Discuss as:

FACT CHECK: Class sizes do matter


Mitt Romney found himself on the opposite side of a skeptical audience on Thursday in Philadelphia, after he seemed to dismiss the impact of class sizes on student achievement.

Mario Tama / Getty Images

Republican presidential candidate Mitt Romney greets students in a music class at Universal Bluford Charter School on May 24 in Philadelphia, Pa.

At an event capping a weeklong messaging effort surrounding the presumptive Republican nominee’s education policy, Romney cited a study by management consulting firm McKinsey to back up his argument that the number of students per teacher in a classroom wasn’t the most important predictor of academic success.

But the former Massachusetts governor’s assertion differs from the evidence produced in large, recent, peer-reviewed academic research showing that class size does, in fact, impact student outcomes.

“Well, if you had a class of five that would be terrific; if you had a class of 50 that’s impossible,” Romney said, when asked his view on class sizes. “So there are points where I think those who have looked at schools in this country and schools around the world, McKinsey for instance … went around the world and looked at schools in Singapore and Finland and South Korea and the United States and looked at differences and said gosh, schools that are the highest-performing in the world, their classroom sizes are about the same as in the United States. So it’s not the classroom size that is driving the success of those school systems.”

The Republican presidential candidate visited a West Philadelphia charter school

A teacher in the audience pushed back, citing a landmark Tennessee study conducted by a Harvard researcher in the 1980s – famous in the world of education research – which looked at the Tennessee STAR program, or Student/Teacher Achievement Ratio, in which the state reduced class sizes across the board by about a third, from 22-25 students per teacher down to 13-17.

The study of the program -- conducted when current U.S. Sen. Lamar Alexander, a Republican, was governor – found “compelling evidence that smaller classes help, at least in early grades.”

Romney didn’t respond directly to the teacher or study during the event in Philadelphia.


Political food fight

The Obama campaign tried to capitalize. “Larger Class Sizes Are the Answer to a Better Education? On What Planet?” blared an email from Obama spokeswoman Lis Smith.

That was echoed on a conference call today. “Romney insisted in face of logic, that small classes don't help,” said Philadelphia Mayor Michael Nutter. “Two years ago, [he] claimed that effort to reduce class size may hurt. I'm not sure what universe he's operating in. Every parent knows that smaller classes are preferable. Everybody knows that except Romney.”

But Romney did acknowledge in Philadelphia that having the smallest classes are optimal, but that they’re not the driver of success in the classroom.

The Romney campaign pointed to Obama’s Education Secretary Arne Duncan, who noted – not as a matter of personal opinion, but of official administration policy – that class sizes should be increased.

“In our blueprint for reauthorizing the Elementary and Secondary Education Act, we support shifting away from class-sized based reduction that is not evidence-based,” Duncan said, according to a transcript of Duncan’s speech, posted by Education Week, at the conservative think tank, the American Enterprise Institute.

Duncan has also called class size "a sacred cow," "and I think we need to take it on," said in March 2011. He later said, "My point there was that I think the quality of the teacher is so hugely important. I've said things like, give me the parent, give me an option of 28 children in a class with a phenomenal teacher or 22 children in a class with a mediocre teacher. If I was given that choice, I would choose a larger class size."

After the Obama conference call, Romney spokesman Ryan Williams boasted on Twitter: “If @BarackObama believes what his campaign is saying, he should fire Arne Duncan for supporting @MittRomney's view on class size.”

Romney campaign spokeswoman Andrea Saul released the following statement:

“If President Obama is as focused on class size as his campaign seems to be, his outdated view of education reform puts him at odds with leaders like Michelle Rhee, Bill Gates, and his own secretary of education -- all of whom have said that improving teacher quality gives kids the best opportunity to learn. Secretary Duncan even said that he ‘would choose a larger class size’ if it meant having a better teacher in the room. President Obama should be ashamed that his campaign is launching such cheap political attacks at the expense of a serious discussion about education policy. If he actually believes what his campaign is saying, he should fire his education secretary for supporting the same view on class size that Governor Romney is advancing.”

A broader reading of Duncan’s remarks before AEI shows he believes smaller classes are a good thing, but because of state budget restrictions, school districts need to find ways to adjust.

“Consider the debate around reducing class size,” he said. “Up through third grade, research shows a small class size of 13 to 17 students can boost achievement. Parents, like myself, understandably like smaller classes. We would like to have small classes for everyone -- and it is good news that the size of classes in the U.S. has steadily shrunk for decades. But in secondary schools, districts may be able to save money without hurting students, while allowing modest but smartly targeted increases in class size.”

In fact, research bares out that smaller class sizes have resulted in gains in K-3, but results are either inconclusive, not significant, or non-existent for older children.

The Obama’s campaign’s Smith responded this way, in an email to First Read: “Both experience and evidence show that smaller classes are better than bigger classes, especially for young children.  But class sizes aren’t the only thing that matters, and President Obama and Secretary Duncan are also working to raise academic expectations, invest in teacher quality, and turn around struggling schools. That’s very different from Mitt Romney, who thinks that smaller class sizes don’t  matter or can even be harmful.”


Taking on unions

Romney has accused Obama of being held captive by teachers unions, but positions like the one taken above by his education secretary, as well as his administration’s push on merit pay, teacher evaluations, and support for charter schools, have rankled those unions.

Obama has said, since the 2008 campaign, that reforms were necessary but that he would try to work “with” unions. Romney has taken a combative tone and blamed unions for promoting class size at a September Republican presidential debate in Florida.

“[A]ll the talk about we need smaller classroom size, look that's promoted by the teachers’ unions to hire more teachers,” Romney said, adding, “[A]s president, I will stand up to the National Teachers Unions.”

Romney’s tough talk toward labor has been a hallmark of not just his education plan, but his overall economic strategy. It’s understandable, in some ways, why Romney, like many mayors and governors of both parties across the country, would want to cut out teachers’ unions. As with many businesses, unions often prove to be an obstacle in an executive’s ability to enact wholesale changes or implement new programs – like pay for performance (merit pay), or fire teachers regardless of whether they’re underperforming.


Body of evidence

Parental involvement, effective teachers, and competent administrators are certainly major factors in how well students do. But studies have found that class sizes, when reduced by more than a couple students, especially in early grades, can have an impact on student achievement.

A study conducted in Tennessee -- published in April 2011 in the Journal of Educational and Behavioral Statistics, conducted by researchers from the University of Chicago and Virginia Commonwealth University -- found improvements as a result of smaller classes in “reading, mathematics, listening, and word recognition test scores” in early grades.

A California study, conducted by economics professors at the University of Kentucky and Amherst College and published in The Journal of Human Resources, also found that test scores improved -- even when taking into consideration the number of inexperienced teachers that had to be hired to fill the 25,000 jobs created by the state’s $1 billion effort to reduce class sizes. After a few years – when the new teachers gained experience – the cost of hiring those teachers was net-even.

“[T]here is little or no support for the hypotheses that the need to hire large numbers of teachers following the adoption of CSR [class-size reduction] led to a lasting reduction in the quality of instruction,” according to the study. “Overall, the findings suggest that CSR increased achievement in the early grades for all demographic groups…”

And on cost: “From a purely distributional point of view, the benefits of CSR were allocated in a quite regressive manner in the short term but in a close to neutral manner as of six years following the implementation of the policy.”

A Florida study, which followed up on the California results with a study of Florida’s similar effort, conducted by a Harvard researcher and government professor, found the class-size reduction had a minimal impact. The results “indicate that the effects … on cognitive and non-cognitive outcomes were small at best and most likely close to zero,” according to the study.

But as it also points out, class sizes were only reduced by two or three students per class: “One might not expect a large effect given that over three years class size was only reduced by 1.9 students more in the treated districts than in the comparison districts, but I also find no evidence of positive CSR effects in grades seven and eight, where the relative reduction in class size was three students.”

The Romney campaign, for its part, when asked about these studies, didn’t deny that class sizes impact student achievement; it’s just not “his focus.”

“The governor said, ‘Just getting smaller classrooms didn't seem to be the key,’” a Romney aide told First Read by email. “His policies address ensuring better teachers in the classroom and rewarding their success which is a very important part of improving student outcomes. That’s his focus.”

The Florida study also notes that providing additional teacher resources and supports, like the STAR Program did, combined with smaller class sizes, could have also had an impact: “It is impossible to disentangle the effect of reducing class size from the effect of providing additional resources.”

That’s something the original Tennessee study made a point of as well: “The benefits derived from these smaller classes persist leaves open the possibility that additional or different educational devices could lead to still further gains. For example, applying to small classes the technique of within-class grouping in which the teacher handles each small group separately for short periods could strengthen the educational process (essentially a second-order use of small class size). The point is that small classes can be used jointly with other teaching techniques which may add further gains.”

Like in many things, and especially in education, there’s no magic bullet. It’s a combination of a variety of tools, including class size.