Class size matters


by Don Fraser • Last weekend the National Invitational Conference of the Early Childhood Research Collaborative was held at the Federal Reserve Bank in Minneapolis. The Bank and the U of M Center for Early Education and Development were joint sponsors.

A discussion of class size was led by Jeremy D. Finn at the University at Buffalo – SUNY. He concluded that smaller class sizes at the earliest grades had substantial benefits especially for minority students.

His paper (co-authored by Allison Suriani & Charles M. Achilles) was titled “Small Classes in the Early Grades: One Policy – Multiple Outcomes”. He began by refuting a widely held opinion that lower pupil-teacher ratios showed little benefit. He argued that pupil-teacher ratios ignore the actual setting in which teachers are teaching and students are learning. He said that “it is not surprising that the relationship of PTR (Pupil-Teacher Ratio) with student achievement is weak at best” partly because studies have alternated between class size and pupil-teacher ratio, with no reference to the actual teaching or learning process in the classroom.

The paper goes on to describe the STAR policy study funded by the Tennessee legislature in 1985. This study began in the earliest grades (K-3) with 17 inner-city, 16 suburban, 8 urban and 38 rural schools. Within each school all students were assigned at random to one of three conditions: small class with 13-17 students, a regular class with 22-25 students, and a regular class with a full-time teacher aide with 6300 students taking part.

Short-term findings were that small classes have academic benefits in every grade and in every subject tested in K-3. Benefits equated to 1/2 to 5-1/2 months of schooling. Starting early and continuing in small classes for multiple years also helped – “in general, the more the better.” The benefits of small classes were greater for minority students or students attending inner-city schools than for white students in non-urban schools. The advantages for minorities were often two to three times as great, thus reducing the white/minority achievement gap – leading to recommendations that small classes should be “targeted” to schools serving low-income and minority students.

The report goes on to say that “these findings have been corroborated in non-experimental research” citing a number of studies.

Other positive outcomes included fewer small-class students recommended for in-grade retention (being “held back”) in grades 1, 2 and 3.

There were no academic benefits from having teacher-aides in the regular classes in any subject matter in any grade “no matter what functions the aides performed in the classroom.”

Long-term outcomes – after students returned to full-size class rooms the academic benefits continued in all subsequent grades. More of the students who attended small classes took college entrance tests (SATs and ACTs) “The benefit for Black students was substantially greater than for white students, reducing the Back-white gap in college-entrance-test taking by 54%.”

Graduation rates were significantly impacted by small class participation. “For all students combined, the effects of attending small classes for 4 years increased the odds of graduation by about 80%.” The rates for low-income students with 3 or more years of small class participation were as high as those of higher-income students.

The randomized assignments of students produced heterogeneous classes. The paper notes that if the class is comprised only of students who are difficult to teach, the same benefits may not be realized.

One analysis suggest that maintaining the same class grouping across the years contributed to the positive benefit of small classes, and noted that certified teachers produced greater learning gains with small classes than with larger classes.

In looking at why small classes affect student performance, systematic interview and classroom observations don’t support the hypothesis that smaller classes provide more individualized instruction. Studies do affirm that teachers of small classes spend more time on instruction and less on classroom management or matters of discipline, and have more time to “listen to children, to get to know their personal lives and concerns.” The most visible changes are not in teacher’s behavior, but in students’ behavior…more engaged in learning, and better behaved.

Other findings: In small classes all students feel pressure to participate, and the teacher can’t “easily ignore” any particular student. Students have a “sense of belonging” being more cohesive with splinter groups rare and benefiting from a “psychological sense of community”.

A substantial portion of the paper is devoted to costs and economic benefits. The paper observes that the costs of reducing class size can be high, and notes that much of the overall savings derived from “reduced dropout rates, reduced health care or welfare costs, or the increased earning power of higher-achieving students” don’t accrue to the school district itself. But the paper also notes that some school districts have implemented lower class size at much lower costs.

One analysis showed that reducing class size from 22 to 15 for grades K-3 would yield an internal rate of return of approximately 6%. “Assuming a 4% interest rate, the benefits of reducing class size to 15 in terms of lifetime earning, would be 43% greater than the costs, and 100% greater than the costs if real wages grow by 1 percent per year.”

The paper makes one further interesting observation: Several surveys have documented that small classes may be an incentive to teachers because of improved working conditions and increased student outcomes. It cites a study of 234 teachers asked to choose between a small class, a full-size class with an aide, or else a $2,500 bonus. “Of teachers who had been teaching small classes, 81% said they preferred small class over a full-size class with an aide, and 70% chose the small class over the salary increase.” Another survey asked “Which do you think is a better use of our education dollars, increasing teacher salaries or decreasing class size?” Of all respondents, 23% chose salary increase and 77% chose decreasing class size. Among teachers, 19% chose salary increase and 81% chose class size reductions.

It may be possible to obtain copies of this paper from the author, J. Finn at (716) 645-2484 ext 1071