On Bill Gates, Class Size, and American Parents

In a Washington Post op-ed and speech yesterday to the National Governors' Association, Bill Gates floated the idea of dealing with state education budget cuts by increasing the class sizes of the most effective teachers:

What should policymakers do? One approach is to get more students in front of top teachers by identifying the top 25 percent of teachers and asking them to take on four or five more students. Part of the savings could then be used to give the top teachers a raise. (In a 2008 survey funded by the Gates Foundation, 83 percent of teachers said they would be happy to teach more students for more pay.) The rest of the savings could go toward improving teacher support and evaluation systems, to help more teachers become great.

This is not an unreasonable proposal, but it is certainly a politcally difficult one. As I've written before, the free market education reform movement has long struggled to attract grassroots support in neighborhoods and among parents; loudly advocating for larger class sizes will only make this problem worse. In New York City, Newark, Denver, and every other district whose school reforms I've covered over the past year, class size remains the single most salient issue for community groups, uniting folks no matter how they feel about school closings, charter school co-locations, or standardized testing. 

Simply put--and decades of polling on class size backs this up--Americans love the idea of their children getting more one-on-one attention from teachers. Indeed, OECD data shows the U.S. has one of the world's largest gaps between average public school class size (23.6 sstudents) and average private school class size (19.4 students), suggesting that small class size is one of the main perks affluent Americans are paying for when they pull their kids out of public schools.

All that said, it's certainly true that small class size is not clearly correlated with high achievement--and neither is large class size. Take a look at this OECD chart.

 

Note that the U.S. already has larger than average class sizes. Of the nations whose class sizes are even larger than ours, Korea, Japan, and Australia are clearly outperforming us; they each rank in the PISA top 10 for multiple academic subjects. Yet other top-performing nations, including Finland, Singapore, and Canada, have made lowering class sizes a cornerstone of their education policy.

Since small class sizes alone don't ensure high achievement, at least as measured by standardized tests, it might make sense to argue that maintaining small classes should not be a priority during lean economic times. The problem is that American parents are concerned not only with their children's test scores, but also with their day to day experiences at school. Parents want their children to have meaningful personal relationships with educators--the sorts of life-changing experiences many of us remember fondly when we think back on our favorite teachers, whether they helped us score higher on a chemistry exam or just got us through a difficult time at home.

Since lowering class sizes is extremely popular, it would take a major public reeducation effort to convince Americans that larger classes are better classes. My hunch is that this issue would not be a political winner. Just remember: There are ways to improve teacher quality through better training, evaluation, and professional development, while simultaneously keeping class sizes steady. It's called progressive taxation, and for some reason, even Democrats are terrified of talking about it.

Comments

[Urban Owens]

While I am all for a progressive tax and a larger investment in education, I do not see a personal connection as necessary for effective teaching. A teacher is not there to be a buddy or friend to his or her pupils.

[Leonie Haimson]

As I pointed out on the Huffington Post, Bill Gates misstates the findings of the survey he financed --

http://www.huffingtonpost.com/leonie-haimson/arne-duncan-and-bill-gate_b_829094.html

Moreover, all surveys that ask the question find that teachers overwhelmingly believe that class size reduction is the best way to improve their effectiveness.

Why do Bill Gates and the other corporate type reformers try so desperately to undermine the importance of class size -- going against commonsense, research and experience? Search me. One would have to suspect that they really don't want to improve the quality of the public schools.

[Greg]

I'd be very happy to see average class size of 24 our urban US public school, but we have class sizes of 28-30 in primary school. Perhaps rural class sizes, which tend to be lower, are distorting the averages?

[Aaron Bady]

You dismiss the class size correlation way too quickly on the basis of that chart. That chart is, at most, suggestive, but proves nothing, and is certainly amenable to alternate explanations. As a teacher, I'd need an awful lot of convincing before I'd believe differently than what my experience teaches me: class size matters a lot, and smaller class sizes make good teachers much better.

[Matt Brown86]

Okay, if you're a teacher, paid a flat fee (aka SALARY) for teaching, and you can ASSIGN WHAT YOU WANT to students (barring administrative interference which is almost nil), and you've been given a class of students that is ten more than you had last year, what kind of work will you assign? And what kind of feedback will you give on that work? Will it be the 2-page essay? Or something that doesn't actually require that you read it?

Working conditions matter; class size matters. You want to burn out the good teachers? Increase class sizes.

[Liberpolly]

The quality of education will continue to decline as long as we cannot fire bad teachers and promote good ones regardless of seniority. Taxes have nothing to do with it.

[joel hanes]

I spent a couple years working in the local elementary schools.

It's amazing to me that none of the excellent teachers I knew there agreed with Liberpolly about the problems facing our schools. If Liberpolly was correct, you'd expect the excellent teachers to agree, since they are the ones whose excellence is not being rewarded -- but in my experience, views like Liberpolly's are found almost exclusively among those who do not work in education at all, and very seldom among teachers, regardless of their experience or ability.

PS: American education is not in any meaningful decline. It has achieved about the same degree of success for many decades, despite the vicissitudes of funding and the sturm and drang of "reform" and "innovation". Anyone who tells you otherwise is selling something.

[Danny DeMichele Entrepreneur]

I'd need an awful lot of convincing before I'd believe differently than what my experience teaches me: class size matters a lot, and smaller class sizes make good teachers much better.

[ThaomasH]

Good teachers need teacher's aides to help with the mindless lesson planning, reporting to parents, homework checking, essay correcting etc. We need larger class sizes and a lower teacher/pupil ratio.

[Tim H]

Class size may not matter for high achievers, who are likely to find a way to succeed, though I think its hard to argue that smaller classes would be bad. But for children who face challenges from learning disabilities, ADD, language, etc. it matters a great deal. Further, class size probably matters more at schools serving lower income, and/or immigrant neighborhoods, or in districts where high achievers are siphoned off into charter schools, magnet schools, and the like, leaving behind those requiring more attention. There is no magic "reform" that will work in all cases, but class size is certainly one major aspect.

[John Salzsieder]

It all hinges upon what you call success in teaching/education. Correct answers on a standardized test? Learning defined by the committee who wrote the test?

[David Stahler]

Perhaps this is an obvious point, but...

Class size matters most in the context of what kind of students are in the class. A group of upper-level, high achieving students who are well-socialized can still thrive in a large class. However, in classes with higher percentages of students who have severe learning disabilities and/or emotional disturbances, large class size is a recipe for disaster no matter how good the teacher is. The problem is teachers are facing more and more of these kinds of students.

The irony, of course, is that the kinds of students who can function perfectly well in large classes are often the ones who end up in private schools with small class sizes.

[Martin]

While limiting this to primary education is helpful, class size matters more in specific areas. Lower grades (K-3) have considerably more behavioral and attention problems than higher grades, where the benefits of small class size are minimal. There's a world more difference between 5 year olds and 9 year olds than there is between 9 year olds and 13 year olds on that front.

The other place where it matters is in subject areas where learning is more experiential. There are to main factors to this: 1) experiential learning in a classroom is less familiar and when problems occur, they tend to completely block learning. Consider a science lab where something goes wrong - until the problem is fixed, learning on that lesson essentially stops. Or a music class where an instrument stops working properly. 2) In group experiential learning settings, students are more likely to find ways to not participate and ultimately many reject the field altogether. This is a long-documented problem where females may become intimidated with science or technology subjects because male lab partners dominate the experience - and they often permanently reject science and technology studies before they even reach middle school. By eliminating the pairing and forcing every student to stand on their own, this problem at least gets reduced.

[Chevy avalanche parts]

Well I think This views like Liberpolly's are found almost exclusively among those who do not work in education at all..Thanks

[Bill]

Korea/Japan have extremely homogeneous populations/cultures compared to the US. This might make it easier to do well with larger class sizes.

[Michael]

I have to call bullshit on this graph and post. Displaying average size of a primary school class on national level doesn’t tell you anything meaningful. At least show us corresponding national achievement levels to go with it. Even then there are several factors that contribute to achievement which make any statement about the correlation of class size meaningless.

If you isolate public schools in well funded, affluent middle class school districts in the United States and then compare class size with achievement you might be able to observe something worthwhile. Better yet, do the same comparison for public and private schools in the same areas. Then add in corresponding low income districts. From there do similar analysis in different nations to identify cultural impacts.

The reality of the situation is that some students will thrive in a large class environment and others will do better in a smaller setting. Likewise, some teachers will excel with a larger number of students, while others will have more success with fewer students.

[Nelson Chamberlain]

It's surprising that not so long ago, Gates proclaimed that large schools were the root cause of low student performance and now he's saying that large classes would not be a problem. Seems contradictory to me.

Perhaps no one has explained the issue to Gates in computer terms. The more tasks (students) you assign to a CPU (teacher), the smaller the time-slice the CPU can spend on each task. If additional run time is not allocated to compensate for reduced time-slices, the likelihood increases that the tasks will not successfully complete (poor student scores). And the more tasks assigned, the more likely it is that a rogue task (disruptive student) will consume all available system resources.

Gates (and most school critics) has probably never actually taught a class of disinterested students. He has probably led many seminars and speeches where all the *adults* listened with great interest to whatever the billionaire felt like nattering on about, so he probably thinks that teaching 9-year olds is simple. I would be more inclined to believe him if I knew that he'd ever successfully hosted one of his kids' birthday parties *without* any other adult in the room. Especially if the party lasted 6 hrs. Five days a week. I might believe him then.

[Cynthia Liu]

Experiment: increase the ratio of children to teacher at the school Bill Gates' children attend, and then see if he thinks it's okay to increase class sizes.

I'm offended that he would urge for others what he would likely resist for his own children.