The Class Size Conundrum

“In fact, the research that shows benefits from class size reduction finds effects that are smaller than most people realize …”

The above quote was taken from Education Myths by J.P. Greene. Greene goes on to say that smaller classes would require too much funding, taking away from other possible reform efforts.

I read Greene’s words and thought that he had captured a snippet of the mess that is American educational research. If researchers could truly control for classroom conditions, I suspect they would find a large effect from class size — in certain populations and specific classes. In districts that historically have sent the great majority of their graduates on to college, class size mostly will not matter since upwardly-mobile students have their eyes on the prize. They do their homework and fight for grades they expect to be reviewed by college admissions counselors. When those districts are included in class size studies, they can be expected to take attention away from schools where class size may make all the difference in academic achievement.

The class size picture in financially- and academically-disadvantaged districts remains murky, but I would like to observe anecdotally that class size has sometimes had a big effect on my teaching life, depending on the class. I found that increasing my math class size from 19 to 24 students caused little disruption. Increasing my high school Spanish class from 28 to 34 (originally 40!), however, was a game changer. Suddenly, behaviors were impacting instruction to a much greater degree. Suddenly, I had firecrackers in bottles in the back of the room and gang symbols etched onto desks.

Research unsurprisingly shows varying benefits from smaller class sizes. Teasing out exact benefits remains impossible, due to variations in teachers and class compositions, among other considerations. I had two of those large Spanish 1 classes, and one was much better behaved and more academically centered than the other. That one class was almost no trouble. The other … Well, I took a day’s absence once, worried about my sub, and when I called to check on him, found the poor fellow’s voice was shaking and the police had been called. (In fairness, a student from the special education department suffering from behavioral disorders had taken advantage of my absence to sneak into the class, where he yanked a girl’s hair hard enough to set off a series of reactions. My students probably would have done fine if not for the unknown marauder.)

Eduhonesty: It only takes one student. Classes of thirty-five students can be easier to teach than classes of 20 students, depending on the students within those classes, one reason why studies on class size keep finding anomalies that researchers struggle to explain. A small group of disaffected kids can keep a large group from learning. The wrong buddies become a teacher’s bad luck. A quick catch may allow a teacher to change schedules and separate Kyle from James at the beginning of the year, but most of the time the best and only option will be a well-designed seating chart that may not work during projects, gallery walks and other group activities.

My problem with educational research can be captured in a snapshot by Greene’s words as he attempts to debunk an idea that ought not to be debunked. Class size matters. I’ve walked the walk that Greene minimizes, and it’s no stroll through a sunny park. While the class size problem cannot be reduced to reliable and exact numbers, dismissing the impact of class size in urban and academically-disadvantaged schools helps no one — not parents, not teachers, not administrators, and most especially not students.