Does class size really matter?

Parents are dying to get their kids into smaller classes. But research shows they may be panicking over nothing

Published August 6, 2011 5:01PM (EDT)

Few things about a school seem to matter more to parents than class size. For many of us it is the litmus test for a well-run school. Small class size speaks of a school that is focused on putting resources in the right place -- not administrative retreats, paneling for the principal’s office, or expensive but rarely used class-room technology. Small class size is a signal to us that a hundred smaller decisions that accompany the running of a school have been shaped with our children as a priority. As a result, a school is able to invest in an appropriate number of teachers.

Classrooms with fifteen students and one teacher usually look better, too -- more controlled than classrooms with thirty kids. At best, we imagine that small classes are environments where our children will be closely observed and where teachers have the opportunity to get to know each child. We assume that in small classes our children will receive personalized attention and that learning can be sprinkled like stardust through the thoughtful, free-ranging give-and-take between student and teacher. Small class size creates an environment that invites parent involvement, as well. If your daughter is one of thirty second-graders, you know without being told that the teacher is going to be hard-pressed to remember which reading group your daughter is in, much less her progress with phonemes. It’s not surprising that so many parents will move heaven and earth to get their children in schools with a low teacher-student ratio.

But you may be surprised to learn that the effects of class size on learning are not 100 percent clear. Conventional wisdom tell us that smaller class size is crucial for learning -- that kids of all ages learn more in smaller groups. And indeed, in the early years of schooling, there is some research to back this up.

But there is a substantial body of research to suggest that kids in small classes don’t necessarily learn more. In the range of things that schools can do to improve outcomes for your child, reducing class size may rank a distant fourth behind solid teacher training, a clear and well-sequenced curriculum, and a staff that is well supported and regularly evaluated. For decades, class size was largely a function of a community’s population. A lot of kids born in a particular year? The local school found a way to cram them into classrooms. In the 1970s, though, as the discussion of the achievement gap sharpened and schools began to be seen as an instrument of racial oppression, "overcrowding" became a catch-all concept for the inequities between poor and middle-class kids in public education. Writers like liberal activist Jonathan Kozol decried the antiquated, crumbling, and overcrowded classrooms where poor children had their dreams denied. "The overcrowded classroom" was associated with poor performance, high truancy, and high rates of juvenile crime.

In the last twenty years, legislators have tried to institute state-wide standards in an effort to keep teacher-student ratios low, especially in poor and underperforming schools. Currently, thirty-two states now set aside funds for a voluntary or mandatory reduction in class size. These policies have had a substantial effect. In the last ten years, class size in America has declined -- and continues to drop. According to the National Center for Education Statistics, the average class size of U.S. elementary schools has been reduced from twenty-four pupils in 1993 to twenty pupils in 2007. Currently, not all poor kids are in overcrowded classes. In schools that serve rural poor kids, for instance, class sizes tend to be small. Urban schools that serve impoverished kids tend to be larger than their more affluent suburban counterparts, though. In public schools in inner-city Chicago, for example, kindergarten through eighth-grade classrooms are 14 percent larger than the average-size classrooms throughout Illinois (and considerably larger than the teacher-student ratio in schools in the affluent suburbs that ring the city). In New York City, fourth-grade and eighth-grade classroom sizes are 10 percent and 17 percent higher, respectively, than the average classroom size in the rest of the state.

Right now, the discussion about class size is a polarized and highly political one. Fiscal conservatives and corporate management-style reformers who have been enjoying center stage in so many recent education debates say that smaller class size is simply too darn expensive for what turns out to be very minor improvements in student achievement. And it comes at a huge cost. Florida alone has spent nearly $20 billion reducing class size after the number of students per classroom was capped in 2003.

In these difficult financial times, as states struggle to balance their budgets, state officials say they need to be more strategic with their dollars and to fund programs that show a greater and more consistent improvement for a wide swath of kids. They suggest that parents are manipulated by teachers unions into wanting small classrooms because it takes the scrutiny off the effectiveness of their rank and file.

Michelle Rhee, the former chancellor of schools for the District of Columbia, now heads the advocacy group Students First. Not long ago, she told a group of reporters that with the right preparation, bigger classes, not smaller ones, would be an effective way to raise test scores and save money. "The way that I think would make sense is to identify the most highly effective teachers in a particular district, and think about assigning a few more students to each of their classrooms," Rhee says. Those same fiscal conservatives and corporate-style school managers accuse the teachers unions of pushing the small-is-good agenda to the detriment of kids.

In general, the powerful teachers unions do endorse small class size. Although it is popular to bash the unions, you can look at their enthusiasm for small class size in a couple of different ways. It may be an honest reflection of the experience of the people who are on the front lines in education. A great number of classroom teachers point out that they can barely learn the names of thirty students by the end of the first month of school, much less pitch instruction to different learning styles so the students can learn best. Teachers also describe a sense of connectedness that can grow in a small class, creating a learning environment that is intimate, flexible, and, when it works, highly productive. A more cynical take is that the union support for small classrooms is part of an effort to protect the working conditions of its members. Smaller class size makes it easier for teachers to teach. It takes much less time to grade fifteen essays than thirty.

The most cynical take is that smaller class size also increases the number of teachers who are hired and strengthens the union that supports them. Randi Weingarten, head of the American Federation of Teachers, acknowledges that raising class size is a branch on a tree of hard decisions that cash-strapped states are facing. But, she says, "if somebody says they want to raise class size, they’re doing it because they want to cut the budget, not because it’s actually going to help children." Teachers’ union representatives point out that the same fiscal conservatives and corporate-type reformers who encourage high student-to-teacher ratios in classrooms are often the ones who send their own children to private schools where -- you guessed it -- the kids receive instruction in small groups, often twelve to fifteen in a class.

Does class size matter? For some interesting reasons, it’s hard for researchers to come up with a definitive answer. A group of analysts followed children over time and correlated the size of the class they were assigned to with how much they learned. Ordinarily, these would be juicy studies, but how kids ended up in classes with small teacher-student ratios created some confounding factors. Children who attended elementary school in affluent Wilmette, Illinois, for example, may have been educated in classes that had fifteen kids and one teacher and showed huge gains in their academic achievement compared to kids in larger classes in nearby inner-city Chicago. But was it the class size or the opportunities that went along with privilege that made the difference?

Doug Ready, an education sociologist from Teachers College at Columbia University in New York, points out that schools with class sizes that are small by default are different in many ways -- and provide different outcomes -- from those that provide smaller classes by design. But even when school administrators purposefully create small classes and researchers explore the learning outcomes, the results need to be taken with a grain of salt. Principals create smaller classes for a variety of reasons. They can assign tough-to-teach kids to smaller classes, or they can create smaller classes for their weakest or inexperienced teachers. In both these cases, graphs and charts of what the students learn are likely to reflect outcomes that depend on many factors that won’t necessarily apply to your child’s school or your community. Then there are studies that involve a large, nationally representative sample of children -- the same proportion of affluent and less-affluent kids that would be found in public schools around the nation -- which are a bit more revealing.

I'll look at two of the most prominent studies from the Early Childhood Longitudinal Study, Kindergarten class of 1998–99 (known in sociology circles by the catchy nickname ECLS- K). Those two studies assessed the learning of twenty thousand kids who were in kindergarten and first grade during the 1998– 99 school year. One study, which looked at the kids who attended public and private schools, found that small class size had no effect on student learning in reading or mathematics. The other study, which used the same data set minus the private-school kids, found that kindergarten and first-grade children who attended classes that were seventeen kids and one teacher or less learned about three weeks more than kindergartners and first graders who learned in large groups -- greater than twenty-five students. In that study, kids in middle- size classrooms -- between eighteen and twenty- five kids per teacher -- also did slightly better than the kids in the very big classrooms. What we learn is that all things being equal, big classes aren’t ideal -- medium and small-size ones are better. It seems that in smaller classes teachers are able to teach more. But the differential between kids who attend big ones and kids who attend small ones amounts to only three weeks of learning per year -- a surprisingly small margin.

Other more formally designed experiments in class size provide us with further clues as to who may benefit from smaller classes. In 1985, researchers in the state of Tennessee randomly assigned over six thousand kindergartners to one of three experimental conditions: (a) a small class of between thirteen and seventeen kids and a single teacher, (b) a large class with twenty-two to twenty-six children and a single teacher, and (c) a large class of twenty-two to twenty-six children led by a teacher and an aide. At the end of kindergarten, the achievement of the children in small classes was almost a month ahead of the achievement of the kids in the other two kinds of classes.

By the end of first grade, the kids in the small class were almost two months ahead. After four years, children in small classes were 5.4 months ahead in reading and 3.1 months ahead in mathematics. In small classes, kids learned more. So in that way, small classes were successful.

But the way some education researchers look at it, this experiment showed that small class size was a big disappointment. Why? I would like my kids to learn as much as they possibly can each school year. But the Holy Grail of educational intervention is a little more ambitious. Researchers are constantly looking for interventions that improve the rate of learning. Small class size doesn’t do it. Every year kids who were in smaller classes learned about a month more than kids in larger classes -- which means if they got smaller classes for three years, they’d be three months ahead of kids from the large classes. But the rate at which they learned was never compounding -- smaller classes gave students the equivalent of an extra month every time. The personal attention didn’t kick-start their ability to learn any faster.

Which doesn’t mean it’s not worth doing. It means that small class size alone is not going to help underachieving kids catch up and stay on par with high achievers. And even though the rate of learning did not continue to accelerate, the positive effects of small class size were long-lasting. When the kids who were assigned to small classes in kindergarten through third grade got to high school, they were earning higher grades and were more likely to complete advanced academic classes, take college admissions tests, and graduate. Later analysis of this data, and additional data from a 1996 study of small class size in low-income schools in Wisconsin, highlights an interesting point: African American kids who attended predominantly African American schools get a bigger boost from small class size than did white kids. In Tennessee, on average, black students in small classes ended third grade with academic achievement that was 7 to 10 percentage points higher than black students who attended the large classes. White students in small classes were only 3 to 5 percentage points ahead of white students from those larger classes.

Before you pull out your cell phone, ask to speak to your child’s principal, and demand that your precious gem gets moved into a smaller class, let’s reflect a bit about what this research means. The best study we have finds a small but persistent difference when you compare kids in a classroom of twelve to seventeen kids to a classroom with twenty-two to twenty-six. If you are making big decisions about your family life and your child’s education to get your child moved from a twenty-four-child classroom to one with twenty children in it, you may be acting on your best instinct, but the research doesn’t necessarily back you up.

The Tennessee study has become the gold standard in most debates about class size. It has also come under a lot of fire. Researchers who revisited the schools where the study was conducted suggested that some parents may have advocated to get their kids in the smaller classes -- thus making the "random" part of the study not so random. Others pointed out that some kids were in big classes and then transferred to smaller ones -- and those tended to be the kids of activist parents, the same parents who would make sure their children got their homework done. Researchers also complained that certain protocols that should have been in place were faulty. For the results to be considered "pure," for example, none of the test subjects should know which of them is getting active treatment and which is getting a placebo. In the Tennessee study, teachers were told that they were taking part in a test to determine the effect of small classes on learning, and that might have, consciously or unconsciously, affected the way they taught.

To be sure, all the nay-saying and bellyaching about the way the Tennessee test was conducted didn’t slow the enthusiasm of class-size-reduction proponents. In fact, the Tennessee project changed education policy in the entire state of California. In 1996, state education officials on the West Coast got legislators to appropriate $1 billion a year to cap elementary school class sizes at a strict twenty kids to one teacher. A pricy undertaking, it led to an unprecedented hiring binge, with the state bringing 28,886 new teachers on board. Six years later, the Rand Corporation published a study examining the results of the California effort-- and it was discouraging. The good news was that, overall, California’s educational performance had gone up. The bad news was this: despite hiring all those new teachers, the kids in the small classes were performing about the same as kids in the larger classes. And those positive downstream effects -- better grades in high school and higher graduation rates -- never materialized, either.

So, what happened? No one is sure. But there are two strong hypotheses: either the Tennessee results were specific to that state and that experiment, or -- and this is one that most educational experts favor -- teacher quality matters more than class size. California went on a hiring spree at a time when there were not a lot of highly qualified teachers waiting around on the sidelines to be hired. Because they had to hire so many teachers so quickly, they paid little attention to hiring the best ones. While very small classes early on seem to provide an advantage for kids -- particularly low-income African American kids -- good teachers, it seems, are even more important for increasing student achievement.

Peg Tyre is the author of the New York Times bestseller "The Trouble with Boys." She was awarded the prestigious Spencer Research Fellowship at the Columbia Graduate School of Journalism, where she began work on this book. Her writing about education has appeared in Newsweek, The New York Times, Family Circle and She lives in Brooklyn, New York.

Excerpted with permission from "The Good School: How Smart Parents Get Their Kids the Education They Deserve."

By Peg Tyre


Related Topics ------------------------------------------

Education Nonfiction