by Brian
I teach high school mathematics, and this year I have 132 students in 5 classes. That's an average of 26.4 students per class. Not bad, right? Except that one of them only has 13 (Pre-calculus) and one has 33 (Geometry). So the "average" really doesn't tell the story.
What about that Geometry class? I know it would feel better to both me and the students if there were only 25 of them, but would they really do better? This Saturday I am going to participate in a Policy Symposium of National Board Certified Teachers, and one of the topics we will discuss is class size. Superintendent Randy Dorn has recommended class sizes of 15 in Kindergarten to third grade, and 25 for fourth grade through high school. So has the Basic Education Funding Task Force, and the Washington Learns Council. Ample research has shown that reducing class size at the elementary level has a real and lasting impact on student achievement. But there is little evidence that smaller classes have a noticeable impact at the secondary level. When my colleagues complain about large classes it is usually about their own workload rather than their students' success.
So given the reality of limited resources, and scant evidence of class size making much difference at the secondary level, shouldn't we concentrate our efforts at the lower elementary grades?
I’m an 8th grade student and I prefer small classes, because I’m a little shy and I don’t like to interrupt when someone else is asking for help, so I wait and wait and in huge classes I just don’t ask questions, because teachers don’t have the time.
Mark, it’s compelling to read your account of your writing conferences. I read a study of the positive effects of writing comments on a student’s paper as opposed to just giving it a letter grade, and your idea really brings that home. Your anecdote gives great support to the idea that if we teach differently with small classes, we will get the gains we seek.
I guess my response to the grades comparison is this: who assigns the grades? I know that when my class sizes are larger, it changes the kinds of assignments I offer, and I adjust my learning goals for the group. I succeed in coverage but without the depth of mastery that my smaller classes can demonstrate.
If the curriculum is lock-step and all graded assessments are uniform and not subject to teacher interpretation, I suppose the grades argument might work.
I am the principal at Brian’s school and I just met with a teacher today to look at the final grades for our school from last semester. It’s interesting that the data analysis she did doesn’t show an correlation between class size and student success. Some of our largest classes had excellent student success while some of our lowest sizes did not.
Thanks for the extended explanations. You obviously know what you’re writing about.
I agree that smaller class sizes permit more attention to certain matters fewer teachers will manage in larger classes. I had what some considered large classes in my fifth grade classes and used instruction to match the size.
From a public policy view, variations in class sizes will likely account for variations in costs of learning rate changes depending on instructional choices of teachers. Perhaps that’s a different discussion.
Although I’ve seen the research that suggests class size doesn’t matter as much as teacher quality, I wonder what would happen if we had both? Let’s put a great teacher into every classroom and let’s put the right number of kids in that teacher’s classroom. And while we’re engaging in all this “crazy talk” why don’t we ask those great teachers what that class size is? I teach third grade. I would love a class size of 20. If I taught high school algebra, maybe I’d want more. If I taught kindergarten, maybe I’d want less. Class size, to me, is one of those things where you don’t really notice the first one or two kids, or maybe not the third or fourth, but after awhile you look around and there’s thirty-one kids in your room and you’ve switched from teaching to lecturing.
And despite the “research,” I have data that suggests I am a more effective teacher when provided smaller class sizes…if my effectiveness is to be measured by state test scores.
With a larger class, it simply is not feasible for me to do many of the things I know have measurable results. As just one example of many practices I can do with smaller classes but not larger, in the last several years I’ve begun using face-to-face writing conferences–15-20 minutes each–where every kid gets my undivided attention on their capstone writing assignments for each given unit. We revise together, I can individualize instruction, we can practice together with grammar and vocabulary in context. I’ve had great pass rates on the state tests. Even more important than the good state test results, I see immediate improvement on their next major writing assignment…much more improvement than when I put comments and editing marks on their drafts. To devote that kind of time to each and every kid means we do in-class meetings as well as lunch and after-school meetings. When we meet in class, I devote full focus to the student in conference, while the other students are working on independent reading, writing workshops, other projects, constructivist learning, etc.
I can do this with 24 kids at the most. The time is manageable, and it’s the right amount of self-directed time for the other kids. With that number, the turnaround on essays is tighter and thus the feedback is more relevant and immediately internalized. The kids find it useful so they beg for writing conferences, so I try to do it with as many major writings as I can.
With 30 kids, I have to add three whole instructional days to the process. That’s three more days between drafting and revision, three more days of teacherless self-monitoring. The last time I had that large of a class and tried the writing conferences, I ended up having to cut conferences drastically because there’s only so long a class full of 15-year-olds can continue to self-manage several days in a row. A 15-20 minute conference is able to go so much further than a 5-8 minute conference…even if I was able to do more of the short conferences, the long conferences still produce such greater gains.
So instead, I now do writing workshops with my larger classes. It is no where nearly as effective…and their writing progress evidences this. The ONLY students of mine who have ever failed the state writing assessment were in my classes with 28 or more students. In my smaller classes, the pass rate has been 100%. There is a correlation and I propose it is not just a coincidental one.
In this case, then, the one-on-one conferences are certainly the better practice. However, it is a practice which cannot be as frequently or consistently applied to larger classes. There are several other projects, activities, and practices which are unfeasible due to time constraints with larger classes.
This is why class size matters, and I think the research supports it: smaller class sizes, when accompanied by better pedagogical approaches (often only feasible with smaller groups), DOES result in greater impact on student learning.
Bob, I think when I teach a large class I do what I need to do to control the “buzz”. I don’t know of any adjustments to make except to protect the learning climate. Everyone needs a chance to learn; no one can take that from another student. The more students I have, the “tighter” the ship has to be.
Learning rates vary for a myriad of reasons. They not only vary from teacher to teacher, they vary from hour to hour. Every mix of students is different. I might succeed one hour, and struggle the next. (See Tom’s post on Data Driven? for my default opinion on educational research.)
Good point, Mark, about no definitive empirical research on effects of class size on learning. To support your point, a landmark study released about 1994 or 1995 as an Illinois State University doctoral dissertation reported no definitive effects of class size on learning. She used a statistical meta-analysis of all available empirical comparative studies (about 1,700) of class size. Can anyone point me to an update of that study?
As I think you must be, Brian, I’m curious: Do you think teachers of larger classes adjust their instruction to make it more likely that students will meet desired learning criteria? If so, why do learning rates vary across teachers under comparable conditions? It doesn’t appear from state required test results that teachers take advantage of available instructional options.
For convenience, I use Jaime Escalante’s calculus instruction of barrio students as an example for my question about effective adaptive teaching. Escalante’s instruction was later depicted in the 1988 movie “Stand and Deliver”.
Excellent point, Mark. Most teachers don’t change their practice to take advantage of smaller classes; they just keep doing what they’ve always done.
The research on class size and student achievement often indicates that the size of the class does not impact student learning as much as the practices applied in that class (thus, people who oppose costly class size reductions or caps cite this part of the research). However, what is clear, is that those better, more promising practices, are often only feasible when class sizes are reduced. For example, one article I read described how if teachers would just drop their dependence on direct instruction and spend class time working individually with students, student learning would increase regardless of class size. True, but my colleague with 40 students in her Spanish I class likely doesn’t have time to do so in a way that continues to move the whole forward in a meaningful way.
To be sure, the research does not say that simply reducing class size is the magic bullet. With that reduction must come a shift in pedagogy and implemention toward better practices not feasible with large classes.
Links to peruse for the research on this topic…some old, some newer:
http://www.ed.gov/pubs/ReducingClass/Class_size.html
http://www.ed.gov/pubs/ClassSize/academic.html
http://www.edweek.org/rc/issues/class-size/
http://www.serve.org/SDImprov/classsizelinks.php
Nancy, I want to live long and prosper:-). I like what you said about jettisoning the lockstep language. I teach math, and I can grade 33 math tests in one night. But my friend who teaches English can’t read 33 essays, mark them with meaningful comments, and give them back the next day. The chemistry teacher can’t safely supervise a lab activity with 33 students. We’ll have to have flexible language to allow for a rational allocation of staffing.
But if we’re going to close any gaps I’m convinced we’ll have to start with small classes as soon as they begin school.
I have a class of 25 First Grade Students. I have taught for over 20 years. It doesn’t get any easier. Each year students bring a whole new set of needs. Carefully designed instruction for each student so that he/she can achieve success takes many hours. Fewer students would allow me to do a more in depth study of each student’s need which would result in improved student learning for all. I might even be able to refrain from staying late and working on weekends.
I haven’t been able to find any research on the effects of total student load per teacher. IS there any? if so, where?
How about room size? 26 students don’t fit in my room, especially when I get the 16 year old football players! If I have kids stuffed in so tight that I can’t even get through the rows, classroom control and efficiency suffer.
I think there is also a big difference between a health or phys ed teacher’s load of 150 students daily and an English teacher’s load of 150–trying to read and respond to 150 essays, even if each only gets 5 minutes, adds up to over 12 hours. Multiply that by 10 essays a year (minimum) and it’s a huge burden on time outside of class. I don’t want to give up a 12 hour day on the weekend!
Interesting post, Brian. Do you have a death wish (laughing)? I admire your use of research (and Kristin’s supporting comment) to point out that class size is critical when your students are 6 and learning to read, and feels different when students have become accustomed to sharing teacher attention in a group-instruction format.
Maybe the deal with class size is getting rid of formulaic, lockstep language. Perhaps there are subjects or instructional strategies where students would benefit from larger groups, like content-rich lectures. What if some secondary teachers were paid more for handling larger groups effectively? What if other jobs were deemed “small group”-intensive and teachers were expected to be responsible for meeting effectiveness targets, rather than just coping with a pre-set number of kids? There are lots of different ways of changing the “X kids to one teacher” ratio to get better results.
I do support smaller class sizes, in general. But teacher effectiveness matters a great deal more than class size, and class size reduction is very expensive. Reducing class size by 2 kids, across a district, will not yield great results. Most parents would rather have their kid in class with a superb teacher and 34 other students than have a them with a mediocre teacher and 20 classmates.
Yes! I agree with you. I have 38 kids in my 4th period and only 17 in my 6th period. You would think my 6th period was a dream, but it’s difficult, like being at an awkward party where not enough people showed up.
Because they are teenagers, they don’t want my one-on-one tutoring every day. They don’t want to have to share every 17th time. They get a little sick of partnering up with the same people again and again. There is definitely a critical mass that needs to be reached in a 10th grade language arts class, and 17 is not it. 38 exceeds it, but 25-32 is just about right.
Young children, on the other hand, love one-on-one attention. They need more supervision when they’re working in groups. My vote is for putting all that classroom-size spending towards elementary.