clear Research: Impact of Class Size on Pupil Performance

District 64 conducted a thorough review of class size policy, including an extensive literature review, in 1995. A Class Size Committee of community members, teachers, and administrators was appointed by the Board of Education to help set “a class size policy which will promote excellence in education throughout the District” and “investigate and research class size issues” (Class Size Committee Report, Final – June 26, 1995). Successive Boards have referred to the committee’s work as class size discussions occur.

Since the committee’s final report, additional studies and analyses investigating the impact of class size on student achievement have been completed. Here is a recap of the academic literature over the last 25 years to provide some perspective based on “real world” studies.

Early Academic Studies
Several hundred academic studies attempting to analyze and document the impact of class size on pupil performance have been conducted in the United States and overseas. Unfortunately, policy makers, including school boards across the country, have been presented with a variety of conflicting findings and conclusions.

One problem is that many of these studies assume that pupil-teacher ratios were an adequate proxy for class sizes. In general, researchers calculate pupil-teacher ratios by dividing the total students in a school district by total teachers. Given the number of teaching professionals employed by school districts without primary classroom responsibilities, a wide discrepancy exists between calculated pupil-teacher ratios and actual classroom sizes per primary teacher. As a result, these studies are thought at best to be directionally correct, but more likely could be misleading.

The STAR Project
To eliminate the problems in earlier studies, several analysis projects were designed to evaluate performance on a class-by-class level. The leading, and most often cited, study of classroom size effects is the STAR Project. (A pdf of the Final Summary Report 1985 - 1990 is available here.)

The Student Teacher Achievement Ratio (STAR) project was a large-scale, four-year experimental study of reduced class sizes, mandated and funded by the state of Tennessee in 1985-1989. No other study of this magnitude has been reported since then. Over 7,000 students in 79 schools were randomly assigned to one of three groups: Small Class (13 to 17 students per teacher), Regular Class (22 to 25 students per teacher), and Regular Class With an Aide (22 to 25 students per teacher with a full-time teacher’s aide).

Pupil reading and math test performance of the Small Class group exceeded the other groups by a statistically significant margin. There were no differences between Regular Class and Regular Class With an Aide groups.

The primary investigators concluded “this research leaves no doubt that small classes have an advantage over larger classes in reading and math in the early grades.”

The same team analyzed this set of students as they progressed through 12th grade. The team concluded that performance benefits persisted: “students placed in small class sizes in grades K-3 have better high school graduation rates, higher grade point averages, and are more inclined to pursue higher education.”

The study drew an implication of particular interest at the national level: “Inner-City (Predominantly Minority) students in small classes always outscored inner-city students in regular and regular/aide classes. This suggests that small classes are very beneficial to minority students.”

Project STAR and related studies are the intellectual rationale behind efforts to reduce classroom sizes in California to 20 pupils and for similar initiatives in Nevada, North Carolina, Texas, and Utah.

Critics note three sets of concerns that call the findings and conclusions of Project STAR and similar studies into question.

Design Flaws: No pre-test information was collected for pupils before placement, the schools were not randomly selected, it is not clear if teachers were truly randomly selected, and teacher expectations may have affected the outcomes (the Hawthorne effect). Therefore, whether the performance effects were as strong as indicated can be questioned.

In addition, the numbers of students moving between Small Classes and Regular Classes was not random, suggesting parental pressure to have children placed in smaller classes. This adds the variable of “parental involvement” into the mix.

Finally, there was a large attrition of students in year-to-year movements. The students who left tended to have lower test scores, driving up the averages for those remaining.

Apparently Limited Impact:  Classroom size as a primary explanation for performance differences is also in question.  Consider Kindergarten, where the effect was found to be the strongest. 79 schools had at least one Small Class, one Regular Class, and one Regular With an Aide Kindergarten class. In 40 schools or about 50% of the total, Small Class groups outperformed, which exceeds random expectations (i.e., 26 or 33%). However, it is not 100% either. Clearly, classroom size is only one variable – it is probably not the primary explanation.

Interpretation Difference: The improvement in performance can be explained by a “one-time” effect in Kindergarten and Grade 1 that persists regardless of whether pupils stayed in smaller class or were in regular classes in Grades 2-3 and beyond. It appears that the study demonstrates, at most, that a very large reduction in class size has positive effects in the first year or two of schooling. After that, the initial effects only manage to survive – they do not continue to increase even when the student remains in much smaller classes.
The interactions between Project STAR’s investigators and its critics have resulted in a carefully-worded benefit statement by a primary investigator: “The STAR study provides solid, experimental evidence of a ‘class size’ effect, its longevity, its academic and non-academic benefits, and the continuing growth of students who start schooling in small classes (15 or 18:1) in Kindergarten or grade one.”

Another issue germane to District 64 is that the outcomes involve student performance as measured by achievement tests. The mission of District 64 is to serve the needs of the whole child, which are measured utilizing more than achievement test results. None of the studies noted above sought to evaluate the impact of class size on other dimensions of the whole child.

What Do We Know?

Considerable disagreement has characterized exchanges among researchers. Although extensive research has been conducted, the literature offers little closure or clear direction for policy makers. Given potential performance and economic ramifications, impartial experts assert that improved studies and analysis are required before an effective classroom size “formula” is discovered.

However, several conclusions, supported by quantitative evidence, seem to represent the academic consensus:

  • Most positive reading and math performance effects from small classes occur in Kindergarten and grade 1.
  • Small classes can positively affect the academic achievement of economically disadvantaged and ethnic minority students.
  • Within a midrange of 23 to 30 pupils, class size has little impact on academic achievement of most pupils in most subjects above the primary grades.
  • The positive effects of class size on student achievement decrease as grade levels increase.
  • The use of teacher assistants does not seem to permit the classroom teacher to provide more individualized attention and thus raise pupil achievement.

From District 64’s perspective, the results of the STAR study and similar projects can help inform class size guidelines. However, it should be noted that these studies are more than 20 years old, often took place in low-income urban areas not analogous to Park Ridge-Niles, and did not attempt to assess class size impact on the whole child. External research is important, but is only one variable in the District’s class size strategies and guideline policies.

Further Reading
The following sources illustrate the range of debate about the efficacy of significant classroom size reductions as a means to improve pupil performance:

For further discussion and information, please visit: