Review of the Research
A review of the research literature shows an extensive amount of research on the effects of class size and student achievement. The literature review indicates a lack of consensus about the benefits of reducing class size. One study concluded: “It is unfortunate that the subject of public education has become so polarized that policy debates, allegedly based on scholarly research, have become more contentious than the research itself seems to require.”(1) Regarding the class size debate, Jennifer Rice King offers, “…although research on the impact of class size has been conducted, the literature offers little closure or clear direction for policy makers considering investments in smaller classes.”

The research is about evenly split into two camps. One camp contends that increasing the number of teachers to reduce the pupil-teacher ratio is a very costly proposition with little improvement of student achievement. The second camp offers that “money matters” and reducing class size is a viable way to improve student achievement.

One of the first conclusions from reviewing the literature on the subject is that researchers have not used the same terms and definitions when examining the issue. It therefore becomes very important to define the terms. What is the difference between pupil-teacher ratio and class size? Class size usually refers to the number of students under a teacher's control, whereas pupil teacher ratio, or PTR, refers to the number of students in the facility divided by the number of staff in the facility - in some cases this may or may not include positions such as aides, teachers working as administrators, etc.

Examination of a core set of data with consistent definitions over time can be found in the 2003 Digest of Education Statistics published by the National Center for Education Statistics. Information presented in the 2003 Digest shows pupil teacher ratios declining because of a decrease in the number of students and an increase in the number of teachers. According to the 2003 Digest , “During the 1970s and early 1980s, public school enrollment decreased, while the number of teachers increased. As a result, the public school pupil/teacher ratio declined from 22.3 in 1970 to 17.9 in 1985. After 1985, the number of pupils per teacher continued downward, reaching 17.2 in 1989. The pupil/teacher ratio was stable during the early 1990s, but began to decline again during the mid-1990s. By 1999, the ratio had stabilized again, and in 2002, a pupil/teacher ratio of 16.1 was estimated in public schools.” This ratio compares favorably to private schools where the pupil/teacher ratio for 2002 was estimated to be 16.2.

The 2003 Digest goes on to say that, “The pupil/teacher ratio includes teachers for disabled students and other special teachers, who generally are excluded from class size calculations. The average class size in 1999–2000 was 21.1 pupils for public elementary schools and 23.6 for public secondary schools.” Corresponding numbers for Pennsylvania show the average sizes were 22.2 for public elementary schools and 23.2 for public secondary schools.

Several research studies indicated that there are about five to ten more students per class based on the pupil/teacher ratio than for “average class size.” The difference becomes an issue when describing the learning environment. The math of “class size or pupil teacher ratio” gets pretty messy when trying to sort between these factors. For example, a school with a large support staff of teaching assistants, special aides, coaches, nurses, etc. might show a favorable pupil/teacher ratio. But in the same school, the actual number of students in a classroom might be substantially higher.

Thus, the distinction between class size and PTR has become more problematic as the number of specialized teachers, teacher's aides and other support staff has expanded in response to requirements from legislation such as IDEA/IDEIA. On the other hand, an “average class size” of 20 might be comprised of one class with 15 students and another with 25 students. However, if a classroom has two teachers and 30 students, the specific PTR is 15 to 1 but the class size is 30. If the research papers have not always been clear about defining which factors are used to measure the issue, the result is the current state of conflict in the research.

A study of the California class size reduction initiative suggested that students in poor urban areas with higher concentrations of minority students would most likely benefit from class sizes of less than 20. Competing studies indicate that reducing class size has minimal improvement on academic achievement.

Dr. Eric A. Hanushek, of the Hoover Institution, Stanford University, and a research associate of the National Bureau of Economic Research, is a noted opponent of smaller class size as a mechanism for improving educational outcomes. Hanushek has noted that class size is not a panacea for improving education. He contends that decreasing class size involves hiring more teachers and has generally resulted in lowering credential requirements. He argues that refocusing resources on improving teacher quality is more important. But even Dr. Hanushek, like other researchers that generally oppose broad application of class size reduction, seems to agree that smaller classes can be beneficial for specific groups of students, in specific circumstances, or for subject areas. This seems to be true especially in primary grades K-3.

According to Dr. Alan B. Krueger, of Princeton University: economics is, “at heart, questions concerning the desirability of spending more money to reduce class size.” Krueger adds, “Aside from the opportunity cost of students' time, teachers are the most important, and most costly, factor of production in education.”

Research by Edward Lazear, points out those students in smaller classes experience fewer disruptions, allowing for more orderly learning time. To this position, Krueger asserts that students in smaller classes have closer supervision and this is likely to re-enforce discipline that carries forward to other classes regardless of size.

Lazear, contends that “optimal” class size can be larger for students who are well behaved; therefore, these students would benefit less from class size reduction efforts. However, Dr. Krueger opines that a tendency to place weaker students in smaller classes, where additional resources can be applied, could lead to erroneous conclusions about the efficacy of smaller class size as the academic performance of these less able students is evaluated in relation to class size.(2)

An April 2003 article in Educational Leadership outlined the results of research studies on three of the largest and most well known class-size reduction programs. These were Indiana's Project Prime Time, Tennessee's Project STAR (Student Teacher Achievement Ratio) and California's CSR (Class Size Reduction) initiative.

Indiana's Project Prime Time was implemented during the 1980s by providing for reduced class size in K-3 classes at 24 randomly selected schools. The initial review of results was positive and encouraging. Subsequent reports concluded that benefits were negligible. The follow up studies also indicated little difference in teaching methods between small classes and larger ones. In an article in the Pennsylvania Educational Leadership magazine, Jacqueline Saylor expressed a concern that teachers might have been biased towards success of the program. However, there was sufficient satisfaction on the part of teachers, administrators and legislators that additional funding was put into the program.

Tennessee's program was the most extensive class size study to date. The Student/Teacher Achievement Ratio (STAR) Project began in the 1980s and continued until 1990. Seventy-nine schools participated in the four-year project. Two groups of students, an experimental group and a control matched for demographic characteristics, were assigned to small classes of 15 or less and larger classes of up to 25 students per teacher. About half of the larger classes were assigned a teaching aide. There was an equal number of each type of class. Based on pre and post-tests, small classes consistently showed greater academic achievement than either of the larger classes and there was no discernable difference between the larger classes with and without aides. Some researchers have criticized the selection process for participating schools as a weakness in the design of the project.

According to Achilles, Finn and Pate-Bain (February 2002, Educational Leadership ), Project STAR researchers and many other class size researchers have concluded that to have substantial, long-term effects, class size reduction efforts must meet three conditions: early intervention, duration and intensity. Early intervention requires that students must begin school in small classes in kindergarten or pre-K.

Duration means that students are in small classes for at least three years, but preferably four years. Intensity is defined as maintaining students in small classes every day, all day.

Hanushek reported, “The gains in performance from the experimental reduction in class size were relatively small (less than 0.2 standard deviations of test performance), especially in the context of the magnitude of the class size reduction (around eight students per class). Thus, even if Project STAR is taken at face value, it has relatively limited policy implications.” Hanushek also raised questions about, “…teachers all knew they were participating in an experiment that could potentially affect the future resources available from the state. Small schools were excluded from the experiment, and all participating schools were willing to provide their own partial funding to cover the full costs. The net result of each of these effects is difficult to ascertain, but there is prima facie evidence that the total impact is to overstate the impact of reduced class size (Hanushek 1999b).”(3)

California's program was implemented in the mid 1990s as the result of a significant budget windfall. Under this initiative, class sizes were to be reduced from often 30 or more pupils, to no more than 20. Due to rapid implementation, there was a derived shortage of qualified teachers to fill the new positions. Compounding the shortage in some schools, many of the more qualified teachers moved from urban/poor districts to the suburban and more affluent districts. The ultimate result was the hiring of many less than fully qualified teachers. Also, due to the increase in the number of classes, many school buildings were short of classrooms. Classroom space shortages resulted in the use of less than adequate facilities such as libraries or band rooms.

Experience from the California CSR Initiative demonstrates that rapid change for smaller class size can create unintended consequences and any change should be phased in. In examining future studies of class size, qualifications of teachers and aides, as required by the No Child Left Behind Act of 2001, will be significant.

Researchers from the Education Policy Studies Laboratory and the University of Wisconsin-Milwaukee reporting in Results of the Student Achievement Guarantee in Education Program Evaluation (SAGE), 2001, found that a 5-year-old program in Wisconsin produced higher achievement results for children living in poverty. The primary objective was to reduce class size 15:1 or less for grades K-3. The program which began in 1995 with 30 schools has been expanded to include more than 600. The state legislature established funding that provides $2,000 for each low-income student in participating schools. Also, some funding has been provided to pay debt service on construction of additional classrooms to accommodate the additional classes. The Wisconsin Department of Public Instruction agrees the results have been “generally favorable.” The DPI has noted that other changes are also occurring. As a result, some of the achievement may be the result of increased collaboration between schools and their communities, along with the implementation of a rigorous curriculum focusing on academic achievement. There is further blurring of express agreement that the program is solely responsible for improvement because of expanded standards that is further supported by expanded professional development and staff evaluation practices. The DPI feels that class size reduction may not be the most significant factor contributing to improvement.

In 2000, the federal government created a class-size-reduction program including funding for states to recruit, hire and train new teachers for grades K-3. U.S. Department of Education published a paper in September 2000 titled “The Class-Size Reduction Program, Boosting Student Achievement in Schools Across the Nation, A First Year Report.” This report outlines the initial program funding to increase the number of teachers as a means of reducing class size. The report indicates that reduced class size generally results in improvement in student academic performance. A close reading of this paper shows that improvement seems to be mostly in terms of documented class size reductions, but only anecdotal evidence of academic success.

Under the reauthorized Elementary and Secondary Education Act, popularly known as "No Child Left Behind," the federal program was consolidated into a more general teacher-quality block-grant program. The final distribution of Class-Size Reduction funds was made to states on October 1, 2001. For FY 2002, the Class-Size Reduction Program was incorporated into the new ESEA Title II Teacher Quality block grant. States and local education agencies were permitted to use any portion of the nearly $3 billion in Title II funds to hire qualified teachers to reduce class size.

An examination of international data by Martin R. West and Ludger Woessmann suggest that, “…the existence of class-size effects is related to the quality of the teaching force. Smaller classes appear to be beneficial only in countries where average teacher quality is low….” The article “Crowd Control, An International Look at the Relationship Between Class Size and Student Achievement” appeared in Education Next , Spring 2003. Their conclusion is to, “…devote the limited resources available for education to employing more capable teachers rather than to reducing class sizes.”

Some researchers, like Bruce Biddle and David Berliner, Education Policy Studies Laboratory, seem to equate two basic positions on class size to politics. Essentially, liberals and teachers want small classes and more teachers, therefore, find nothing but benefits from smaller classes. Fiscal conservatives, on the other hand, want to hold the line on spending and so find little benefit to smaller classes, contending that other efforts might have greater return on resources.

Summary
Several common themes appear in a review of the literature on class size. Almost all agree that there is some benefit to smaller class size in kindergarten through third grade. However, this benefit has been limited to large numbers of poor and/or minority students. Most of the research agrees that there is carry over of the benefits from smaller class size in the early grades to academic performance in later grades regardless of class size in the later grades. It might be worthwhile to consider smaller class size in K-3.

Research on class size reduction generally has provided mixed results. Most researchers find some beneficial outcomes, especially in class sizes of 15 or less. But whether the improvement warrants a significant addition of resources is problematic. Some interpret the results as too ambiguous to justify the huge expenditure for a concerted effort to reduce class sizes in all schools. It is almost universally agreed that the greatest benefit accrues to students in K-3, where there are students from backgrounds that do not typically encourage academic achievement. Most researchers and commentators conclude that any effort to reduce class size should begin at the lowest grade (kindergarten) and phase in year-by-year to at least grade 3 for the beneficial results to carry forward.

Individual attention when coupled with early intervention tends to instill discipline in student educational conduct. Students are more likely to be well behaved, which results in less disruption and more time for teaching and learning as a result of closer attention from the teacher. Most research on the topic indicates that there is carry over of the academic gains from small class sizes in K-3, even when class sizes in the higher grades are significantly larger. One of the corollaries noted is that where students are well behaved, class size has less impact on student academic achievement.

Conclusions
Targeted expenditures aimed at class size reduction appear to generate gains in academic outcomes with some sustained educational benefits. Expenditures work best according to the research when focused on the K-3 grade levels for those students from high poverty and/or high minority backgrounds where little or no emphasis is placed on educational success.

The impact of a decision to reduce class size has substantial financial impact. In addition to the salary and benefits of additional teaching staff, space needs associated with renovations or additions to existing buildings may be necessary. To reduce class size without appropriate academic space does not provide benefits. When reducing class size, space considerations are essential. The best gains occur when the class size reduction is maintained through the third grade.

Benefits also may be gained when small class size is targeted to students with learning difficulties (as distinguished from disabilities). Thus, small class size may be beneficial for a small portion of the district's students.

Professional development needs to be expanded for those teachers assigned to smaller classes to assure that teaching methods are appropriate for small classes vis-à-vis large group instructional techniques.

A school board should consider several items when discussing the implementation of class size reduction. The following is intended as a guide for school boards to evaluate the benefits of any class size reduction action.

Prior to implementing a program, establish the mechanism for documenting the impact of smaller class size. Identify the grades, groupings of students, buildings or other items that will exhibit the success/failure of class size reduction associated with student achievement. What are the needed resources related to facilities, staffing and professional development associated with the class size reduction program being proposed? Is there an ability to make any type of trade-off between class size reduction and other programs?