Quality. Equality. Equity. Each of these words generates debate. In seeking quality, educators continuously change, evaluate, and refine the educative process. In seeking equality, they attempt to provide equal educational treatments for all. Sometimes both efforts can result in practices that accentuate a gap in desired outcomes among groups of students. Equity, an attempt to close achievement gaps, then becomes an issue.
One current education idea—reduced class size—addresses all three concepts: When applied equally, it reduces the equity gap while simultaneously improving quality of outcomes. But reduced class size itself generates controversy. Although some people claim that class size is not important, data, logic, and common sense contradict their conclusion. Further, some critics of reducing class size base their objections on misinformation or misuse of terms. Other criticisms reflect a simple lack of will and creativity. The time has come to respond to detractors who would delay reducing class size for little or no substantive reason.
What the Data Tells Us
Much of the debate around class size emerges from the larger discussion of test scores. Reports about schooling outcomes continuously express concern for test-score gaps among various groups identifiable by race, gender, socioeconomic status, or other variables. The need to address such gaps generates seemingly endless special projects designed to help students who are identified as being "at-risk" of school failure as traditionally defined by low scores on standardized tests. Seeking ways to reduce achievement gaps drives much of what education is all about, even to the extent that American education is less a process than a disjointed series of projects, each cheered on by its own special interests. One task in the broader effort to reduce achievement gaps is to keep the floor level, at least in the early grades.
Tennessee's Project STAR (Student Teacher Achievement Ratio), a large, longitudinal project involving students in kindergarten through 3rd grade, has provided important information about class-size effects on pupil achievement and development. STAR began as an experiment (1985-1989) and continued as the Lasting Benefits Study (LBS) (1989-1997) and Project Challenge (1989-1997), which was adopted as a policy application of STAR's experimental results. Each school participating in STAR housed a small class of 13 to 17 pupils, as well as a regular class of 23 to 26 pupils, and a regular class with a full-time instructional aide. Pupils and teachers were assigned at random to the various classes. Since its inception, STAR has generated a longitudinal database of more than 11,000 pupils—more if the many related studies are included.
Project STAR showed clearly that small classes provided higher student outcomes and better student behaviors than either regular or regular-with-aide classes. Variables coded into the STAR database included pupil ethnicity, gender, and socioeconomic status as determined by free and reduced-lunch status. These entries allowed researchers to consider gaps related to these characteristics.
Figure 1 suggests that all students benefit from small class size and that nonwhite students benefit more than white students. When both white and nonwhite students began kindergarten in small classes, 87 percent of white and 86 percent of nonwhite 1st graders passed Tennessee's Basic Skills First criterion-referenced test. When they began kindergarten in regular classes, 84 percent of white and 72 percent of nonwhite children passed, opening a large gap that had the possibility of becoming the futile and expensive target of equity projects.
Figure 1—Percentage of 1st Graders Passing Basic Skills First Test
Using Class Size to Reduce the Equity Gap - Table
Class Type and Student Category | Attended Kindergarten | Did Not Attend Kindergarten |
|---|---|---|
| Small Class (13-17 students) | ||
| White | 87 | 79 |
| Nonwhite | 86 | 70 |
| Regular Class (23-26 students) | ||
| White | 84 | 70 |
| Nonwhite | 72 | 60 |
Whereas some test score results can reveal an equity gap, others can show equity. Analysis of the average scores on a Stanford Achievement Test of STAR students who passed from kindergarten to grade 1 and of those retained in kindergarten provides an example of the equity factor. The typical pupil retained in kindergarten is a minority male. In small STAR classes, the cutoff point for retaining students is a score of 422; in regular classes the cutoff point is 427. In other words, on average, students in small classes who have scores ranging from 423 to 426 are promoted; students in regular classes with those scores would be retained. The difference between passing and retention scores in small and regular classes reflects teachers' recognition that in small classes the student will succeed, and in regular classes the student likely will not succeed.
Based on STAR's results, policy leaders in at least a dozen states have enacted or are discussing class-size initiatives. Results from STAR and other studies, such as evaluations of Prime Time in Indiana (Mueller, Chase, and Walden 1988) and class-size reductions in Burke County (Achilles, Harman, and Egelson 1995) and Winston-Salem, North Carolina, have provided answers that beg new questions about the entire process of education in the United States today.
But the STAR outcomes should not have been a surprise. Robinson (1990) summarized the effects of class size: "The research [on class-size effects] rather consistently finds that students who are economically disadvantaged or from some ethnic minorities perform better... in smaller classes" (p. 85, emphasis added). "Fourth graders in smaller-than-average classes are about a half year ahead of 4th graders in larger-than-average classes," stated Wenglinsky (1997, p. 24), who also found differential small-class benefits: "The largest effects seem to be for poor students in high-cost areas." Looking at the first two years of STAR, we found thatfor all five measures, the advantage of being in a small class is greater for minority students than for whites... minorities in small classes outperformed their peers by an average of 16.7 points..., more than twice the effect size for whites (Finn and Achilles 1990, p. 567).
Small classes provide quality (higher scores), equality (pupils are assigned at random and every child gets a smaller class), and equity (those who usually do less well get greater benefits). Why, then, aren't small classes in widespread use, at least for all pupils in early primary grades?
Confusion Over Terms
Confusion over two terms—"class size" and "pupil-teacher ratio"—causes problems and misunderstandings in the debate over increasing student achievement in grades K–3 through focusing on size of classes. Class size and pupil-teacher ratio are not the same thing. Clarifying meaning and using the terms correctly will advance more widespread application of small classes in early primary grades.
Critics of public education often claim that added funds will not improve education quality. As evidence, they point out that over the years the pupil-teacher ratio has declined, but test scores have not risen proportionately. The critics have a good point. The pupil-teacher ratio has gone down. Most remedial and special projects are pullouts or add-ons for small numbers of students with one teacher (for example, Reading Recovery, Success for All, and, especially, Chapter 1). The plethora of special projects drives down the pupil-teacher ratio, and with few exceptions this plug-the-dike approach has limited success. For example, the $7.2 billion Chapter 1 program drew the following headline in Education Week (March 2, 1997, p. 1): "Chapter 1 Aid Failed to Close Learning Gap." Chapter 1 teachers do change a school's pupil-teacher ratio, but they do not reduce class size. In fact, average class size has increased. Many primary teachers attest that class size creeps upward as school districts seek ways to save money.
A study done more for economic than for education purposes made clear the distinction between pupil-teacher ratio and class size and their effects. Even here the equity gap was an issue:In order for school quality to explain the black-white gap in achievement, two relationships must hold: black students must attend schools of inferior quality, and school quality must matter for achievement (Boozer and Rouse 1995, p. 1).
The authors showed that "the correlation between pupil-teacher ratio and the average class size is relatively low" (p. 5), and they cited key pupil-teacher ratio versus class-size conclusions related to equity gaps:While the [pupil-teacher ratio] suggests no black-white differences in class size, measures of the school's average class size indicate that blacks are in larger classes (p. 2).Thus, according to the pupil-teacher ratio, blacks and hispanics [sic] do not attend schools of inferior quality. The conclusion reverses for the average class size (p. 7, emphasis added); students in schools with larger average class sizes have significantly smaller test score gains (p. 8).
Indeed, when taken together, the STAR findings, the Boozer and Rouse findings, the Chapter 1 evaluation results, and other findings provide a puzzling, provocative paradox: Projects undertaken to reduce the equity gap may actually increase it.
Responding to the Detractors
With research evidence as a guide, it seems logical that we begin at the outset to reduce the equity gap, instructing primary-grade children in classes of about 15 students—the average class size of the small classes in the STAR project. Classes of this size benefit all children, but particularly those who traditionally need some extra help.
Knowing what to do is one thing; actually doing it is another. Vocal detractors are abetted by some educators who either don't dare to change or who are just too complacent. Other educators are not willing to let pupils in someone else's class get off to a good start if their class size can't be reduced, too; they see reduced class size as an unfair benefit to primary-grade teachers. A response to these arguments should point out that teachers in classes of 15 students will do more work with parents, will work intensely with hard-to-teach pupils, and most likely will pass along students who are on or close to grade level, making teaching more rewarding for everyone.
To people who say that the lack of classrooms is a problem (and it is), a response is, "Think space and be creative with technology." Consider satellite K–1 spaces in neighborhoods that could be linked to your school through technology. Look for suitable space in libraries, YMCA/YWCA settings, and clubhouses. Define the problem as space, not classrooms.
To those who say that teachers need staff development before they teach small classes, we say that staff development probably would help, but the STAR results were achieved with no staff development in kindergarten and 1st grade, and the staff development in grade 2 provided no apparent advantages. Teachers can now use those teaching strategies they've always said they would like to use but have not had the opportunity to try.
Cost forms the basis of another argument in favor of small class size. Class-size reduction offers areas of potential savings. Grade retention is expensive, and small classes result in less grade retention. In addition, teachers in small classes quickly identify learning problems that go undiagnosed in regular classes. Lack of identification leads to costly special projects later on.
In the move to create appropriately sized classes for young children, educators may be their own worst enemies. They should all know the data: Class size matters. They should correct those who say that class size doesn't matter when what is really involved is pupil-teacher ratio. They should begin systemic change at the foundation of schooling; with small classes at the primary level, schools can reduce the myriad of remediation projects that later attempt to repair education failures.
Quality education builds on a quality start. The frantic search for equity starts with quality and equality, and all three can be addressed with smaller class size. Let's begin. Now!
