An Examination of Class Size Reduction

By David Boze, Policy Analyst, Evergreen Freedom Foundation

Introduction: The Controversy Over Class Size

As Aristotle once observed, "the fate of empires depends on the education of youth."It should be comforting, therefore, that voters consistently rank education as a top priority. Yet, despite innumerable reform experiments, public schools continue to be mired in the same problems today that troubled us twenty years ago. Of the solutions proposed to solve this dilemma, one of the most popular appears to be class size reduction.

According to the National Education Association (NEA), the nation’s largest teacher union, "excellence in the classroom can best be attained by small class size." ¹ These sentiments are echoed by numerous parents, legislators and community leaders who have been involved in class size reduction efforts. ² Lighter workloads for teachers, more one-on-one instructional time for students, fewer discipline problems and increased student learning are all benefits ascribed to reduced class size. Furthermore, class size reduction is a concrete, easily measurable and observable reform for which politicians can take immediate credit. Indeed, common sense would seem to support reducing class size as a means to improve student achievement. But does the available evidence support this assumption?

Many education groups, including the Department of Education, claim that research on class size reduction "points more and more clearly toward the beneficial effects of reducing class size" to improve student achievement.³ Other sources suggest an entirely different pattern of evidence. For example, according to the National Conference of State Legislators, more than 1,100 studies examine the relationship between class size and student achievement, yet no definitive conclusions have been reached.⁴ And Education Week has reported that deep divisions among scholars remain concerning class size reduction results.⁵

Other research has suggested that studies linking increased student achievement and class size reduction efforts have been unsound.⁶ Critics maintain that after isolating and evaluating co-existing influences, evidence suggests that most class size reduction efforts offer no significant increases in student learning unless the pupil-teacher ratio reaches tutorial levels.⁷

National Trend Data

As unbelievable as it seems on its face, national trend data supports the argument that class reduction efforts, by itself, will not improve student achievement.⁸ Class size trends from the ‘50s to the ‘80s provide no evidence of a significant relationship between student achievement and class size. Compare the 1950s national average teacher-pupil ratio of 30 to 1 with today’s average of about 19 to 1 (actual class size averages 23 students per class,⁹,¹⁰ —see the section entitled "The Federal Government, Washington State, and Class Size" for an explanation of the differences between pupil-per-teacher ratio and actual class size ). And from the 1970s to 1996 class sizes have fallen nearly 25 percent while Scholastic Aptitude Test (SAT) scores have tumbled during the same time period.¹¹ While SAT scores are not the only indicator of student performance, they are an important thermometer in determining the academic health of our brightest, college-bound youth. Smaller classes did not seem to help this segment of the student population.

Meanwhile, over the last three decades, public education spending has increased by at least 61 percent above inflation with much of those funds going toward the hiring of new teachers.¹² Yet few would argue that there has been a corresponding increase in student learning.

Advocates of class size reductions maintain that these trend data ignore negative social factors, such as increased numbers of single parents and the corresponding levels of poverty, which have changed the face of education over the years.¹³ Yet these critics fail to point out the many positive social factors gained during the past few decades, such as increased K-12 funding and higher parental education levels, which we are told are precursors to improved student test scores.¹⁴ (Data also indicate that no more than one-third of the drop in the pupil-teacher ratio can be attributed to the increase in special education populations who tend to require smaller classes and greater individualized attention).¹⁵

International Trend Data

Like the national data, international trend data suggest a further look is warranted before class size reduction measures are adopted as policy. Nations with far larger class sizes than those found in the U.S. are experiencing significantly higher student achievement. For example, the average academic scores of eighth graders in Korea and Japan are noticeably higher than those of U.S. students. Yet thirteen-year old students in Korea and Japan learn in average class sizes of 49 and 36 respectively.¹⁶

Actual class size averages in the U.S. and Japan are quite different due to the way schools in the two nations prioritize and organize academic disciplines and instructional practices.¹⁷ Teachers must instruct large classes (often exceeding forty) to ensure that schools can still provide students with art, music and computer instruction.¹⁸ We are not suggesting that American schools ought to adopt a Japanese or Korean instructional model. But the point must not be lost that the academic success enjoyed in countries where classrooms are brimming over with students suggests that the key factor for student achievement is something other than smaller classes.

Despite the evidence forwarded from national trend data and international comparisons, class size reduction advocates are not cautious or dissuaded. Many states have already invested heavily in reducing class size, with more states promising to follow. Examining the results of a few of the more prominent of these experiments helps provide perspective as to the wisdom of class size reduction policies.

Tennessee’s STAR Experiment

In 1985, Tennessee began a four-year study of the effects of class size reduction on kindergarten, first, second and third grades. In relative terms, Project STAR (Student-Teacher Achievement Ratio) was to be a controlled experiment designed to provide reliable, scientific evidence on the effects of class size reduction. The project was inspired by earlier research indicating student achievement was roughly constant until class size was reduced to 15 or fewer pupils per teacher.¹⁹

More than 300 classrooms in 79 schools totaling 7000 students participated in the program. Assignment of students and teachers to classrooms was random. STAR compared smaller classes (13-17 students) with normal-sized, larger classes (22 to 26 students) both with and without an instructional aide in the larger classes. To control for possible differences in school quality, every participating school had at least one of each type of class (a class of 13 to 17 students, a class of 22 to 26 students without an instructional aide, and a class of 22 to 26 students with an instructional aide).²⁰

The results of the STAR experiment led many class size reduction advocates to conclude that STAR had "definitively" proved that "bringing class size down in the primary grades in and of itself has positive effects on student achievement in all subject areas."²¹ Initial results revealed that students from smaller classes outperformed students from larger classes and that smaller classes had significantly less student retention and improved assessment of special education needs.²²

Performance of students participating in STAR was followed for several years after the program. Test scores indicated that students from smaller classes significantly outperformed students from the larger classes in all academic subjects. These results led Tennessee to implement class size reduction in 16 of the state’s poorest schools in a program called "Project Challenge." These schools experienced an increase in their ranking compared to other school districts on statewide student achievement tests.²³ Unlike Project STAR, however, Project Challenge was not a controlled experiment.

Despite these seemingly clear cut results, researchers remain divided on the effectiveness of comprehensive class size reduction, especially since a closer examination of the STAR experiment suggests that its results are not as conclusive as is commonly believed.²⁴

A review of the available STAR data, which generally has not been available to researchers,²⁵ was conducted by Eric A. Hanushek, Professor of Economics, University of Rochester. Professor Hanushek observed that many advocates of class size reduction report the STAR project showed that small classes led to improved student achievement in each grade. The data, however, actually showed no significant cumulative achievement in students who remained in small classes, leaving researchers to wonder why learning gains did not accumulate from year to year.²⁶

Gains were made by students who began their initial grade (kindergarten or first grade) in a small class. Evidence obtained from the STAR experiment and its follow up studies suggests that, although these first-year, one-time gains remained with these students, no increase in subsequent years occurred. This was true even when students remained in small classes.²⁷ If class size reduction changed the aggregate rate of student learning, students remaining in small classes would advance above their counterparts each year.²⁸ This first-year gain in student achievement may reflect a one-time acquisition of social and learning behaviors useful to students in subsequent years.²⁹

Professor Hanushek cited numerous potential problems with the experiment (such as the reassignment of some students due to parental pressures ³⁰) and warned against concluding too much based on the STAR experience. He further observed that the STAR study has never been replicated and remarked that, although such experiments [class size reduction] are expensive, they are far less expensive as a demonstration project than the potential costs of an across-the-board policy of class size reduction, especially if the sweeping policy change is based on faulty or incomplete data.³¹

Contrary to those who believe that the STAR project conclusively proved that comprehensive class size reduction efforts will improve academic achievement, Professor Hanushek concludes that evidence obtained from the STAR project supports only targeted class size reduction, and that comprehensive class size reform would likely be an inefficient use of scarce resources.

Professor Hanushek’s concerns have been echoed by other researchers as well, including Stanford University professor Mike Kirst, who co-directs the Policy Analysis Center for Education (PACE). According to Professor Kirst, "There is certainly no research to justify reducing all grades."³²

Professor Hanushek’s interpretation of the STAR data can be summarized as follows:

The most expansive conclusion that can be reached from Project STAR and the Lasting Benefits Study is that they might support an expectation of positive achievement effects from moving toward small kindergartens, and maybe small first grades. None of the STAR data support a wholesale reduction of class sizes across grades in schools. The achievement results also come from large reductions (one-third of the existing regular class sizes) that take the small classes to quite low levels compared to most existing situations (15 students per class). It does not provide evidence about what might happen with smaller changes that take class sizes down from the current levels to levels above the Tennessee experiment, say, 18-20 students. (Remember that the original motivation for Project STAR involved research results suggesting no effects for class sizes above 15 to 1). ³³

Other State Experiments in Class Size Reduction

Tennessee’s STAR experiment is the most significant in terms of class size reduction, particularly since it was designed with some controls to measure the effects of class size reduction as an independent variable. But the experiences of other states are also instructive. Wisconsin and California, are currently experimenting with reduced class size.

Wisconsin’s class size reduction program, called the Student Achievement Guarantee in Education (SAGE), was enacted and financed as part of Wisconsin’s 1995-97 state budget.³⁴ The objective for Wisconsin’s program was to target class size reduction in kindergarten through third grade in schools serving low-income populations. The program was intended to reduce class sizes in targeted schools to a student/teacher ratio of 15 to 1 or less. Unlike STAR, however, SAGE was not designed to directly compare classes targeted by the program with control classes not involved in the reforms. In fact, SAGE reforms went beyond class size reduction.

Under the SAGE program, reductions in class size were accompanied by additional reforms including a revised, rigorous academic curriculum, professional development, and accountability initiatives.³⁵ Schools participating in the SAGE program are also "required to embrace the "lighted schoolhouse" concept" which requires that schools provide family, student, and community events from very early in the morning until late in the evening.³⁶

Preliminary reports in the December 9, 1997 Milwaukee State Journal suggest that this approach was successful in raising test scores, particularly for minority students. But the degree to which class size reduction alone is to be credited with these improvements remains unknown, ³⁷ and because SAGE reforms targeted specific populations, results of the reform may not be generally applicable. Thus, when SAGE is used as an example of the positive effects of class size, it is important to remember that the reduction of class size under the program was only one of several significant alterations. The effect of the class size reform (or any of the other reforms) has not been measured as an isolated variable. As a result, one cannot assume that any increases in student learning are due to class size reduction alone—or that these increase would not have been achieved without reducing class size.

California has also taken a lead in state-wide class size reduction efforts. In 1996, inspired by the reported success of Tennessee’s STAR report, California lawmakers set up a program to implement class size reduction in grades K-3 throughout the state. California’s Class Size Reduction Program mandates that top priority be given to first grade, followed by second grade and then third grade or kindergarten.³⁸ In its third year of implementation, $1.5 billion has been appropriated to California’s Superintendent of Public Instruction in order to provide incentive money for schools to implement ratios of twenty students to one teacher.³⁹

The California experiment extends well beyond any state’s previous experimentation with class size reduction. The scale of the experiment dwarfs the STAR program and is based on assumptions that lay outside the realm of the available STAR data.

For example, STAR data suggested increases in student learning at the earliest grade, but California’s program prioritizes kindergarten after first, second and third grades. Also, Tennessee’s STAR program, largely carried out by experienced teachers, did not create teacher shortages. On the other hand, California’s plan (requiring extra classrooms with a teacher in each one) has created a severe teacher shortage resulting in the hiring of many inexperienced, new teachers. This practice is likely to negatively impact the possibilities for increased student achievement.

Another significant difference: Small classes for the STAR experiment were approximately 15 students per teacher. California’s "small" classes of twenty pupils per teacher better resemble the large classes of the STAR experiment. So, though STAR provided no evidence that a reduction to twenty students per teacher would improve student learning, California has set twenty students per class as its target size.

California appears to be gambling its educational resources on unfounded extrapolations of already optimistic conclusions on class size reduction. Since California began making massive expenditures on class size, it has been reported that test scores have improved slightly, but whether or not reduced class size is the cause of these modest gains remains unknown.⁴⁰ Some researchers contend that California could have improved student achievement using other methods for a fraction of the cost.⁴¹

Class size reduction experiments in other states have yielded inconclusive and/or unsupportive evidence as well. For example, the Nevada Department of Education’s "1993 Class Size Evaluation Study" determined that "achievement levels remained about the same when small classes were compared with larger classes."⁴² During the 1980s, in Austin, Texas, sixteen schools were given $300,000 a year for five years which was used to reduce class size. After five years, fourteen of the schools had failed to improve student test scores or attendance, while two schools posted significant improvements in student test scores. The two schools that improved student achievement implemented a number of reforms in addition to a reduction in class size, including intensive teacher training, and rigorous academic standards. Whether or not reduced class size was a significant factor in the improved student performance in the two successful schools is uncertain.⁴³

The uncertainty of class size reduction results coupled with the modest gains attributed (justly or unjustly) to these efforts leave many questions unanswered. For example, even if one assumes that reducing class size improves student achievement, do the benefits justify the costs? Also consider that the STAR experiment reduced classes to an average of fifteen pupils per class, a rate lower than many policymakers are proposing. Still, many researchers considered the actual improvements in test scores to be questionable or modest. ⁴⁴

The Costs of Class Size Reduction

Class size reduction faces numerous obstacles to its implementation. Although the concept (having fewer students per teacher) is simple to understand relative to other education reform possibilities, implementation of class size reduction is far more complex than a mere allocation of dollars.

After evaluating effectiveness, the first factor to consider regarding class size reduction efforts is cost—but monetary costs cannot be considered in a vacuum. The costs of class size reduction must be compared to its benefits. The opportunity costs of choosing class size reduction over other education reforms must also be examined. For example, one must consider if the resources spent on class size reduction would be better spent on another type of reform. The possibility that other reforms might have a greater positive effect on student achievement for the same or fewer resources demonstrates the need for reliable research on education reform options. Investments of resources must be made carefully, not only for the sake of the taxpayers who foot the bill, but for the students whose educational opportunities are at stake. Additional investment alone does not always improve student learning or achievement.⁴⁵

Class size reduction is among the most expensive education reform options, and costs must be calculated using a number of factors, including:

Initial average class size. The larger the drop to "small" the greater the cost.

Imposition of a rigid cap, or flexibility in the number of students per teacher. A rigid cap will increase the cost by decreasing the final average class size. Schools will keep numbers lower than necessary to ensure staying below the cap. (For example, if schools only receive class size reduction funding for staying below 17 students per class, it is likely that schools will remain at 15 or 16 per class in case of a student transfer. Therefore, rigid caps are likely to increase the costs of class size reduction by creating a need for classes to be even smaller than mandated.)

The cost of teachers hired for class size reduction. This depends on the experience level of the teachers hired. Teacher costs will increase with time as teachers move up the salary ladder, but experienced teachers are valuable. (The costs of teacher support may also need to be factored in.)

The cost of providing new classrooms.

Added operational costs, such as costs for utilities, custodial and clerical services.

Potential cost offsets, e.g., due to less grade retention.⁴⁶ (If fewer students are held back or sent to special education services, taxpayers would save on the extra costs of those students.)

Method by which class size is implemented (year-round schooling, block scheduling, etc.).⁴⁷

The extreme financial commitment inherent in class size reduction can be readily observed through the California example where legislation has mandated that class size be reduced below twenty pupils per teacher. The actual per-pupil costs vary by district from $0 to $1000 excluding facilities and staff development costs.⁴⁸ Since 1996 California has allocated more than $4 billion to class size reduction.⁴⁹ Continuing costs for the program start at $1-1.5 billion every year with costs steadily rising as teachers’ salaries climb due to experience.⁵⁰ Also note that California’s twenty students per pupil target is approximately the same size as the large classes in the STAR experiment. Were California or any other state to attempt to achieve the levels of the small classes in STAR, the financial costs would skyrocket. Yet the monetary price tag of class size reduction is not the only cost to consider.

Opportunity Costs

The opportunity costs of class size reduction are immense and must be considered. Before committing to reductions, policymakers should consider if the required money spent might have greater impact if it were invested in better curricula; rigorous academic standards; teacher training; or technology. Money spent on class size reduction cannot be recovered for other areas of educational reform that may be better suited toward improving student learning.

Chester Finn, a former Assistant Secretary for Research and Improvement for the U.S. Department of Education, recently commented on the opportunity costs of President Clinton’s $12 billion proposal to reduce class size nationwide. Finn remarked that the $12 billion proposal could be used to train teachers lacking in specialized knowledge of their subjects, or to provide each of the nation’s 2.7 million teachers a $1000 tuition grant to learn effective techniques for teaching reading or math.⁵¹ Putting aside the question of the propriety of such a federal expenditure, Mr. Finn’s observations illustrate well the opportunity costs. Finn’s comments become even more relevant when one considers that teacher quality is a more important factor in increasing student learning than is class size.⁵²

Related Implementation Obstacles to Class Size Reduction

As already noted, cost alone is not the only problem facing class size reduction. Other implementation obstacles loom large, and policymakers should consider these issues before any decision is made to implement class size reduction.

Class size reduction will lead to the need for more classes, which will in turn increase the need to hire more teachers. However, as one superintendent of a school district near Houston remarked, "The fact that there’s more money [for class size reduction] doesn’t mean there are more teachers."⁵³

Already some states are finding it difficult to obtain qualified, knowledgeable teachers. In Massachusetts, for example, when new teachers took basic academic tests, a majority failed to pass. Policymakers, already struggling to find ways to ensure that teachers are qualified, will face shortages as classroom sizes are reduced. Unless ways are found to find and hire qualified individuals, these shortages can lead to the increased hiring of unqualified teachers. This is especially troubling considering that teacher quality, not class size, is the most important factor in improving student achievement.⁵⁴

Adding more classes and teachers mean more facilities. Portables, new schools, and/or adding rooms to existing facilities require additional resources. These facilities also increase utility and maintenance costs. Furthermore, these additional costs do not affect all schools equally. Some schools may have additional facilities readily available. Some may already have small classes. On the other hand, some schools may experience extreme difficulty obtaining additional facilities (particularly in urban areas).⁵⁵

A number of other issues, such as the priority in the type of classes to be reduced in size; whether or not team teaching should count as a reduced class; the varying abilities of teachers; etc., combine to make class size reduction a much more complex issue than it may appear at first glance. The costs and difficulties in implementation and the here-to-date disappointing results appear to make comprehensive class size reduction a weak strategy for meaningful education reform.

Nevertheless, class size reduction remains politically popular.⁵⁶ In fact, the reform’s popularity may actually hinder its potential for effectiveness.⁵⁷ When targeted class size reform has been implemented, teachers of all grades may become envious (understandably) of the smaller class sizes and lighter work load of their colleagues in the early grades. And because of the perception that smaller classes are necessary for increased student learning, parents will want smaller class sizes so their children (of all ages) possess the same advantages (real or perceived) as other children. Targeted reductions, therefore, will be politically difficult to implement.⁵⁸ Still, policymakers may be able to avoid the excessive costs and ultimate disappointment of comprehensive class size reduction by targeting student populations who appear to be most positively affected by class size reduction.

Would Class Size Reduction Help Anybody?

To summarize what we have already learned, evidence that reductions in class size have a positive effect is strongest in the primary grades.⁵⁹ Data collected from the Tennessee STAR experiment suggests that this benefit is achieved in the initial small kindergarten or first grade class a child attends. The evidence also indicates that, because benefits to reduced class size have not proven to be cumulative, providing additional grades (above the first grade) with reduced class size may have no significant positive affect on student learning.⁶¹ In other words, the assumption of the STAR experiment was that a reduction in class size of any grade would change the rate of student learning. If this were true, then every year a student is in a smaller class, he or she should get farther ahead.⁶² Experiments to date, including the STAR project indicate that learning gains, do not build cumulatively from year to year.

Effectiveness may also depend on the extent of the drop in class size. STAR classes dropped by about a third to an average size of 15 students per pupil and registered modest gains, while some class size reduction proposals offer reductions of only a student or two, with average class sizes remaining significantly larger than 15. Minor reductions in class size are likely to be an ineffective and expensive means of improving student achievement.⁶³

Aside from the first couple of grades, certain student populations seem to be more affected by reduced class size than others. Disadvantaged students and minorities seem to be more sensitive to class size variation than other students.⁶⁴ (This does not necessarily mean that these students would not be better served by another reform; say, increased teacher quality.⁶⁵)

The Federal Government, Washington State, and Class Size

Although specifically targeted class size reduction shows some promise, it is comprehensive class size reduction that has captured the attention of policymakers. Resolutions from both of the nation’s leading teacher unions strongly advocate overall class size reduction. President Clinton has succeeded in getting Congress to agree on a $1.2 billion "down payment" toward his initiative to hire 100,000 new teachers in order to reduce class size. This "down payment" of $1.2 billion is a fraction of his seven-year, $12 billion plan to get the nation’s class average down to 18 students to 1 teacher.⁶⁶

Currently, Washington state law provides schools with funds to provide, at a minimum, one teacher for every 20.4 students and, at a maximum, one teacher for every 18.42 students. Washington’s actual pupil per teacher ratio is among the highest in the nation at 20.32 pupils-per-teacher.⁶⁷ But it should also be noted that our state’s pupil-per-teacher ratio has been steadily falling. In 1976 Washington our average was 23.21 pupils per teacher.

Of further interest is the fact that the pupils per total instructional staff ratio is dramatically different from the pupil per teacher ratio. Though Washington has about 20 pupils per teacher, we calculate that there are 16 students per certified staff member. And the American Legislative Exchange Council’s "Report Card on American Education," reports that National Education Association statistics indicate that Washington state has only 8.55 pupils per instructional staff employee which includes teachers, principals, supervisors, and other non-supervisory instructional staff.⁶⁸

Obviously, if Washington can have an official pupil-to-teacher ratio of 20.32 pupils per teacher while, according to the Nation Education Association, the actual average class size* is more than 28 pupils per teacher,⁶⁹ pupil-teacher ratio is not necessarily the same as actual class size. That pupil-to-teacher ratios do not always accurately reflect class size is due to the accounting formula used for non-teaching certified staff (such as librarians and counselors). Another reason may be found in class scheduling. For example, if a school has a 22-to-1 pupil-to-teacher ratio, the school may be able to hold class size down to 15 students in reading classes by scheduling larger classes in a less intensive discipline. Further illustration of the point may help clarify the issue.

Under current state law, an elementary school of 300 students would be provided a minimum of 15 certified instructional staff. If every teacher was an in-class instructor, the school’s actual class size would be twenty students per class. But if the school used the funds to provide two out-of-class instructors, say a librarian and a counselor, the teacher-to-student ratio would technically remain 1:20 but the actual class size would be 23 students per in-class instructor (assuming all classes are of equal size). Regardless, no significant improvement in student learning occurs until class sizes are reduced far below this level—which could require nearly doubling the resources currently allotted to K-12 public education.

Although pupil-to-teacher ratios and actual class size rankings receive a great deal of political attention, the real value of these measures should be weighed against their impact on student achievement. This effect, according to the evidence explored here, is, in most circumstances, likely quite minimal.

While the Washington Education Association, numerous elected officials, and many parents have expressed discontent with Washington’s comparative ranking among states in the pupil-per-teacher ratio, the rank actually provides yet another illustration of the elusiveness of the true effect of class size on student achievement. If Washington state’s 48th in-the-nation class size ranking was the determining factor in student achievement, one would expect Washington’s student achievement ranking to be among the worst in the nation as well. Washington, however, ranked 14th in the average mathematics score in the 8th grade National Assessment of Educational Progress (NAEP) and 2nd among the 22 states using SAT scores as a primary college entrance examination.⁷⁰ The fact that state pupil-teacher ratios and student achievement measures do not coincide suggests that factors other than class size may be better determinates of student success.**

Recently, the President of the Washington Education Association (WEA) announced that the WEA planned to push for smaller class sizes. And while President Clinton’s new federal initiative will provide temporary dollars for the hiring of approximately five hundred additional teachers in our state, Governor Locke’s 1999-01 budget proposal sought to provide enough matching funds to double that number.*** When the federal funding expires however, Washington (and other states) will be left in the unenviable position of either firing teachers or finding more money to keep them. These facts suggest that the debate over the appropriate class size in Washington state is unlikely to subside anytime soon.

If Not Class Size Reduction, What?

We cannot ignore the evidence that class size is not necessarily a determining factor in student achievement⁷¹ and that other reforms, such as new curricula, teaching methods, or teacher development may have greater value.⁷²

We already know that among the factors more strongly and consistently linked by evidence to student achievement is teacher quality. ⁷³ Teachers who are more experienced, more knowledgeable in their subject matter and have high quality professional training are the chief remedy for underachieving students.

Basic incentives and political realities, however, make implementing meaningful improvements in teacher quality a formidable political task. Any attempt at tying teacher pay to teacher quality will be opposed by the National Education Association, the resolutions of which specifically oppose any form of merit pay. Mediocre teachers would consider competency tests threatening to their job security. And talented teachers may not gain materially from increased expectations, but could experience weaker job security as the result of such measures. As a result, teachers as a group will have little incentive to approve of plans to raise standards for teacher performance, to test teachers for competency, or to support measures to increase individual teacher accountability. With class size reduction offering no such disincentives as well as the advantages of a reduced workload, it is likely to remain a much more attractive reform for teachers.

Policymakers need to ask themselves how they can make the best, most effective, use of limited resources. Numerous reform options are available. It is important to remember that schools will vary in strengths and weaknesses, so a one-size-fits-all strategy is likely to be inefficient. Decisions are best made locally, school by school, class by class, teacher by teacher.

Conclusion

Concern about the condition of our public school system has not dissipated over the last decade and a half. Parents and policymakers remain desperate to improve public education and, despite the expense and limited success of class size reduction, it remains a popular solution.

So why does class size reduction remain a popular education reform? Parents like class size reduction because they believe it will allow more one-on-one instruction for their children. Taxpayers in general like it because smaller classes are credited with increasing student achievement. For teachers, class size reduction decreases their workload and has the potential to reduce classroom discipline problems. For politicians, it offers a popular, simple, easily measured (and observed) education reform idea for which they can take credit. In short, class size reduction appears to have something for everyone—except students. It is here that the oft-promised benefits of class size reduction fail to materialize. It is the among the most expensive and least effective education reform investments and, as with any other investment, we should not invest in something that does not offer a reasonable return.

The primary harm inherent in comprehensive class size reduction efforts is in the opportunity cost of the expended taxpayer funds and the psychic energy of the individuals involved. Because the results of class size reduction efforts are negligible, the considerable resources (human and financial) are mostly wasted. More worthy, but more controversial efforts, such as improving teacher quality, are not pursued. In the end, the political popularity of an issue should not be prioritized above the needs of students.

For some students, class size reduction may offer increased potential of achievement. It may also offer teachers more time and less stress. But as a comprehensive attempt at increasing student achievement, it has little, if any, positive impact.

Class size reduction may be an easy reform for which to take political credit, and it is likely to remain popular with the teachers unions, administrators, and parents.⁷⁴ But our education system desperately needs improvement, and while class size reduction has proved expensive, it has not proved effective. Taxpayers are too heavily burdened for the government to make inefficient use of the massive additional resources that would be required for large-scale class size reduction. If children are to be well-educated, we need to focus less on what special interests may want, and focus more on what works for students.

NOTES

1. National Education Association Resolution B-6 1998-99.

2. United States, Department of Education, Reducing Class Size: What Do We Know? SAI 98-3027 (May 1998): 6 (http://www.ed.gov/pubs/ReducingClass/research.html)

3. United States, Department of Education, Reducing 1.

4. National Conference of State legislators, Class Size Reduction, (updated April 1, 1998) 1.

5. West Ed Policy Support Program, "Class Size Update Page" (February-March 1998) quoted Education Week Article (Viadero Feb. 18,1998).

6. Andrew J. Coulson, Education Policy Analysis, Human Life, Human Organizations and Education Vol. 2 No. 9 ISSN 1068-2341 (June 3, 1994) 22.

7. Andrew J. Coulson, Human Life and Andrew J. Coulson, Education Policy Analysis Archieves, The Center for Applied Philosophy, A Response to John Covaleskie Vol. 2 No 12 ISSn-1068-2341 (August 10, 1994) 5.

8. U.S. Sept of Ed. Reducing 2.

9. National Conference of State Legislators, Class Size Reduction (April 1, 1998) 1.

10. Julie Davis Bell, "Smaller= Better?" State Legislatures, National Conference of State Legislatures, June 1998: 1.

11. "Do More Teachers Mean Better Education?" Investor’s Business Daily Sept. 30, 1998 National Center for Policy Analysis (website)

12. Thomas Toch and Betsy Streisand, "Does Class Size Matter?" U.S. News and World Report October 13, 1997 (www.middleweb.com/Class Size.html)

13. "Do More Teachers Mean Better Education?" Investor’s Business Daily Sept. 30, 1998 National Center for Policy Analysis

14. Eric A. Hanushek, The Evidence on Class Size, Occasional Paper 98-1, W. Allen wallis Institute of Political Economy, University of Rochester, February 1998: 10.

15. Eric A. Hanushek, The Evidence on Class Size, 16.

16. National Conference of State legislators "Class Size Reducation" updated April 1, 1998 p.1

17. Eric A. Hanushek, The Evidence on Class Size, 19.

18. John Gittelsohn, "Lessons from Japan," Orange County Register April 26, 1998, as found in West Ed Policy Support Program "Class Size Update."

19. Eric A. Hanushek, The Evidence on Class Size, 26.

20. United States, Department of Education, Reducing Class Size: What Do We Know? 3-4

21. West Ed Policy Brief, Class Size Reduction Lessons Learned from Experience, No. 23 (August 1998) 2.

22. United States, Department of Education, Reducing Class Size: What Do We Know? 4.

23. United States, Department of Education, Reducing Class Size: What Do We Know? 6-7

24. Debra Viadero, "Small Classes: Popular, But Still Unproven," Education Week Febuary 18, 1998 (www.edweek.org/ew/vol-17/23class.h17).

25. Eric A. Hanushek, The Evidence on Class Size, 27.

26. Debra Viadero, "Small Classes," Education Week 2.

27. Eric A. Hanushek, The Evidence on Class Size, 27-29

28. Debra Viadero, "Small Classes," Education Week 2

29. Eric A. Hanushek, The Evidence on Class Size, 30.

30. Eric A. Hanushek, The Evidence on Class Size, 31.

31. Eric A. Hanushek, The Evidence on Class Size, 32.

32. Julie Davis Bell, "Smaller=Better?" State Legislatures Magazine, 2.

33. Eric A. Hanushek, The Evidence on Class Size, 30.

34. State of Wisconsin, Department of Public Instruction, Governor signs SAGE initiative into law!, SAGE Newsletter.

35. United States, Department of Education, Reducing Class Size: What Do We Know?" 5-7.

36. Joseph H. Quick, "The Wisdom of Investing in Class Size Reduction," Wisconsin School News (reprinted by Madison School District, Wisconsin, www.madison.k12.wi.us/sage.htm) Jan. 1998: 1.

37. United Sates, Department of Education, Reducing Class Size: What Do We Know?" 7.

38. David Boze, personal interview, California Department of Education.

39. David Boze, personal interview, California Department of Education.

40. Ken Hoover, "State Data hint Smaller Classes Are Effective, Modest improvement noted on test scores," San Francisco Chronicle 29 December, 1998 (www.sfgate.com/cgi-bin/article.cgi?file=/chronicle/archive/1998/12/29/MN6622.DTL).

41. Thomas Toch and Betsy Streisand, "Does Class Size Matter?" U.S. News and World Report 13 October 1997 (www.middleweb.com/Class Size.html).

42. Erica Olsen, Class size reduction is not the answer to Nevada’s failing education system, Nevada Policy Research Institute, 3 February 1997 (www.npri.org/issues/class_size.htm).

43. Thomas Toch and Betsy Streisand, "Class Size?," U.S News and World Report 5. AND Debra Viadero, "Small Classes: Popular, But still Unproven," Education Week 2-3.

44. Thomas Toch and Betsy Streisand, "Class Size?" U.S. News and World Report 5.

45. Andrew J. Coulson, Human Life, 14-16.

46. "Class Size Reduction: Lessons Learned from Experience," West Ed Policy Brief (No. 23) August 1998: 4-5.

47. "Class Size" Education Week on the Web (Www.edweek.org/context/topics/class.htm)

48. "Class Size Reduction: Lessons Learned," West Ed Policy Brief 5.

49. Ken Hoover, "State Data Hint Smaller Classes Are Effective, Modest Improvement Noted on Test Scores," San Francisco Chronicle 29 December 1998 (www.sfgate.com/cgi-bin/article.cgi?file=/chronicle/archive/1998/12/29/MN6622.DTL) 2.

50. Julian Guthrie, "Teachers: One clas size fits all," San Francisco Examiner, 8 March 1998: 1

    (www.sfgate.com/cgi-bin/article.cgi?file=/examiner/archive/1998/03/08/NEWS5653.dtl) and Thomas Toch and Betsy Streisand, "Class Size?" U.S. News and World Report 2.

51. National Conference of State Legislators, Class Size Reduction 1 April 1998: 2.

52. National Conference of State Legislators, Class Size Reduction 2-3.

53. Joetta Sack, "Critics Doubt Teacher Plan’s Effectiveness," Education Week 28 October 1998.

54. Jay Mathews and Valerie Strauss, "Should Classes Be Smaller?," Washington Post, Dec. 1997: 1 (www.middleweb.com?WPCLSz.html). And S. Paul Wright, Sandra P. Horn, and William L. Sanders, "Teacher and Classroom Context Effects on Student Achievement: Implications for Teacher Evaluation," Journal of Personnel Evaluation in Edication, Volume 11 (1997) 61.

55. "Class Size Reduction: Lessons," West Ed Policy Brief 8.

56. Thomas Toch and Betsy Streisand,"Class Size?," U.S. News and World Report 3. And "Class Size Research" (www.nashville.net/~heros/class_size.htm) February 1999:3.

57. Thomas Toch and Betsy Streisand, "Class Size?," U.S. News and World Report 6.

58. Julian Guthrie, "Teachers," San Francisco Examiner, 1.

59. United States, Department of Education, Reducing Class Size, 2.

60. Eric A. Hanushek, The Evidence on Class Size, 27.

61. Eric A. Hanushek, The Evidence on Class Size, 27.

62. Debra Viaders, "Small Classes," Education Week

63. Thomas Toch and Betsy Streisand, "Class Size?," U.S. News and World Report 5.

64. United States, Department of Education, Reducing Class Size: What Do We Know?, and Thomas Toch and Betsy Streisand, "Class Size," U.S. News and World Report 4.

65. S. Paul Wright, Sandra P. Horn, and William L. Sanders, "Teachers and Classroom Context Effects on Student Achievement: Implications for Teacher Evaluation," Journal of Personnel Evaluation in Education, Volume 11 (1997) 57-62.

66. Joetta Sack, "Critics Doubt Teacher Plan’s Effectiveness," Education Week 28 October 1998.

67. John S. Barry and Rea S. Hederman Jr., Report Card on American Education: A State-by State Analysis 1976-1998, American Legislative Exchange Council (December 1998) 39.

68. John S. Barry and Rea S. Henderman Jr., Report Card: 1976-1998, 37, 39-41.

69. Thomas Toch and Betsy Streisand, "Class Size?," U.S. News and World Report 4.

70. Lee MacVaugh with Taylor Nguyen and James Ahn, Report Card on American Education 1996, American Legislatvie Exchange Council (October 1997) 4-7.

71. S. Paul Wright, Sandra P. Horn, and William L. Sanders, "Teacher and Classroom Context Effects on Student Achievement" Journal of Personnel Evaluation in Education, 62. And Eric A. Hanushek, The Evidence on Class Size, and David J. Hoff, "Federal Class-Size Reports Do an About-Face" Teacher Magazine on the Web, 10 June 1998: 2. And Julie Davis Bell, "Smaller=Better?," State Legislatures, 2-3. And "Class Size Research" February 1999: 3.

72. West Ed Policy Brief, Class Size Reduction: Lessons Learned from Experience, No. 23 August 1998: 7

73. S. Paul Wright, Sandra P. Horn, and William L. Sanders, "Teacher and Classroom Context Effects on Student Achievement," Journal of Personnel Evaluation in Education, 57-67. And United States, Department of Education, Reducing Class Size: What Do We Know?, 3.

74. Julie Davis Bell, "Smaller=Better?," State Legislatures Magazine, 1.

    
    

    Send this Page to a Friend
    

    Evergreen Freedom Foundation P.O. Box 552, Olympia, WA 98507 Phone: (360) 956-3482, Fax: (360) 352-1874 Email: effwa@effwa.org