3. The state provides the full additional per pupil cost of $800.

Not only was the California program the largest and most ambitious ever, it was implemented in a remarkably short period of time. And it has turned out to be a political bonanza. The 1996 legislation was so popular with the public and voters that everyone has taken credit for it - including the governor, the superintendent of public instruction and legislative leaders - Democrats and Republicans. In a state often marked by its political contentiousness this was quite a story.

Class size reduction has sparked a whole new outlook on education and education reform in California. Over the last 10 years the state's education system has been down in the dumps; the property tax limiting Proposition 13, coupled with a poor economy and rapidly rising school enrollments, caused tremendous strain. California consistently was ranked at the bottom of the 50 states for per pupil spending and achievement. Public confidence in the schools was low.

The original measure moved quickly. Senator Leroy Greene wrote the first piece of legislation in 1996. The proposal was a modest one, targeting one grade a year. But he was concerned about the lack of research supporting whether this was a good investment. "Five months later the governor discovered it, and it became his class size reduction program. He wanted to do it all at once," Greene says.

As a result of the governor's strong endorsement and support from top education leaders, the legislation was signed into law in a few months. The response among parents and teachers was immediately positive. Perhaps the most telling statistic about the program's political success is that 120 legislators listed class size reduction as a major accomplishment in their 1996 campaign brochures.

HURDLES

But there were significant hurdles and very little planning time. California immediately needed 26,000 new teachers. There were not enough certified and substitute teachers for the new classes. The Legislature passed emergency certification legislation, enabling districts to draft candidates from a "reserve pool" of about 200,000 people who had valid state licenses. Some teachers were reassigned from upper grades with little, if any, training. Between the reduction program, the nation's fastest growing student enrollment and an aging teacher work force, estimates are that California will need to hire as many as 250,000 teachers over the next 10 years - more than the current total on the job.

Also immediately needed were 18,000 new classrooms. In California, every unused building was tapped, classrooms were reconfigured, libraries, auditoriums and gyms became classrooms and portable classrooms were put into service. In doing this, California may have already tapped out its low-cost options. The main alternatives left include more portables or new construction. Having already spent about $200 million on new classrooms, the state estimates the program will top $1.1 billion.

That's a hefty price tag, particularly since evaluation plans were not put in place nor were long-term strategies for teachers and facilities adequately mapped out. This year, California has addressed some of the shortcomings of the original legislation to make future steps easier. For example, PACE is among several organizations that will conduct a comprehensive evaluation of the effects of class size reduction.

Senator Greene thinks California is moving too fast. He is concerned about the program's popularity and the response by lawmakers to expand the program by reducing class size in fourth and fifth grades and targeting specific middle and high school classes, such as eighth grade writing. Greene thinks the state would be better off looking at more proven reforms, such as strengthening reading, writing and arithmetic. "Here we are spending a billion dollars without knowing what it's buying."

Assemblywoman Mazzoni continues to be a strong proponent. "Class size reduction has been worth every penny. The deal has had a huge effect on public opinion and brought people together. it has changed public perception about schools and brought back children from the private schools. It's brought the need for adequate facilities to the forefront and is helping to secure the needed public support for funding."

Some good information may soon be forthcoming in California through the PACE study. The evaluation will examine the effect of class size reduction on student achievement; educational opportunities for minority students, students with limited English proficiency, special education and poor students; the quantity and quality of teaching, including professional development and instruction; educational facilities; and parallel effects of other educational reforms.

The California study is likely to offer important information not only to policymakers in that state, but to others considering this substantial investment. In the meantime, plenty of personal testimony points out the benefits of smaller classes. One district in California - the San Juan Unified School District in suburban Sacramento - reported that suspensions were down 19 percent and referrals of students with severe learning disabilities to special education programs were down 16 percent. Significant improvement was made in mathematics scores - an increase of 7 percent to 8 percent. Reading scores went up slightly - 4 percent. The San Francisco Unified School District also has seen some gains in standardized test scores since reducing class size.

States planning to reduce the size of classes can learn a great deal by observing the successes and problems of those who have gone before them. Most states can expect problems in finding available and qualified teachers. At least in the early stages, states must make some important decisions: Is it better to have a small class with a less qualified teacher or a large class with a highly qualified teacher? Those who argue that an underqualified teacher is never a good choice suggest that states should focus on teacher quality, not class size. Some studies have suggested that teacher training and experience are more important indicators of student achievement than is class size. Teacher groups also argue that smaller classes are meaningless if teachers have not been trained to work with fewer students. States must deal with important issues of equity as well: When teachers are in high demand, the wealthier districts are mote likely to be able to pay for the most qualified.

Stanford's Mike Kirst's advice is "target your resources to the very early grades, disadvantaged pupils, 20 or fewer students per class, and a phase-in with ample supply of quality teachers and sufficient space. Deviations from that are hard to justify."

Many policymakers, however, may find that something so popular is much too hard to resist.

RELATED ARTICLE: DOES IT WORK? THE RESEARCH IS MIXED

Although over 1,100 studies examine the relationship between class size and student achievement, no definitive conclusions have been reached. Positive results have been demonstrated in Tennessee and Wisconsin, but other research finds little connection between student-teacher ratios and student performance, especially when measured against other types of educational reforms.

Critics argue that class sizes have fallen for decades, but a corresponding increase in student performance has not occurred. Further, international comparisons demonstrate little support for class size reduction. The average score of eighth graders in Korea and Japan was significantly higher than the United States in both math and science on the Third International Mathematics and Science Study, yet the average class sizes in those countries for 13-year-olds were 49 and 36 respectively vs. 23 in the United States.

The largest and perhaps most convincing study supporting class size reduction comes from Tennessee as a result of Project STAR (Student-Teacher Achievement Ratio) initiated in 1985. Spurred by a study detailing small class size success in Indiana, the legislature authorized a four-year program and study. A second phase of the project, Lasting Benefits, began in 1989 and continues to track students from the initial study to determine whether small class benefits persist over time.

Project STAR examined 7,000 students and teachers randomly assigned to either a small class with approximately 15 students, a regular class with approximately 25 students or a regular class with approximately 25 students and a teacher aide. The four-year study found a consistent pattern of students in smaller classes achieving substantially higher scores than students in regular classes, with or without an aide. The lasting benefits study demonstrates the positive effects of these small classes in grades K-3. Higher achievement from small-class pupils has continued through ninth grade, although the disparity between students in small and regular classes has diminished slightly over time.

A new study of the Wisconsin's SAGE (Student Achievement Guarantee in Education) initiative has also yielded positive results. The state passed legislation in 1995 to phase in reductions in classes to 15 students in poor schools. During the 1996-97 school year, average test scores in smaller first grade classes increased 12 percent to 14 percent more than those of students in regular classes. The average gains were even greater for African American males, with total scores increasing by over 40 percent compared with similar students in regular classes.

While class size reductions cannot guarantee better student performance, both the Tennessee and Wisconsin experience along with other studies suggest that successful programs share key characteristics:

* Class size reduction should be concentrated in the primary years, particularly kindergarten through third grade. Tennessee students returning to regular classes as early as fourth grade maintained significantly higher achievement levels.

* Classes should be reduced to fewer than 20 students. Programs that reduce groups to below 20 have been found to be more effective than programs that retain more than 20 students but use teacher aides and other techniques to lower student-teacher ratios.

* Urban students, particularly minority pupils, benefit more than their peers from smaller classes. In Tennessee, inner city (predominantly minority) students also had significantly higher self-concept and third grade motivation scores than other inner city students.

* Class size reduction works best when coupled with professional development opportunities for teachers. Educators should be trained in new teaching techniques that take advantage of smaller class sizes.

Even if the research did demonstrate a clear link between class size and student performance, the question remains whether limiting class size is the smartest investment compared to other education reforms.

When asked about President Clinton's initiative, Chester Finn of the Hudson Institute think tank noted, "For $12 billion you could retrain today's teachers so they knew their subjects. You could give each of the nation's 2.7 million teachers a $1,000 tuition grant to go learn math or really effective techniques for teaching reading." A recent study using data from 60 previous research projects supports Finn's claim. Measuring the effect of a $500 investment per student, it was found that spending on teacher education had the greatest impact on student achievement. Lowering the student-teacher ratio was found to have a smaller effect than increasing teacher education, experience and salaries.

The study examined only the relationship with student performance. Lowering class size also has been linked to other benefits such as positive emotional development, better and more frequent communication between home and school, fewer special education referrals and decreased teacher absenteeism. However, when measured against the costs of limited school facilities and the difficulty of finding and hiring qualified teachers, support for reducing class size is far from overwhelming.

- Eric Hirsch, NCSL

Julie Davis Bell is an education expert at NCSL.