Bain, Helen and C. M. Achilles, "Interesting Developments in Class Size," Phi Delta Kappan (May 1986)
Connor, Kim and Rosemarie Day, "Class Size: When Less is More," Senate Office of Research, California Legislature, Sacramento (February 1989)
Educational Research Services, Class Size Research: A Critique of Recent Meta-Analyses, Corporate Author, (Arlington, VA) 1980
Finn, Jeremy and Charles Achilles, "Answers and Questions About Class Size: A Statewide Experiment," 27 American Educational Research Journal (Fall 1990)
Finn, Jeremy, DeWayne Fulton, Jayne Zaharias, and Barbara Nye, "Carry-Over Effects of Small Classes," 67 Peabody Journal of Education, (Fall 1989, published 1992)
Glass, Gene and Mary Lee Smith Meta-Analysis of the Relationship of Class Size and Student Achievement San Francisco, Far West Laboratory for Educational Research 1978
Glass, Gene, Leonard Cahan, Mary Lee Smith, and Nikola Filby, School Class Size: Research and Policy Sage Publications (Beverly Hills, London, New Delhi) 1982
Hanushek, Eric, "The Economics of Schooling: Production and Efficiency in Public Schools," 24 Journal of Economic Literature (September 1986)
Hanushek, Eric, with other authors, Making Schools Work: Improving Performance and Controlling Costs, The Brookings Institution (Washington D.C.) 1994
Hanushek, Eric, "Money Might Matter Somewhere: A Response to Hedges, Laine, and Greenwald," 23 Educational Research (May 1994)
Hedges, Larry V., Richard Laine, and Rob Greenwald, "Does Money Matter? A Meta-Analysis of Studies of the Effects of Differential School Inputs on Student Outcomes," 23 Educational Research, (April 1994)
Levin, Henry M., "The Economics of Education for At-Risk Students," in Emily P. Hoffman, ed., Essays on the Economics of Education, W.E. Upjohn Institute (Kalamazoo, MI) 1993
Maddan, Nancy, Robert Slavin, et al, "Success for All: Longitudinal Effects of a Restructuring Program for Inner-City Elementary Schools," 30 American Educational Research Journal, (Spring 1993)
Mitchell, Douglas, Cristie Carson, and Gary Badarak, How Changing Class Size Affects Classrooms and Students California Educational Research Cooperative, University of California at Riverside (May 1989)
Mitchell, Douglas, Sara Ann Beach, and Gary Badarak, "Modeling the Relationship Between Achievement and Class Size: A Re-Analysis of the Tennessee Project STAR Data," 67 Peabody Journal of Education (Fall 1989, published 1992)
Mosteller, Frederick, "The Tennessee Study of Class Size in the Early School Grades," 5 The Future of Children (Summer/Fall 1995)
Nye, Barbara, et al, "The Lasting Benefits Study: Grade 6 Technical Report," Center of Excellence for Research in Basic Skills, Tennessee State University 1993
Nye, Barbara et al, "The Lasting Benefits Study: Eighth Grade Technical Report," Center of Excellence for Research in Basic Skills, Tennessee State University 1995
Odden, Allan "Class Size and Student Achievement: New and Affordable Strategies That Make Sense," School of Education, USC, Report for the California Department of Education (May 1989)
Robinson, Glen, "Synthesis of Research on the Effects of Class Size," 47 Educational Leadership 1990
Robinson, Glen and J. H. Wittebols, Class Size Research: A Related Cluster Analysis for Decision Making, Educational Research Service (Arlington VA) 1986
Sadowski, Michael, "The Numbers Game Yields Simplistic Answers on the Link Between Spending and Outcomes," 11 The Harvard Education Letter (March/April 1995)
Slavin, Robert E., "Class Size and Student Achievement: Small Effects of Small Classes," 24 Educational Psychologist, (1989)
Slavin, Robert, "Class Size and Student Achievement: Is Smaller Better?" 62 Contemporary Education (Fall 1990)
Snow, Mary, The 1993 Class Size Reduction Evaluation Study Nevada Department of Education (August 1993)
Tillitski, Christopher, "The Longitudinal Effect Size of PRIME TIME: Indiana's State Sponsored Reduced Class Size Program," 62 Contemporary Education (Fall 1990)
Tomlinson, Tommy, "Class Size and Public Policy: The Plot Thickens," 62 Contemporary Education (Fall 1990)
Word, Elizabeth, Director, et al, "Project STAR Final Report, 1985-1990," Tennessee State Department of Education, 1990
2 Project STAR was conceived in the context of an already enacted set of comprehensive education reforms. These reforms included a new career ladder for teachers (with master teacher designations and merit pay) and a Tennessee Center for Excellence program (designed to provide incentives to universities in Tennessee to work toward better educational outcomes). Also, the legislature created a new skill-based early elementary curriculum called Basic Skills First. Finally, the legislature funded a class-size reduction demonstration project at one Nashville elementary school.
3 The Tennessee Department of Education provided summaries of the research literature and other documents to the legislator for his review.
4 Thus, three basic comparisons were possible: comparisons of small to regular class, small to regular classes with aides, and regular classes to regular classes with aides. Since most regular classes and small classes had part-time teaching aides, the comparisons between the regular and small classes, and the two regular class types was, in fact, a comparison between a part-time and a full-time aide.
5 Kindergarten was not mandatory in Tennessee during the time of Project STAR, therefore some students completed four years in the same condition while others completed only grades 1-3.
6 Performance or achievement means class average scores on the tests given at the end of each year. Performance differential or achievement gain means statistically significant differences in the average scores between the groups being compared.
7 Glass, et al 1982.
8 Minority students were almost exclusively African American. Latinos, Asians, and other minorities comprised less than 2 percent of the minority population in the demonstration.
9 It is important to note that the small class differential began declining after the first year children were in the small classes and by the fourth grade it was only about 50 percent of its value at the end of the first year.
10 This section relies on several recent reviews of the literature, including: Project STAR 1989, Odden 1989, Mitchell, et al 1989, Slavin 1990, Robinson 1990, Hanushek 1994, and Sadowski 1995.
11 Glass, et al 1978.
12 Educational Research Service 1980.
13 Summarized in Glass, et al 1982. See also Educational Research Service 1980, and Robinson and Wittebols 1986.
14 Project Prime Time was approved by the Indiana legislature in 1981 and began in the 1983-84 school year in 24 kindergarten through second grade classes. The demonstration tested class sizes at 14 students per teacher in a variety of schools.
15 Slavin 1989.
16 The Glass-Smith finding is derived from the use of statistical techniques to estimate a curve that is based on extrapolation from the results of 14 "well controlled" studies reviewed by the authors. Slavin's analysis examined the actual data underlying the Glass-Smith curve, and found that these data suggest much smaller gains than the gains based on the Glass-Smith estimates. Slavin also suggested that other studies could be excluded either because the study examined only short duration small classes (30 minutes), or the study examined post-secondary classes. Leaving these studies in the analysis, however, would not materially affect the results.
17 The study Slavin excluded was an experiment that tested whether class size affects motor skill training. The achievement test used in the experiment consisted of rallying a tennis ball off a wall for a specified period of time.
18 Snow 1993.
19 Sadowski 1995.
20 The Project STAR team and other researchers also suggest that class size reductions should be accompanied with other changes including changes in teaching style. Some of these changes may evolve naturally when classes are smaller; but these changes cannot be assumed. A recent book published by the Brookings Institute draws similar conclusions (Hanushek 1994). This book summarizes the work of a prominent group of education economists with diverse backgrounds. It supports the notion that smaller classes must be imbedded in more comprehensive reforms.
21 Some of the concerns mentioned in the Project STAR report also have been identified by other researchers. Issues that could affect the results include: (1) removal of 108 children from small classes and reassigning them to regular or regular with aide classes; (2) three schools dropped out of the experiment at first grade and another dropped out at second grade (two were inner city schools, one was an urban school and one was either a suburban or rural school); (3) comparisons between the regular and regular with aide classes are affected by regular classes having part-time aides; (4) schools and teachers in the study were chosen from among schools and teachers that volunteered; thus, may not be representative; and (5) attrition from the sample at all schools and for all class types was significant and as the report suggests the children who remained for the entire study may not be representative. Other researchers have noted that the manner of random assignment inadvertently may have led to relatively more low achievers in the large classes while distributing the high achievers more equally between the large and small classes.
22 The reading initiative includes funds for a new curriculum stressing reading skills, staff development, and materials.
23 These include: desegregation funds, economic impact funds, Healthy Start funds, and special education funds.
24 The Governor's class size reduction proposal includes an evaluation component and it could be a component of the evaluation needs discussed in this paragraph.
25 Levin 1993.
26 Madden, et al 1993.
27 These studies are summarized in several review articles published in Long-Term Outcomes of Early Childhood Programs, a special issue of The Future of Children, Center for the Future of Children Vol. 5 No. 3 (Winter 1995) and Home Visiting, a special issue of The Future of Children Vol. 3 No. 3 (Winter 1993).