Column

Despite popularity with parents and teachers, review of research finds small benefits to small classes

Researchers in Denmark sorted through 127 studies, many conducted in the United States

Photo of Jill Barshay

Proof Points

A classroom with only a few desks in Richton, Mississippi.

A small classroom in Richton, Mississippi.

Small classes are very popular with parents. Fewer kids in a room can mean more personal attention for their little ones. Teachers like them too. Fewer kids mean fewer tests to mark and fewer disruptions. Communities across the United States have invested enormously in smaller classes over the past 50 years. Pupil-teacher ratios declined from 22.3 in 1970 to 17.9 in 1985 and dropped to a low of 15.3 in 2008. But after the 2008 recession, local budget cuts forced class sizes to increase again, bumping the pupil-teacher ratio up to 16.1 in 2014, the most recent federal data available.

There’s a general consensus among education researchers that smaller classes are more effective. (In graduate school, I was taught that the benefits of small classes kick in once the class size falls below 16 students.)  The benefits of small classes have become something of an informal yardstick. When I have written about unrelated educational reforms, researchers often compare them to the effectiveness of class size reductions to give me a sense of their relative impact.

But that general consensus masks some important disagreements. Experts have long known that the research evidence doesn’t consistently support the notion that smaller classes increase how much students learn. In 2002, the debate about the merits of small class sizes erupted into a public spat between Eric Hanushek of Stanford University’s Hoover Institution and Alan Kreuger of Princeton University.

Now, a new October 2018 review of class size research around the world finds at most small benefits to small classes when it comes to reading. In math, it found no benefits at all.

“Class size reduction is costly,” the researchers wrote. “The available evidence points to no or only very small effect sizes of small classes in comparison to larger classes. Moreover, we cannot rule out the possibility that small classes may be counterproductive for some students.”

The study, “Small class sizes for improving student achievement in primary and secondary schools: a systematic review,” was produced by three researchers at the Danish Center for Social Science Research, an independent research center, and published by the Campbell Collaboration, an Oslo-based organization named after an American scientist who argued that government reforms could generate scientific evidence to produce better policies in the future.

Related: Documenting Maine’s failure to implement proficiency-based education

Robert Slavin, director of the Center for Research and Reform in Education at Johns Hopkins University, brought this study to my attention, explaining that this new review confirms a similar review of the research on small class sizes he conducted in 1989. “[I]n sending my own kids to school, I’d prefer to send them to schools with smaller class sizes,” Slavin wrote to me by email. “But…no studies of high quality have ever found substantial positive effects of reducing class size.”

Reviews by the Campbell Collaboration use very stringent criteria for including a study in its analyses. The researchers began by collecting 127 high-quality studies on class size reduction from 41 countries from kindergarten through grade 12. More than a third of them — 45 studies — analyzed Tennessee’s Student-Teacher Achievement Ratio (STAR) experiment in the 1980s. The researchers then looked carefully at all of these studies, making sure that the studies compared test results between students in smaller classes and with those in larger classes.

They rejected studies that could be biased or influenced by other “confounding” factors. For example, if parents had the ability to influence which classroom their kids were put in for the experiment, that could bias the results in favor of small classes. If the comparison classrooms didn’t have similar demographics or academic achievement at the start of the experiment, it would be discarded. Some studies were missing key pieces of information such as failing to specify whether the test results were in math or reading. Another common problem was failing to analyze the results by the actual number of students in each class. Many studies used a proxy of what the school district’s rule was for the maximum number of students allowed or average pupil-teacher ratios.

In the end, only 10 studies, including four on the Tennessee experiment, made it to the final analysis. The others were from the United States, the Netherlands and France. To understand how small the overall benefit to reading was, the researchers calculated that there was only a 53 percent chance that a randomly selected score of a student from a small class would be greater than a randomly selected test score of a student from a larger class. That’s not much better than tossing a coin. “This is a very small effect,” the researchers wrote.

In math, overall test scores were slightly lower in the small classes, but it wasn’t statistically significant. That’s why the authors say they cannot rule out that some children may be adversely affected by small math classes. How can small classes possibly be harmful? The logic goes like this: when you reduce class sizes, you need to form more classes, say, five classes of 15 students instead of three classes of 25. The school suddenly needs five teachers instead of three. Assuming that the two new teachers it hires are of average capabilities, that could be a downgrade in teacher quality if the existing teachers were talented veterans.

Would you rather your child were in a small class with an average teacher versus a larger class with a good teacher?  Reasonable people can disagree. Slavin suggests that it might be more effective to operate larger classes, supplemented by one-to-one tutoring for kids who need it.

Related: Putting an ed tech darling in context

Andreas Schleicher, director of the education and skills unit at the Organization for Economic Cooperation and Development, has long been arguing that the U.S. overemphasizes small classes at the expense of good teaching. In his 2018 Book, “World Class,” Schleicher debunks the “myth” that “smaller classes always mean better results.” He points out that top performing nations, such as Japan, South Korea and China, tend to have larger classes but spend more of their resources on giving their teachers time during the workday to produce good lessons or give students extra attention outside of class.

“Even if … smaller classes might yield small gains, the more important question to ask is if it is the best investment of an additional dollar,” said Schleicher by email. “You can spend a dollar only once and you need to decide whether you invest it into smaller classes, better salaries, more non-teaching working time, or longer student learning hours. If you make that calculation, almost any of the other factors yield better outcomes.”

This story about small classes was written by Jill Barshay and produced by The Hechinger Report, a nonprofit, independent news organization focused on inequality and innovation in education. Sign up for the Hechinger newsletter.

The Hechinger Report provides in-depth, fact-based, unbiased reporting on education that is free to all readers. But that doesn't mean it's free to produce. Support our mission today and, thanks to NewsMatch, your one time donation will be doubled or your new monthly donation will be multiplied 12 times.
Letters

Jill Barshay

Jill Barshay is a contributing editor who writes the weekly “Proof Points” column about education research and data. She taught algebra to ninth graders for… See Archive

Letters to the Editor

Send us your thoughts

At The Hechinger Report, we publish thoughtful letters from readers that contribute to the ongoing discussion about the education topics we cover. Please read our guidelines for more information.

By submitting your name, you grant us permission to publish it with your letter. We will never publish your email. You must fill out all fields to submit a letter.





The reason researchers find something so counterintuitive as "small classes don't matter," is because they assume intuition has no value, for starters. This study defines "school quality" in terms of a few standardized achievement test scores. Before making life altering conclusions that bring danger, depression, loneliness, drop outs, and a host of other negative conditions into the schools, please look at the data on class size vis a vis measurements that matter more for a great nation and culture: happiness, curiosity, engagement, connectedness/belonging, kindness, graduation, and health. It's tragic to see quality measured in areas that bring neither joy nor productivity to our nation. In classes below 15, student engagement levels rise along with a host of positive outcomes--even if our students care LESS about standardized outcomes in those classes.

- from Dr. Stuart R Grauer, Ed.D, Nov 06, 2018

I am always surprised when education study results do not look at the limitations of the studies affecting results to say that the results need to be examined in some other way, by going into classrooms with small class size and large class size to do a qualitative analysis of what is going on. I am always surprised that the only way we decide that something is efficacious is through test data.

To me, a missing element in this article is a discussion about the curriculum/instructional materials that were chosen to guide classroom instruction for which the students in small and large classes were tested. This is a major flaw and it occurs over and over in quantitative education studies. Also, even when studies identify the instructional materials used by the teachers, controlling for fidelity of how teaching is conducted in each class is an independent variable that will affect the results and is almost impossible to control.

I will read the study for which there is a link in this article. Perhaps that will clear up some of my criticisms but I doubt it.

- from Nora, Nov 07, 2018