I freely admit that I don’t understand how to run a multi billion dollar software company. I’m still intrigued, therefore, that Bill Gates continues to believe that he is qualified to make education policy.
In today’s New York Times, Sara Mosle discusses a plan from Gates, along with Michael Bloomberg and Arne Duncan, that would “increase class sizes for the best teachers,” in order to increase the chance that students could study under an excellent teacher. This idea from Gates has been around at least since 2011, when he wrote an op-ed arguing for this in the Washington Post. In the world of many reformers, class size has never been conclusively shown to increase student achievement, whereas “excellent teachers” have been. Therefore, why not disregard the statistically inconclusive factor (class size), and focus on getting students in front of a high performing teacher. Mosle agrees that large numbers of educators are not jumping at the plan, so I don’t fear that this plan is something that we’ll have much reason to worry about. Still, the level of thinking that it reflects from some of our most important people in education policy is irritating.
It is obvious to anyone who has ever taught before that it is easier to reach individual students the smaller the class is. The reason, I think, that class size cannot be proven to have an effect on student achievement is because all of the hundreds other variables that come in to play in any classroom that may be hard to statistically measure. Let me give one example with which I am familiar: Japan has much larger class sizes than the United States, and also scores better on international exams. This could lead one to the simple conclusion that the Japanese experience proves that class sizes do not matter. In fact, to bolster their arguments, ed reformers often mention Japan and South Korea as examples of countries that have huge class sizes but boast impressive test scores. Indeed, every class that I taught in Japan had around 40 students, but what the reformers who ridicule the notion that class size matters do not mention is that a vast majority of Japanese students past the age of elementary school attend private prep schools called juku several times a week in the evenings after school. As the Economist reported in 2011, Japanese test scores tend to rise in direct proportion to the amount of money spent on private classes. In South Korea as well, about 74% of students attend private hagwon schools to supplement their education.
Every Japanese teacher that I knew complained that the class sizes were too large and made it difficult to respond to students’ individual needs. Teachers were typically strapped with workloads that kept them at work until around 8 PM, and all agreed that they could be more attentive to students if the class sizes could be reduced. The fact is that Japanese education is currently sustained by a massive network of private supplemental schools, the same as its neighbor South Korea. Neither of these countries provides good evidence that class size doesn’t matter. If anything, the number of students attending remedial instruction show that the public education systems are not equipped to meet the needs of all their students.
When statistical analysis fails, sometimes it’s best to trust intuition. A teacher with fewer students can give more individual attention to each one, and that gives them a better chance of success. Just because this can’t be definitively proven with data doesn’t mean it isn’t true.