About the Consortium
Research Plan
Consortium Members
Advisory Panel
Summary of Findings
Technical Report
Press Releases


July of 1996 marked the beginning of California’s Class Size Reduction (CSR) Initiative, a bold state effort to boost student achievement by limiting the size of kindergarten through third-grade classes. The program is a response to the continuing poor performance of California students. California ranks at the bottom on the National Assessment of Educational Progress fourth-grade reading achievement: In 1994 fewer than 20 percent of the state’s fourth graders scored at the proficient level, and more than half did not even reach the basic level, a benchmark that indicates only partial mastery of grade-level reading. The CSR program reduces class size from an average of 30--the highest in the nation--to a target of 20 or fewer students.

Size alone gives the initiative significance. It dwarfs other ongoing reforms in the state, and across the nation. With a FY-97 price tag of over $1 billion, or $800 for every participating K-3 student, it represents by far the largest educational reform in the history of this, or any other, state.1 Its impact is likely to be felt nationally since California currently educates more students than any other state, about one of every eight students in the nation. This school year, 1.9 million young children will be assigned to smaller classes because of the initiative. Although participation is not mandatory, over 95 percent of California’s districts took part, attesting to the popularity of the initiative.

CSR has the potential to reverse years of decline and serve as a model for other states. In a state with one of the lowest per-pupil expenditures in the country, many see the CSR’s injection of new funds into early grades education as cause for celebration. Furthermore, many teachers who used to feel that they spent more time keeping order in overcrowded classrooms than they did teaching, now believe it is possible for them to give each child enough individual attention to master core literacy and analytic skills. Educational field trips and projects that were once impossible because of large class sizes now become possible. Anecdotes of relieved and energized teachers set the tone of every discussion and news article about the initiative.

A Complex Undertaking

But the initiative is complex and its success is not assured; implementing and maintaining it are, and will remain, difficult. Working through the logistics of the initiative’s space and personnel needs has been a sizable task. The urgent demand for new teachers and more classroom space has refocused much of the state’s education agenda around the CSR initiative. For example, in 1996, the first year of the initiative, districts hired 18,000 new teachers to cover the new classes it generated. Almost one quarter of these teachers were uncredentialed, placing new demands on alternative certification programs. In the next two years, the demand for new teachers will be greater, and the proportion of under-prepared teachers will increase.2 Teachers in these smaller classes­both new and experienced­will also need additional training to master recently revised curriculum frameworks and to learn to teach small groups effectively for improved student achievement. Additional staffing problems are occurring in other grades and program areas because some of the teachers from upper grades and from programs addressing special-needs students have been attracted to CSR classrooms. This reassignment may cause less qualified and less experienced teachers to be assigned to the more challenging and larger classes.

To be successful, the program will require administrative support; but the time and energy the program absorbs from the state’s educators outside the classroom, including principals, superintendents, their staffs, as well as administrators at the State Department of Education, is time unavailable for other functions. Similarly, while CSR enhances education opportunities for students in the primary grades, it may pull resources away from the higher grades, leaving older students in worse shape than they are now. Finally, the initiative may conflict with and divert resources from other state and district reform programs, thereby interfering with ongoing efforts to improve schools. For these reasons, any evaluation plan must not only consider whether CSR is a good use of new state education funds, but whether it is the best possible use of these funds.

The Potential for Inequities

The initiative provides uniform funding per additional reduced size classroom to all districts, irrespective of local costs, which may accentuate inequities in educational resources and services across the state. For example, according to the California Research Bureau, the cost in FY-97 to districts for reducing class size ranged from zero to over $1,000 per student. In most districts, the FY-97 state allocation of $650 per student was below cost. While the enhanced FY-98 state appropriation will allocate more than enough to cover costs for many districts, others will still need to redirect funds from other budget lines to cover the additional cost of implementing the CSR program.

Furthermore, in order to participate, schools must find space. Early data suggest that the larger and more urban districts, as well as those faced with the highest enrollment growths, may have already reached space limits. These schools may be able to implement the program in just one or two grades. Many have had to convert libraries, computer rooms, and music rooms into classrooms. The majority of school districts are using portable classrooms to implement CSR, but many sites do not have places to put them. In some schools, CSR teachers "share" classrooms; the children are actually assigned to classes of 40 with two teachers. In addition, with the many new teacher openings and a statewide shortage of experienced credentialed staff, veteran teachers are transferring from lower- to higher-paying districts, possibly exacerbating resource differentials between the poor and the rich.

An Uncertain Impact on Teaching and Learning

CSR raises important questions about teaching and learning in smaller classes. Many believe that unless CSR teachers make conscious changes in their teaching strategies, the smaller classes alone will not improve students’ reading and mathematics abilities. Teachers and principals need continuous feedback about strategies that work to help them make the initiative succeed.

CSR comes at a time when California enrollments are growing and the proportion of children who are not proficient in English is at an all-time high ­ one out of three in grades K-3. In addition, one out of four children attending California’s public schools lives in poverty; the same proportion live in single-parent homes. It will be a great challenge for CSR to meet the language instruction and other special needs of all the state’s schoolchildren.

No Definitive Answers in the Literature

Research generally supports the effectiveness of class size reduction on a small scale, but leaves unanswered questions about why the reform works, what conditions are necessary to make it work; how to maximize its effectiveness when there are constraints on space, facilities and staffing; how well it meets the needs of limited English proficient (LEP) students; and whether it represents the best use of educational resources.

Although research on the effects of class size reduction is by no means unanimous, the balance of evidence suggests that substantial reductions in class size improve student achievement (Blatchford and Mortimore, 1994; Finn and Voelkl, 1992; Glass et al., 1982; Illig, 1997; Mosteller, 1995). The effects are strongest for students in the early primary grades (Educational Research Service, 1980), for low-achieving students (Angrist and Lavy, 1997; Krueger, 1997), and for students from poor socio-economic backgrounds (Finn and Achilles, 1990). It also appears that achievement gains are greater when classes are smaller ­ 20 students or fewer (Glass and Smith; 1978). Reducing class size also appears to decrease retention and referrals to special education (Snow, 1993; Illig, 1997), and it boosts teachers' morale and job satisfaction (Glass, et al., 1982; Shapson, et al., 1980).

However, "the question of why these effects are realized remains largely unanswered" (Finn and Achilles, 1990). As a result, research offers little guidance for implementing the reform on a large scale. What should teachers do to take advantage of small classes? What specific professional development should districts provide to newly-hired staff? The few studies of teaching practices in reduced-size classes do not provide clear answers. Cahen et al. (1983) studied four classes intensively after enrollments were reduced and found that changes in practice did occur, but were not dramatic. "Teachers and students were happier and more productive" but "the process of instruction looked very much the same" (p. 201). Shapson et al. (1980) found similar results in a study of fourth-grade classes in Toronto. These researchers noted marked improvements in teachers' attitudes, but little corroborating evidence of changes, for example, in the proportion of time teachers allocated to whole class, group, or individual activities.

The literature is silent on a number of other important questions. How should administrators allocate scarce resources, such as classroom space and experienced teachers? The Tennessee STAR program, which was the largest and best-controlled study of class size reduction, involved 79 schools (Mosteller, 1995). In general, the schools all had the necessary facilities and were able to recruit trained staff to make the program operate smoothly. Other experiments were conducted on smaller scales and did not have to address such problems. Consequently, the literature does not address the larger policy questions that arise when implementing such a reform statewide (Murnane and Levy, 1996; Mitchell and Beach, 1990). What supportive policies are needed to realize the benefits of class size reduction? How does class size reduction compare to other uses of educational resources?

The Value of an Evaluation

Without a thorough, independent evaluation of this expansive effort, timely responses to these vital policy issues are not likely. Below, an evaluation plan is developed that addresses the implementation of the program as well as its effects on schools, classroom practices, and student achievement. The evaluation should be both formative, providing feedback for improvement during the life of the program, and summative, generating results on cumulative impact.

The formative component provides state education policy makers with information about how best to implement this program. It identifies problems as they arise, points to potential solutions and provides insights into how educators throughout the state are responding to CSR issues.

The summative component helps answer the question foremost in people’s minds: Do smaller classes help improve achievement?

In addition, however, the summative component should uncover the program’s subtler effects:

  • Has the reform led to beneficial changes in classroom practices?
  • Under what circumstances?
  • Has the reform reinforced, mitigated or created inequities among California students?
  • How have other educational programs and upper grades been affected by the focus given to the early grades?
  • And how have limited English proficient (LEP), special education, and other "at-risk" students been affected?

Finally, the summative component should help answer those questions that must be addressed for all major public initiatives ­ whether it attains its goals, whether it has unanticipated consequences that need addressing, and whether it is worth the energy and resources it commands.

The findings from an evaluation of the CSR Initiative will have relevance for the current national discussion about what works to improve student achievement, which is the goal of numerous ongoing state and national reform efforts. At the moment, a growing list of states (including Connecticut, Florida, Georgia, Hawaii, Iowa, Kansas, Louisiana, Massachusetts, Michigan, Minnesota, Nevada, New York, and Utah) are implementing or considering reductions in class size.

1 Legislators enacted CSR as part of a package of reforms to raise reading achievement. CSR funds were accompanied by $80 per student for K-3 reading materials.

2 In most participating schools, Year 1 implementation reduced class size in one or two grades. Schools were required to reduce class size in first-grade classrooms first. Second grade was second priority, and either third grade or kindergarten was third priority. In 1997, some schools will implement the initiative in three or even four grades, from kindergarten through third grade.


Next Section: The Evaluation Plan's Design and Key Research Questions